Hi folks,
For a long time we've used "algae," which was designed by Britton about eight years ago, as the default colormap. This has been really nice for "branding" yt -- if you see an algae plot, it's probably (not definitely) made with yt. But it's also not accessible from a colorblindness perspective. Stefan van der Walt has been giving some really great talks lately about building a better colormap for matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU ) which culminated in viridis, which is shipping in recent versions of matplotlib and will become the default.
In support of this, he built a tool called viscm which can generate reduced versions of colormaps to show what they would be like with varying degrees of insensitivity to color. I've generated outputs from viscm of three of the custom colormaps we ship with yt:
Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I believe this is now also shipped with MPL) Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/
I love algae, but it's not the best from an accessibility perspective.
I'd like to propose that we use a new default colormap. If we do this, I see two options:
* Retain a "branding" by developing a new one either by using the techniques used by matplotlib (or one of the maps they opted not to use) or by modifying algae to be more accessible; looking at the response functions, I suspect it would be reasonably possible to modify it. (Modifying algae is my preference.) * Use viridis (which we may then have to ship if we have older versions of matplotlib to support)
-Matt _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
I would go for modifying algae.
On Jan 6, 2016, at 11:30 AM, Matthew Turk matthewturk@gmail.com wrote:
Hi folks,
For a long time we've used "algae," which was designed by Britton about eight years ago, as the default colormap. This has been really nice for "branding" yt -- if you see an algae plot, it's probably (not definitely) made with yt. But it's also not accessible from a colorblindness perspective. Stefan van der Walt has been giving some really great talks lately about building a better colormap for matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU ) which culminated in viridis, which is shipping in recent versions of matplotlib and will become the default.
In support of this, he built a tool called viscm which can generate reduced versions of colormaps to show what they would be like with varying degrees of insensitivity to color. I've generated outputs from viscm of three of the custom colormaps we ship with yt:
Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I believe this is now also shipped with MPL) Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/
I love algae, but it's not the best from an accessibility perspective.
I'd like to propose that we use a new default colormap. If we do this, I see two options:
- Retain a "branding" by developing a new one either by using the
techniques used by matplotlib (or one of the maps they opted not to use) or by modifying algae to be more accessible; looking at the response functions, I suspect it would be reasonably possible to modify it. (Modifying algae is my preference.)
- Use viridis (which we may then have to ship if we have older
versions of matplotlib to support)
-Matt _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
I would also be for coming up with our own colormap. That said, I think simply modifying algae won't be enough, since it is too perceptually nonlinear.
On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com wrote:
I would go for modifying algae.
On Jan 6, 2016, at 11:30 AM, Matthew Turk matthewturk@gmail.com wrote:
Hi folks,
For a long time we've used "algae," which was designed by Britton about eight years ago, as the default colormap. This has been really nice for "branding" yt -- if you see an algae plot, it's probably (not definitely) made with yt. But it's also not accessible from a colorblindness perspective. Stefan van der Walt has been giving some really great talks lately about building a better colormap for matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU ) which culminated in viridis, which is shipping in recent versions of matplotlib and will become the default.
In support of this, he built a tool called viscm which can generate reduced versions of colormaps to show what they would be like with varying degrees of insensitivity to color. I've generated outputs from viscm of three of the custom colormaps we ship with yt:
Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I believe this is now also shipped with MPL) Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/
I love algae, but it's not the best from an accessibility perspective.
I'd like to propose that we use a new default colormap. If we do this, I see two options:
- Retain a "branding" by developing a new one either by using the
techniques used by matplotlib (or one of the maps they opted not to use) or by modifying algae to be more accessible; looking at the response functions, I suspect it would be reasonably possible to modify it. (Modifying algae is my preference.)
- Use viridis (which we may then have to ship if we have older
versions of matplotlib to support)
-Matt _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
I think there are several colourmaps that were created when Viridis was invented. I personally like Inferno.
-erik
On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum nathan12343@gmail.com wrote:
I would also be for coming up with our own colormap. That said, I think simply modifying algae won't be enough, since it is too perceptually nonlinear.
On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com wrote:
I would go for modifying algae.
On Jan 6, 2016, at 11:30 AM, Matthew Turk matthewturk@gmail.com wrote:
Hi folks,
For a long time we've used "algae," which was designed by Britton about eight years ago, as the default colormap. This has been really nice for "branding" yt -- if you see an algae plot, it's probably (not definitely) made with yt. But it's also not accessible from a colorblindness perspective. Stefan van der Walt has been giving some really great talks lately about building a better colormap for matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU ) which culminated in viridis, which is shipping in recent versions of matplotlib and will become the default.
In support of this, he built a tool called viscm which can generate reduced versions of colormaps to show what they would be like with varying degrees of insensitivity to color. I've generated outputs from viscm of three of the custom colormaps we ship with yt:
Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I believe this is now also shipped with MPL) Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/
I love algae, but it's not the best from an accessibility perspective.
I'd like to propose that we use a new default colormap. If we do this, I see two options:
- Retain a "branding" by developing a new one either by using the
techniques used by matplotlib (or one of the maps they opted not to use) or by modifying algae to be more accessible; looking at the response functions, I suspect it would be reasonably possible to modify it. (Modifying algae is my preference.)
- Use viridis (which we may then have to ship if we have older
versions of matplotlib to support)
-Matt _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
There is a really excellent paper on designing color maps called "Color Sequences for Univariate Maps: Theory, Experiments, and Principles" that you can get here: http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq...
If we design a new colormap, this would be a good reference along with those scipy resources. I personally would love to have an accessible, yt-custom colormap.
On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter schnetter@gmail.com wrote:
I think there are several colourmaps that were created when Viridis was invented. I personally like Inferno.
-erik
On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum nathan12343@gmail.com wrote:
I would also be for coming up with our own colormap. That said, I think simply modifying algae won't be enough, since it is too perceptually nonlinear.
On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com wrote:
I would go for modifying algae.
On Jan 6, 2016, at 11:30 AM, Matthew Turk matthewturk@gmail.com
wrote:
Hi folks,
For a long time we've used "algae," which was designed by Britton about eight years ago, as the default colormap. This has been really nice for "branding" yt -- if you see an algae plot, it's probably (not definitely) made with yt. But it's also not accessible from a colorblindness perspective. Stefan van der Walt has been giving some really great talks lately about building a better colormap for matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU ) which culminated in viridis, which is shipping in recent versions of matplotlib and will become the default.
In support of this, he built a tool called viscm which can generate reduced versions of colormaps to show what they would be like with varying degrees of insensitivity to color. I've generated outputs from viscm of three of the custom colormaps we ship with yt:
Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I believe this is now also shipped with MPL) Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/
I love algae, but it's not the best from an accessibility perspective.
I'd like to propose that we use a new default colormap. If we do this, I see two options:
- Retain a "branding" by developing a new one either by using the
techniques used by matplotlib (or one of the maps they opted not to use) or by modifying algae to be more accessible; looking at the response functions, I suspect it would be reasonably possible to modify it. (Modifying algae is my preference.)
- Use viridis (which we may then have to ship if we have older
versions of matplotlib to support)
-Matt _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
-- Erik Schnetter schnetter@gmail.com http://www.perimeterinstitute.ca/personal/eschnetter/ _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
I always liked this reference:
http://www.kennethmoreland.com/color-maps/ColorMapsExpanded.pdf
On Wed, Jan 6, 2016 at 1:13 PM, B.W. Keller kellerbw@mcmaster.ca wrote:
There is a really excellent paper on designing color maps called "Color Sequences for Univariate Maps: Theory, Experiments, and Principles" that you can get here: http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq...
If we design a new colormap, this would be a good reference along with those scipy resources. I personally would love to have an accessible, yt-custom colormap.
On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter schnetter@gmail.com wrote:
I think there are several colourmaps that were created when Viridis was invented. I personally like Inferno.
-erik
On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum nathan12343@gmail.com wrote:
I would also be for coming up with our own colormap. That said, I think simply modifying algae won't be enough, since it is too perceptually nonlinear.
On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com wrote:
I would go for modifying algae.
On Jan 6, 2016, at 11:30 AM, Matthew Turk matthewturk@gmail.com
wrote:
Hi folks,
For a long time we've used "algae," which was designed by Britton about eight years ago, as the default colormap. This has been really nice for "branding" yt -- if you see an algae plot, it's probably
(not
definitely) made with yt. But it's also not accessible from a colorblindness perspective. Stefan van der Walt has been giving some really great talks lately about building a better colormap for matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU )
which
culminated in viridis, which is shipping in recent versions of matplotlib and will become the default.
In support of this, he built a tool called viscm which can generate reduced versions of colormaps to show what they would be like with varying degrees of insensitivity to color. I've generated outputs from viscm of three of the custom colormaps we ship with yt:
Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I believe this is now also shipped with MPL) Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/
I love algae, but it's not the best from an accessibility
perspective.
I'd like to propose that we use a new default colormap. If we do this, I see two options:
- Retain a "branding" by developing a new one either by using the
techniques used by matplotlib (or one of the maps they opted not to use) or by modifying algae to be more accessible; looking at the response functions, I suspect it would be reasonably possible to modify it. (Modifying algae is my preference.)
- Use viridis (which we may then have to ship if we have older
versions of matplotlib to support)
-Matt _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
-- Erik Schnetter schnetter@gmail.com http://www.perimeterinstitute.ca/personal/eschnetter/ _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
There was a fair bit of discussion about colormaps - terrible, useful, beautiful - at IEEE Vis last October. The viridis colormap was a featured one. So was the traditional rainbow, which lots of info-vis and perceptual people piled on to criticize.
Among design criteria for a continuous-valued colormap is whether it's "sequential" (like the typical yt colormap, or viridis) or "diverging". You'd want a diverging colormap to show signed deviations from a norm - where the eye should be caught by places where a value is either much less than, or much more than, something in the middle. Is it worth offering a typical divergent colormap, as well as a new typical sequential one, in yt?
Note that among the Stefan van der Walt & Nathaniel Smith writeup ( http://bids.github.io/colormap/ ) on their development of better cmaps, they use Nathan Goldbaum's galaxy evolution as a test case for six (sequential) examples! => http://vorpus.org/%7Enjs/goldbaum-galaxies-all-colormaps.mkv
A neat web site with sample colormaps - aimed at mapping discrete values on geographic maps, so not directly applicable but cool - is this, by Cynthia Brewer and Mark Harrower at PSU: http://colorbrewer2.org/ It has a library of predesigned cmaps, and lets you sift them by being colorblind-safe, photocopy safe, etc. ________________________________ From: yt-dev [yt-dev-bounces@lists.spacepope.org] on behalf of B.W. Keller [kellerbw@mcmaster.ca] Sent: Wednesday, January 06, 2016 12:13 To: yt-dev@lists.spacepope.org Subject: Re: [yt-dev] Default colormap
There is a really excellent paper on designing color maps called "Color Sequences for Univariate Maps: Theory, Experiments, and Principles" that you can get here: http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq...https://urldefense.proofpoint.com/v2/url?u=http-3A__ccom.unh.edu_sites_default_files_publications_Ware-5F1988-5FCGA-5FColor-5Fsequences-5Funivariate-5Fmaps.pdf&d=BQMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=hgcBC3x6dKFoTrmFmMYYbKNfiHZlGLKliIidd1LwmHI&m=FMHeB3pNOAeqOzZIyPgy5-uxuN2ghGC9-8XQuthPiIU&s=5Yg0N-A3hbVCKwltZjUQBJOQZK6OFbo4G3VWKT9uVZ0&e=
If we design a new colormap, this would be a good reference along with those scipy resources. I personally would love to have an accessible, yt-custom colormap.
On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter <schnetter@gmail.commailto:schnetter@gmail.com> wrote: I think there are several colourmaps that were created when Viridis was invented. I personally like Inferno.
-erik
On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum <nathan12343@gmail.commailto:nathan12343@gmail.com> wrote:
I would also be for coming up with our own colormap. That said, I think simply modifying algae won't be enough, since it is too perceptually nonlinear.
On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone <jzuhone@gmail.commailto:jzuhone@gmail.com> wrote:
I would go for modifying algae.
On Jan 6, 2016, at 11:30 AM, Matthew Turk <matthewturk@gmail.commailto:matthewturk@gmail.com> wrote:
Hi folks,
For a long time we've used "algae," which was designed by Britton about eight years ago, as the default colormap. This has been really nice for "branding" yt -- if you see an algae plot, it's probably (not definitely) made with yt. But it's also not accessible from a colorblindness perspective. Stefan van der Walt has been giving some really great talks lately about building a better colormap for matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lUhttps://urldefense.proofpoint.com/v2/url?u=https-3A__www.youtube.com_watch-3Fv-3DxAoljeRJ3lU&d=BQMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=hgcBC3x6dKFoTrmFmMYYbKNfiHZlGLKliIidd1LwmHI&m=FMHeB3pNOAeqOzZIyPgy5-uxuN2ghGC9-8XQuthPiIU&s=fpeIMGnkAI5_PfL699BNCowLgIQr4xXy-oK90Bub2MU&e= ) which culminated in viridis, which is shipping in recent versions of matplotlib and will become the default.
In support of this, he built a tool called viscm which can generate reduced versions of colormaps to show what they would be like with varying degrees of insensitivity to color. I've generated outputs from viscm of three of the custom colormaps we ship with yt:
Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/https://urldefense.proofpoint.com/v2/url?u=https-3A__images.hub.yt_u_fido_m_d275d5e1-2Dpng_&d=BQMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=hgcBC3x6dKFoTrmFmMYYbKNfiHZlGLKliIidd1LwmHI&m=FMHeB3pNOAeqOzZIyPgy5-uxuN2ghGC9-8XQuthPiIU&s=5En_z8QNwijy3Ou9Ah6_Fzfdp8gSXcs4bphf-SZBGuE&e= Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/https://urldefense.proofpoint.com/v2/url?u=https-3A__images.hub.yt_u_fido_m_8e698928-2Dpng_&d=BQMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=hgcBC3x6dKFoTrmFmMYYbKNfiHZlGLKliIidd1LwmHI&m=FMHeB3pNOAeqOzZIyPgy5-uxuN2ghGC9-8XQuthPiIU&s=y4k79vB300hkxSOvNBPaMYSeMSdgWY3yJP2TJ1e61YA&e= (I believe this is now also shipped with MPL) Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/https://urldefense.proofpoint.com/v2/url?u=https-3A__images.hub.yt_u_fido_m_e0e40efa-2Dpng_&d=BQMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=hgcBC3x6dKFoTrmFmMYYbKNfiHZlGLKliIidd1LwmHI&m=FMHeB3pNOAeqOzZIyPgy5-uxuN2ghGC9-8XQuthPiIU&s=xvnX_JDGJFjLaPam1gf5BwfAS4lF7qjJ5CLmd6ExJzQ&e=
I love algae, but it's not the best from an accessibility perspective.
I'd like to propose that we use a new default colormap. If we do this, I see two options:
- Retain a "branding" by developing a new one either by using the
techniques used by matplotlib (or one of the maps they opted not to use) or by modifying algae to be more accessible; looking at the response functions, I suspect it would be reasonably possible to modify it. (Modifying algae is my preference.)
- Use viridis (which we may then have to ship if we have older
versions of matplotlib to support)
-Matt _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.orgmailto:yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.orghttps://urldefense.proofpoint.com/v2/url?u=http-3A__lists.spacepope.org_listinfo.cgi_yt-2Ddev-2Dspacepope.org&d=BQMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=hgcBC3x6dKFoTrmFmMYYbKNfiHZlGLKliIidd1LwmHI&m=FMHeB3pNOAeqOzZIyPgy5-uxuN2ghGC9-8XQuthPiIU&s=BaeHL9R3SqrWDRh-sTc8pLPmBg83iPA2WHUwC8XJkdM&e=
yt-dev mailing list yt-dev@lists.spacepope.orgmailto:yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.orghttps://urldefense.proofpoint.com/v2/url?u=http-3A__lists.spacepope.org_listinfo.cgi_yt-2Ddev-2Dspacepope.org&d=BQMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=hgcBC3x6dKFoTrmFmMYYbKNfiHZlGLKliIidd1LwmHI&m=FMHeB3pNOAeqOzZIyPgy5-uxuN2ghGC9-8XQuthPiIU&s=BaeHL9R3SqrWDRh-sTc8pLPmBg83iPA2WHUwC8XJkdM&e=
yt-dev mailing list yt-dev@lists.spacepope.orgmailto:yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.orghttps://urldefense.proofpoint.com/v2/url?u=http-3A__lists.spacepope.org_listinfo.cgi_yt-2Ddev-2Dspacepope.org&d=BQMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=hgcBC3x6dKFoTrmFmMYYbKNfiHZlGLKliIidd1LwmHI&m=FMHeB3pNOAeqOzZIyPgy5-uxuN2ghGC9-8XQuthPiIU&s=BaeHL9R3SqrWDRh-sTc8pLPmBg83iPA2WHUwC8XJkdM&e=
-- Erik Schnetter <schnetter@gmail.commailto:schnetter@gmail.com> http://www.perimeterinstitute.ca/personal/eschnetter/https://urldefense.proofpoint.com/v2/url?u=http-3A__www.perimeterinstitute.ca_personal_eschnetter_&d=BQMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=hgcBC3x6dKFoTrmFmMYYbKNfiHZlGLKliIidd1LwmHI&m=FMHeB3pNOAeqOzZIyPgy5-uxuN2ghGC9-8XQuthPiIU&s=7jImD2ypg2lTI_MTSODmVdgQnwBMJvGou2W8_PSAADg&e= _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.orgmailto:yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.orghttps://urldefense.proofpoint.com/v2/url?u=http-3A__lists.spacepope.org_listinfo.cgi_yt-2Ddev-2Dspacepope.org&d=BQMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=hgcBC3x6dKFoTrmFmMYYbKNfiHZlGLKliIidd1LwmHI&m=FMHeB3pNOAeqOzZIyPgy5-uxuN2ghGC9-8XQuthPiIU&s=BaeHL9R3SqrWDRh-sTc8pLPmBg83iPA2WHUwC8XJkdM&e=
_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
I am in favor of retraining a unique yt colormap but making it work better than algae for the reasons already mentioned. I'd like to point out that about a year ago I added the capacity in yt to create custom colormaps that could easily be used for this purpose.
http://yt-project.org/docs/dev/visualizing/colormaps/index.html?highlight=co...
Cameron
On Wed, Jan 6, 2016 at 1:48 PM, Levy, Stuart A salevy@illinois.edu wrote:
There was a fair bit of discussion about colormaps - terrible, useful, beautiful - at IEEE Vis last October. The viridis colormap was a featured one. So was the traditional rainbow, which lots of info-vis and perceptual people piled on to criticize.
Among design criteria for a continuous-valued colormap is whether it's "sequential" (like the typical yt colormap, or viridis) or "diverging". You'd want a diverging colormap to show signed deviations from a norm - where the eye should be caught by places where a value is either much less than, or much more than, something in the middle. Is it worth offering a typical divergent colormap, as well as a new typical sequential one, in yt?
Note that among the Stefan van der Walt & Nathaniel Smith writeup ( http://bids.github.io/colormap/ ) on their development of better cmaps, they use Nathan Goldbaum's galaxy evolution as a test case for six (sequential) examples! => http://vorpus.org/%7Enjs/goldbaum-galaxies-all-colormaps.mkv
A neat web site with sample colormaps - aimed at mapping discrete values on geographic maps, so not directly applicable but cool - is this, by Cynthia Brewer and Mark Harrower at PSU: http://colorbrewer2.org/ It has a library of predesigned cmaps, and lets you sift them by being colorblind-safe, photocopy safe, etc.
*From:* yt-dev [yt-dev-bounces@lists.spacepope.org] on behalf of B.W. Keller [kellerbw@mcmaster.ca] *Sent:* Wednesday, January 06, 2016 12:13 *To:* yt-dev@lists.spacepope.org *Subject:* Re: [yt-dev] Default colormap
There is a really excellent paper on designing color maps called "Color Sequences for Univariate Maps: Theory, Experiments, and Principles" that you can get here: http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq... https://urldefense.proofpoint.com/v2/url?u=http-3A__ccom.unh.edu_sites_default_files_publications_Ware-5F1988-5FCGA-5FColor-5Fsequences-5Funivariate-5Fmaps.pdf&d=BQMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=hgcBC3x6dKFoTrmFmMYYbKNfiHZlGLKliIidd1LwmHI&m=FMHeB3pNOAeqOzZIyPgy5-uxuN2ghGC9-8XQuthPiIU&s=5Yg0N-A3hbVCKwltZjUQBJOQZK6OFbo4G3VWKT9uVZ0&e=
If we design a new colormap, this would be a good reference along with those scipy resources. I personally would love to have an accessible, yt-custom colormap.
On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter schnetter@gmail.com wrote:
I think there are several colourmaps that were created when Viridis was invented. I personally like Inferno.
-erik
On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum nathan12343@gmail.com wrote:
I would also be for coming up with our own colormap. That said, I think simply modifying algae won't be enough, since it is too perceptually nonlinear.
On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com wrote:
I would go for modifying algae.
On Jan 6, 2016, at 11:30 AM, Matthew Turk matthewturk@gmail.com
wrote:
Hi folks,
For a long time we've used "algae," which was designed by Britton about eight years ago, as the default colormap. This has been really nice for "branding" yt -- if you see an algae plot, it's probably
(not
definitely) made with yt. But it's also not accessible from a colorblindness perspective. Stefan van der Walt has been giving some really great talks lately about building a better colormap for matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU
culminated in viridis, which is shipping in recent versions of matplotlib and will become the default.
In support of this, he built a tool called viscm which can generate reduced versions of colormaps to show what they would be like with varying degrees of insensitivity to color. I've generated outputs from viscm of three of the custom colormaps we ship with yt:
Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/
this is now also shipped with MPL) Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/
I love algae, but it's not the best from an accessibility
perspective.
I'd like to propose that we use a new default colormap. If we do this, I see two options:
- Retain a "branding" by developing a new one either by using the
techniques used by matplotlib (or one of the maps they opted not to use) or by modifying algae to be more accessible; looking at the response functions, I suspect it would be reasonably possible to modify it. (Modifying algae is my preference.)
- Use viridis (which we may then have to ship if we have older
versions of matplotlib to support)
-Matt _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
-- Erik Schnetter schnetter@gmail.com http://www.perimeterinstitute.ca/personal/eschnetter/ https://urldefense.proofpoint.com/v2/url?u=http-3A__www.perimeterinstitute.ca_personal_eschnetter_&d=BQMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=hgcBC3x6dKFoTrmFmMYYbKNfiHZlGLKliIidd1LwmHI&m=FMHeB3pNOAeqOzZIyPgy5-uxuN2ghGC9-8XQuthPiIU&s=7jImD2ypg2lTI_MTSODmVdgQnwBMJvGou2W8_PSAADg&e= _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.spacepope.org_listinfo.cgi_yt-2Ddev-2Dspacepope.org&d=BQMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=hgcBC3x6dKFoTrmFmMYYbKNfiHZlGLKliIidd1LwmHI&m=FMHeB3pNOAeqOzZIyPgy5-uxuN2ghGC9-8XQuthPiIU&s=BaeHL9R3SqrWDRh-sTc8pLPmBg83iPA2WHUwC8XJkdM&e=
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
Hi Stuart and everyone else,
This is great info. I appreciate everyone's thoughtful replies.
Having both a sequential colormap (which would replace algae) and a diverging colormap, would be awesome. The Paraview devs shipped the new matplotlib ones (like Inferno) in 5.0. I think it would be a fun experiment to see if we can come up with something sufficiently "branded" or different. And then if we can't, fall back on something like Inferno?
-Matt
On Wed, Jan 6, 2016 at 12:48 PM, Levy, Stuart A salevy@illinois.edu wrote:
There was a fair bit of discussion about colormaps - terrible, useful, beautiful - at IEEE Vis last October. The viridis colormap was a featured one. So was the traditional rainbow, which lots of info-vis and perceptual people piled on to criticize.
Among design criteria for a continuous-valued colormap is whether it's "sequential" (like the typical yt colormap, or viridis) or "diverging". You'd want a diverging colormap to show signed deviations from a norm - where the eye should be caught by places where a value is either much less than, or much more than, something in the middle. Is it worth offering a typical divergent colormap, as well as a new typical sequential one, in yt?
Note that among the Stefan van der Walt & Nathaniel Smith writeup ( http://bids.github.io/colormap/ ) on their development of better cmaps, they use Nathan Goldbaum's galaxy evolution as a test case for six (sequential) examples! => http://vorpus.org/%7Enjs/goldbaum-galaxies-all-colormaps.mkv
A neat web site with sample colormaps - aimed at mapping discrete values on geographic maps, so not directly applicable but cool - is this, by Cynthia Brewer and Mark Harrower at PSU: http://colorbrewer2.org/ It has a library of predesigned cmaps, and lets you sift them by being colorblind-safe, photocopy safe, etc. ________________________________ From: yt-dev [yt-dev-bounces@lists.spacepope.org] on behalf of B.W. Keller [kellerbw@mcmaster.ca] Sent: Wednesday, January 06, 2016 12:13 To: yt-dev@lists.spacepope.org Subject: Re: [yt-dev] Default colormap
There is a really excellent paper on designing color maps called "Color Sequences for Univariate Maps: Theory, Experiments, and Principles" that you can get here: http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq...
If we design a new colormap, this would be a good reference along with those scipy resources. I personally would love to have an accessible, yt-custom colormap.
On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter schnetter@gmail.com wrote:
I think there are several colourmaps that were created when Viridis was invented. I personally like Inferno.
-erik
On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum nathan12343@gmail.com wrote:
I would also be for coming up with our own colormap. That said, I think simply modifying algae won't be enough, since it is too perceptually nonlinear.
On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com wrote:
I would go for modifying algae.
On Jan 6, 2016, at 11:30 AM, Matthew Turk matthewturk@gmail.com wrote:
Hi folks,
For a long time we've used "algae," which was designed by Britton about eight years ago, as the default colormap. This has been really nice for "branding" yt -- if you see an algae plot, it's probably (not definitely) made with yt. But it's also not accessible from a colorblindness perspective. Stefan van der Walt has been giving some really great talks lately about building a better colormap for matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU ) which culminated in viridis, which is shipping in recent versions of matplotlib and will become the default.
In support of this, he built a tool called viscm which can generate reduced versions of colormaps to show what they would be like with varying degrees of insensitivity to color. I've generated outputs from viscm of three of the custom colormaps we ship with yt:
Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I believe this is now also shipped with MPL) Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/
I love algae, but it's not the best from an accessibility perspective.
I'd like to propose that we use a new default colormap. If we do this, I see two options:
- Retain a "branding" by developing a new one either by using the
techniques used by matplotlib (or one of the maps they opted not to use) or by modifying algae to be more accessible; looking at the response functions, I suspect it would be reasonably possible to modify it. (Modifying algae is my preference.)
- Use viridis (which we may then have to ship if we have older
versions of matplotlib to support)
-Matt _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
-- Erik Schnetter schnetter@gmail.com http://www.perimeterinstitute.ca/personal/eschnetter/ _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
Hi all,
I've experimented a bit and come up with this:
https://images.hub.yt/u/fido/m/9bbe3cf6-png/
The script:
http://paste.yt-project.org/show/6151/
This was designed with the viscm project, which is awfully cool. What do folks think? I think Kacper and Nathan also experimented with viscm and have some ideas too, so maybe we should put it up for an eventual vote.
Also, I would campaign for calling whatever our new colormap turns out to be one of these three things, in increasing order of my preference:
agar kelp kanten
-Matt
On Thu, Jan 14, 2016 at 10:57 AM, Matthew Turk matthewturk@gmail.com wrote:
Hi Stuart and everyone else,
This is great info. I appreciate everyone's thoughtful replies.
Having both a sequential colormap (which would replace algae) and a diverging colormap, would be awesome. The Paraview devs shipped the new matplotlib ones (like Inferno) in 5.0. I think it would be a fun experiment to see if we can come up with something sufficiently "branded" or different. And then if we can't, fall back on something like Inferno?
-Matt
On Wed, Jan 6, 2016 at 12:48 PM, Levy, Stuart A salevy@illinois.edu wrote:
There was a fair bit of discussion about colormaps - terrible, useful, beautiful - at IEEE Vis last October. The viridis colormap was a featured one. So was the traditional rainbow, which lots of info-vis and perceptual people piled on to criticize.
Among design criteria for a continuous-valued colormap is whether it's "sequential" (like the typical yt colormap, or viridis) or "diverging". You'd want a diverging colormap to show signed deviations from a norm - where the eye should be caught by places where a value is either much less than, or much more than, something in the middle. Is it worth offering a typical divergent colormap, as well as a new typical sequential one, in yt?
Note that among the Stefan van der Walt & Nathaniel Smith writeup ( http://bids.github.io/colormap/ ) on their development of better cmaps, they use Nathan Goldbaum's galaxy evolution as a test case for six (sequential) examples! => http://vorpus.org/%7Enjs/goldbaum-galaxies-all-colormaps.mkv
A neat web site with sample colormaps - aimed at mapping discrete values on geographic maps, so not directly applicable but cool - is this, by Cynthia Brewer and Mark Harrower at PSU: http://colorbrewer2.org/ It has a library of predesigned cmaps, and lets you sift them by being colorblind-safe, photocopy safe, etc. ________________________________ From: yt-dev [yt-dev-bounces@lists.spacepope.org] on behalf of B.W. Keller [kellerbw@mcmaster.ca] Sent: Wednesday, January 06, 2016 12:13 To: yt-dev@lists.spacepope.org Subject: Re: [yt-dev] Default colormap
There is a really excellent paper on designing color maps called "Color Sequences for Univariate Maps: Theory, Experiments, and Principles" that you can get here: http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq...
If we design a new colormap, this would be a good reference along with those scipy resources. I personally would love to have an accessible, yt-custom colormap.
On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter schnetter@gmail.com wrote:
I think there are several colourmaps that were created when Viridis was invented. I personally like Inferno.
-erik
On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum nathan12343@gmail.com wrote:
I would also be for coming up with our own colormap. That said, I think simply modifying algae won't be enough, since it is too perceptually nonlinear.
On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com wrote:
I would go for modifying algae.
On Jan 6, 2016, at 11:30 AM, Matthew Turk matthewturk@gmail.com wrote:
Hi folks,
For a long time we've used "algae," which was designed by Britton about eight years ago, as the default colormap. This has been really nice for "branding" yt -- if you see an algae plot, it's probably (not definitely) made with yt. But it's also not accessible from a colorblindness perspective. Stefan van der Walt has been giving some really great talks lately about building a better colormap for matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU ) which culminated in viridis, which is shipping in recent versions of matplotlib and will become the default.
In support of this, he built a tool called viscm which can generate reduced versions of colormaps to show what they would be like with varying degrees of insensitivity to color. I've generated outputs from viscm of three of the custom colormaps we ship with yt:
Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I believe this is now also shipped with MPL) Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/
I love algae, but it's not the best from an accessibility perspective.
I'd like to propose that we use a new default colormap. If we do this, I see two options:
- Retain a "branding" by developing a new one either by using the
techniques used by matplotlib (or one of the maps they opted not to use) or by modifying algae to be more accessible; looking at the response functions, I suspect it would be reasonably possible to modify it. (Modifying algae is my preference.)
- Use viridis (which we may then have to ship if we have older
versions of matplotlib to support)
-Matt _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
-- Erik Schnetter schnetter@gmail.com http://www.perimeterinstitute.ca/personal/eschnetter/ _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
The one I came up with is here:
The script is here:
http://paste.yt-project.org/show/6165/
On Mon, Jan 18, 2016 at 9:45 AM, Matthew Turk matthewturk@gmail.com wrote:
Hi all,
I've experimented a bit and come up with this:
https://images.hub.yt/u/fido/m/9bbe3cf6-png/
The script:
http://paste.yt-project.org/show/6151/
This was designed with the viscm project, which is awfully cool. What do folks think? I think Kacper and Nathan also experimented with viscm and have some ideas too, so maybe we should put it up for an eventual vote.
Also, I would campaign for calling whatever our new colormap turns out to be one of these three things, in increasing order of my preference:
agar kelp kanten
-Matt
On Thu, Jan 14, 2016 at 10:57 AM, Matthew Turk matthewturk@gmail.com wrote:
Hi Stuart and everyone else,
This is great info. I appreciate everyone's thoughtful replies.
Having both a sequential colormap (which would replace algae) and a diverging colormap, would be awesome. The Paraview devs shipped the new matplotlib ones (like Inferno) in 5.0. I think it would be a fun experiment to see if we can come up with something sufficiently "branded" or different. And then if we can't, fall back on something like Inferno?
-Matt
On Wed, Jan 6, 2016 at 12:48 PM, Levy, Stuart A salevy@illinois.edu
wrote:
There was a fair bit of discussion about colormaps - terrible, useful, beautiful - at IEEE Vis last October. The viridis colormap was a
featured
one. So was the traditional rainbow, which lots of info-vis and
perceptual
people piled on to criticize.
Among design criteria for a continuous-valued colormap is whether it's "sequential" (like the typical yt colormap, or viridis) or "diverging". You'd want a diverging colormap to show signed deviations from a norm - where the eye should be caught by places where a value is either much
less
than, or much more than, something in the middle. Is it worth
offering a
typical divergent colormap, as well as a new typical sequential one, in
yt?
Note that among the Stefan van der Walt & Nathaniel Smith writeup ( http://bids.github.io/colormap/ ) on their development of better
cmaps, they
use Nathan Goldbaum's galaxy evolution as a test case for six
(sequential)
examples! => http://vorpus.org/%7Enjs/goldbaum-galaxies-all-colormaps.mkv
A neat web site with sample colormaps - aimed at mapping discrete
values on
geographic maps, so not directly applicable but cool - is this, by
Cynthia
Brewer and Mark Harrower at PSU: http://colorbrewer2.org/ It has a library of predesigned cmaps, and lets you sift them by being colorblind-safe, photocopy safe, etc. ________________________________ From: yt-dev [yt-dev-bounces@lists.spacepope.org] on behalf of B.W.
Keller
[kellerbw@mcmaster.ca] Sent: Wednesday, January 06, 2016 12:13 To: yt-dev@lists.spacepope.org Subject: Re: [yt-dev] Default colormap
There is a really excellent paper on designing color maps called "Color Sequences for Univariate Maps: Theory, Experiments, and Principles"
that you
can get here:
http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq...
If we design a new colormap, this would be a good reference along with
those
scipy resources. I personally would love to have an accessible,
yt-custom
colormap.
On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter schnetter@gmail.com
wrote:
I think there are several colourmaps that were created when Viridis was invented. I personally like Inferno.
-erik
On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum <
nathan12343@gmail.com>
wrote:
I would also be for coming up with our own colormap. That said, I
think
simply modifying algae won't be enough, since it is too perceptually nonlinear.
On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com
wrote:
I would go for modifying algae.
> On Jan 6, 2016, at 11:30 AM, Matthew Turk matthewturk@gmail.com > wrote: > > Hi folks, > > For a long time we've used "algae," which was designed by Britton > about eight years ago, as the default colormap. This has been
really
> nice for "branding" yt -- if you see an algae plot, it's probably > (not > definitely) made with yt. But it's also not accessible from a > colorblindness perspective. Stefan van der Walt has been giving
some
> really great talks lately about building a better colormap for > matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU )
which
> culminated in viridis, which is shipping in recent versions of > matplotlib and will become the default. > > In support of this, he built a tool called viscm which can
generate
> reduced versions of colormaps to show what they would be like with > varying degrees of insensitivity to color. I've generated outputs > from viscm of three of the custom colormaps we ship with yt: > > Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ > Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I
believe
> this is now also shipped with MPL) > Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/ > > I love algae, but it's not the best from an accessibility > perspective. > > I'd like to propose that we use a new default colormap. If we do > this, I see two options: > > * Retain a "branding" by developing a new one either by using the > techniques used by matplotlib (or one of the maps they opted not
to
> use) or by modifying algae to be more accessible; looking at the > response functions, I suspect it would be reasonably possible to > modify it. (Modifying algae is my preference.) > * Use viridis (which we may then have to ship if we have older > versions of matplotlib to support) > > -Matt > _______________________________________________ > yt-dev mailing list > yt-dev@lists.spacepope.org > http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
-- Erik Schnetter schnetter@gmail.com http://www.perimeterinstitute.ca/personal/eschnetter/ _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
On 01/18/2016 09:45 AM, Matthew Turk wrote:
Hi all,
I've experimented a bit and come up with this:
https://images.hub.yt/u/fido/m/9bbe3cf6-png/
The script:
http://paste.yt-project.org/show/6151/
This was designed with the viscm project, which is awfully cool. What do folks think? I think Kacper and Nathan also experimented with viscm and have some ideas too, so maybe we should put it up for an eventual vote.
This is my experiment:
https://images.hub.yt/u/fido/m/f180a901-png/
Source:
http://paste.yt-project.org/show/6166/
Cheers, Kacper
Also, I would campaign for calling whatever our new colormap turns out to be one of these three things, in increasing order of my preference:
agar kelp kanten
-Matt
On Thu, Jan 14, 2016 at 10:57 AM, Matthew Turk matthewturk@gmail.com wrote:
Hi Stuart and everyone else,
This is great info. I appreciate everyone's thoughtful replies.
Having both a sequential colormap (which would replace algae) and a diverging colormap, would be awesome. The Paraview devs shipped the new matplotlib ones (like Inferno) in 5.0. I think it would be a fun experiment to see if we can come up with something sufficiently "branded" or different. And then if we can't, fall back on something like Inferno?
-Matt
On Wed, Jan 6, 2016 at 12:48 PM, Levy, Stuart A salevy@illinois.edu wrote:
There was a fair bit of discussion about colormaps - terrible, useful, beautiful - at IEEE Vis last October. The viridis colormap was a featured one. So was the traditional rainbow, which lots of info-vis and perceptual people piled on to criticize.
Among design criteria for a continuous-valued colormap is whether it's "sequential" (like the typical yt colormap, or viridis) or "diverging". You'd want a diverging colormap to show signed deviations from a norm - where the eye should be caught by places where a value is either much less than, or much more than, something in the middle. Is it worth offering a typical divergent colormap, as well as a new typical sequential one, in yt?
Note that among the Stefan van der Walt & Nathaniel Smith writeup ( http://bids.github.io/colormap/ ) on their development of better cmaps, they use Nathan Goldbaum's galaxy evolution as a test case for six (sequential) examples! => http://vorpus.org/%7Enjs/goldbaum-galaxies-all-colormaps.mkv
A neat web site with sample colormaps - aimed at mapping discrete values on geographic maps, so not directly applicable but cool - is this, by Cynthia Brewer and Mark Harrower at PSU: http://colorbrewer2.org/ It has a library of predesigned cmaps, and lets you sift them by being colorblind-safe, photocopy safe, etc. ________________________________ From: yt-dev [yt-dev-bounces@lists.spacepope.org] on behalf of B.W. Keller [kellerbw@mcmaster.ca] Sent: Wednesday, January 06, 2016 12:13 To: yt-dev@lists.spacepope.org Subject: Re: [yt-dev] Default colormap
There is a really excellent paper on designing color maps called "Color Sequences for Univariate Maps: Theory, Experiments, and Principles" that you can get here: http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq...
If we design a new colormap, this would be a good reference along with those scipy resources. I personally would love to have an accessible, yt-custom colormap.
On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter schnetter@gmail.com wrote:
I think there are several colourmaps that were created when Viridis was invented. I personally like Inferno.
-erik
On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum nathan12343@gmail.com wrote:
I would also be for coming up with our own colormap. That said, I think simply modifying algae won't be enough, since it is too perceptually nonlinear.
On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com wrote:
I would go for modifying algae.
> On Jan 6, 2016, at 11:30 AM, Matthew Turk matthewturk@gmail.com > wrote: > > Hi folks, > > For a long time we've used "algae," which was designed by Britton > about eight years ago, as the default colormap. This has been really > nice for "branding" yt -- if you see an algae plot, it's probably > (not > definitely) made with yt. But it's also not accessible from a > colorblindness perspective. Stefan van der Walt has been giving some > really great talks lately about building a better colormap for > matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU ) which > culminated in viridis, which is shipping in recent versions of > matplotlib and will become the default. > > In support of this, he built a tool called viscm which can generate > reduced versions of colormaps to show what they would be like with > varying degrees of insensitivity to color. I've generated outputs > from viscm of three of the custom colormaps we ship with yt: > > Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ > Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I believe > this is now also shipped with MPL) > Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/ > > I love algae, but it's not the best from an accessibility > perspective. > > I'd like to propose that we use a new default colormap. If we do > this, I see two options: > > * Retain a "branding" by developing a new one either by using the > techniques used by matplotlib (or one of the maps they opted not to > use) or by modifying algae to be more accessible; looking at the > response functions, I suspect it would be reasonably possible to > modify it. (Modifying algae is my preference.) > * Use viridis (which we may then have to ship if we have older > versions of matplotlib to support) > > -Matt > _______________________________________________ > yt-dev mailing list > yt-dev@lists.spacepope.org > http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
-- Erik Schnetter schnetter@gmail.com http://www.perimeterinstitute.ca/personal/eschnetter/ _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
Wow, all of these look great. I think I like Matt's best for painting our bikeshed, but I would be happy with any of them.
On Mon, Jan 18, 2016 at 11:10 AM, Kacper Kowalik xarthisius.kk@gmail.com wrote:
On 01/18/2016 09:45 AM, Matthew Turk wrote:
Hi all,
I've experimented a bit and come up with this:
https://images.hub.yt/u/fido/m/9bbe3cf6-png/
The script:
http://paste.yt-project.org/show/6151/
This was designed with the viscm project, which is awfully cool. What do folks think? I think Kacper and Nathan also experimented with viscm and have some ideas too, so maybe we should put it up for an eventual vote.
This is my experiment:
https://images.hub.yt/u/fido/m/f180a901-png/
Source:
http://paste.yt-project.org/show/6166/
Cheers, Kacper
Also, I would campaign for calling whatever our new colormap turns out to be one of these three things, in increasing order of my preference:
agar kelp kanten
-Matt
On Thu, Jan 14, 2016 at 10:57 AM, Matthew Turk matthewturk@gmail.com
wrote:
Hi Stuart and everyone else,
This is great info. I appreciate everyone's thoughtful replies.
Having both a sequential colormap (which would replace algae) and a diverging colormap, would be awesome. The Paraview devs shipped the new matplotlib ones (like Inferno) in 5.0. I think it would be a fun experiment to see if we can come up with something sufficiently "branded" or different. And then if we can't, fall back on something like Inferno?
-Matt
On Wed, Jan 6, 2016 at 12:48 PM, Levy, Stuart A salevy@illinois.edu
wrote:
There was a fair bit of discussion about colormaps - terrible, useful, beautiful - at IEEE Vis last October. The viridis colormap was a
featured
one. So was the traditional rainbow, which lots of info-vis and
perceptual
people piled on to criticize.
Among design criteria for a continuous-valued colormap is whether it's "sequential" (like the typical yt colormap, or viridis) or "diverging". You'd want a diverging colormap to show signed deviations from a norm - where the eye should be caught by places where a value is either much
less
than, or much more than, something in the middle. Is it worth
offering a
typical divergent colormap, as well as a new typical sequential one,
in yt?
Note that among the Stefan van der Walt & Nathaniel Smith writeup ( http://bids.github.io/colormap/ ) on their development of better
cmaps, they
use Nathan Goldbaum's galaxy evolution as a test case for six
(sequential)
examples! =>
http://vorpus.org/%7Enjs/goldbaum-galaxies-all-colormaps.mkv
A neat web site with sample colormaps - aimed at mapping discrete
values on
geographic maps, so not directly applicable but cool - is this, by
Cynthia
Brewer and Mark Harrower at PSU: http://colorbrewer2.org/ It has a library of predesigned cmaps, and lets you sift them by being colorblind-safe, photocopy safe, etc. ________________________________ From: yt-dev [yt-dev-bounces@lists.spacepope.org] on behalf of B.W.
Keller
[kellerbw@mcmaster.ca] Sent: Wednesday, January 06, 2016 12:13 To: yt-dev@lists.spacepope.org Subject: Re: [yt-dev] Default colormap
There is a really excellent paper on designing color maps called "Color Sequences for Univariate Maps: Theory, Experiments, and Principles"
that you
can get here:
http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq...
If we design a new colormap, this would be a good reference along with
those
scipy resources. I personally would love to have an accessible,
yt-custom
colormap.
On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter schnetter@gmail.com
wrote:
I think there are several colourmaps that were created when Viridis was invented. I personally like Inferno.
-erik
On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum <
nathan12343@gmail.com>
wrote:
I would also be for coming up with our own colormap. That said, I
think
simply modifying algae won't be enough, since it is too perceptually nonlinear.
On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com
wrote:
> > I would go for modifying algae. > >> On Jan 6, 2016, at 11:30 AM, Matthew Turk matthewturk@gmail.com >> wrote: >> >> Hi folks, >> >> For a long time we've used "algae," which was designed by Britton >> about eight years ago, as the default colormap. This has been
really
>> nice for "branding" yt -- if you see an algae plot, it's probably >> (not >> definitely) made with yt. But it's also not accessible from a >> colorblindness perspective. Stefan van der Walt has been giving
some
>> really great talks lately about building a better colormap for >> matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU )
which
>> culminated in viridis, which is shipping in recent versions of >> matplotlib and will become the default. >> >> In support of this, he built a tool called viscm which can generate >> reduced versions of colormaps to show what they would be like with >> varying degrees of insensitivity to color. I've generated outputs >> from viscm of three of the custom colormaps we ship with yt: >> >> Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ >> Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I believe >> this is now also shipped with MPL) >> Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/ >> >> I love algae, but it's not the best from an accessibility >> perspective. >> >> I'd like to propose that we use a new default colormap. If we do >> this, I see two options: >> >> * Retain a "branding" by developing a new one either by using the >> techniques used by matplotlib (or one of the maps they opted not to >> use) or by modifying algae to be more accessible; looking at the >> response functions, I suspect it would be reasonably possible to >> modify it. (Modifying algae is my preference.) >> * Use viridis (which we may then have to ship if we have older >> versions of matplotlib to support) >> >> -Matt >> _______________________________________________ >> yt-dev mailing list >> yt-dev@lists.spacepope.org >> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org > > _______________________________________________ > yt-dev mailing list > yt-dev@lists.spacepope.org > http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
-- Erik Schnetter schnetter@gmail.com http://www.perimeterinstitute.ca/personal/eschnetter/ _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
Hi folks,
I've put up a comparison image:
http://i.imgur.com/Afxdb0G.jpg
Left is Kacper, middle is me, right is Nathan.
Honestly I think all could go in, but we should pick a default -- whether it's one of these or a different one. Anyone have a strong opinion?
-Matt
On Mon, Jan 18, 2016 at 10:19 AM, B.W. Keller kellerbw@mcmaster.ca wrote:
Wow, all of these look great. I think I like Matt's best for painting our bikeshed, but I would be happy with any of them.
On Mon, Jan 18, 2016 at 11:10 AM, Kacper Kowalik xarthisius.kk@gmail.com wrote:
On 01/18/2016 09:45 AM, Matthew Turk wrote:
Hi all,
I've experimented a bit and come up with this:
https://images.hub.yt/u/fido/m/9bbe3cf6-png/
The script:
http://paste.yt-project.org/show/6151/
This was designed with the viscm project, which is awfully cool. What do folks think? I think Kacper and Nathan also experimented with viscm and have some ideas too, so maybe we should put it up for an eventual vote.
This is my experiment:
https://images.hub.yt/u/fido/m/f180a901-png/
Source:
http://paste.yt-project.org/show/6166/
Cheers, Kacper
Also, I would campaign for calling whatever our new colormap turns out to be one of these three things, in increasing order of my preference:
agar kelp kanten
-Matt
On Thu, Jan 14, 2016 at 10:57 AM, Matthew Turk matthewturk@gmail.com
wrote:
Hi Stuart and everyone else,
This is great info. I appreciate everyone's thoughtful replies.
Having both a sequential colormap (which would replace algae) and a diverging colormap, would be awesome. The Paraview devs shipped the new matplotlib ones (like Inferno) in 5.0. I think it would be a fun experiment to see if we can come up with something sufficiently "branded" or different. And then if we can't, fall back on something like Inferno?
-Matt
On Wed, Jan 6, 2016 at 12:48 PM, Levy, Stuart A salevy@illinois.edu
wrote:
There was a fair bit of discussion about colormaps - terrible, useful, beautiful - at IEEE Vis last October. The viridis colormap was a
featured
one. So was the traditional rainbow, which lots of info-vis and
perceptual
people piled on to criticize.
Among design criteria for a continuous-valued colormap is whether it's "sequential" (like the typical yt colormap, or viridis) or
"diverging".
You'd want a diverging colormap to show signed deviations from a norm
where the eye should be caught by places where a value is either much
less
than, or much more than, something in the middle. Is it worth
offering a
typical divergent colormap, as well as a new typical sequential one,
in yt?
Note that among the Stefan van der Walt & Nathaniel Smith writeup ( http://bids.github.io/colormap/ ) on their development of better
cmaps, they
use Nathan Goldbaum's galaxy evolution as a test case for six
(sequential)
examples! =>
http://vorpus.org/%7Enjs/goldbaum-galaxies-all-colormaps.mkv
A neat web site with sample colormaps - aimed at mapping discrete
values on
geographic maps, so not directly applicable but cool - is this, by
Cynthia
Brewer and Mark Harrower at PSU: http://colorbrewer2.org/ It has a library of predesigned cmaps, and lets you sift them by being colorblind-safe, photocopy safe, etc. ________________________________ From: yt-dev [yt-dev-bounces@lists.spacepope.org] on behalf of B.W.
Keller
[kellerbw@mcmaster.ca] Sent: Wednesday, January 06, 2016 12:13 To: yt-dev@lists.spacepope.org Subject: Re: [yt-dev] Default colormap
There is a really excellent paper on designing color maps called
"Color
Sequences for Univariate Maps: Theory, Experiments, and Principles"
that you
can get here:
http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq...
If we design a new colormap, this would be a good reference along
with those
scipy resources. I personally would love to have an accessible,
yt-custom
colormap.
On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter schnetter@gmail.com
wrote:
I think there are several colourmaps that were created when Viridis was invented. I personally like Inferno.
-erik
On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum <
nathan12343@gmail.com>
wrote: > I would also be for coming up with our own colormap. That said, I
think
> simply modifying algae won't be enough, since it is too perceptually > nonlinear. > > On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com
wrote:
>> >> I would go for modifying algae. >> >>> On Jan 6, 2016, at 11:30 AM, Matthew Turk matthewturk@gmail.com >>> wrote: >>> >>> Hi folks, >>> >>> For a long time we've used "algae," which was designed by Britton >>> about eight years ago, as the default colormap. This has been
really
>>> nice for "branding" yt -- if you see an algae plot, it's probably >>> (not >>> definitely) made with yt. But it's also not accessible from a >>> colorblindness perspective. Stefan van der Walt has been giving
some
>>> really great talks lately about building a better colormap for >>> matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU )
which
>>> culminated in viridis, which is shipping in recent versions of >>> matplotlib and will become the default. >>> >>> In support of this, he built a tool called viscm which can
generate
>>> reduced versions of colormaps to show what they would be like with >>> varying degrees of insensitivity to color. I've generated outputs >>> from viscm of three of the custom colormaps we ship with yt: >>> >>> Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ >>> Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I
believe
>>> this is now also shipped with MPL) >>> Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/ >>> >>> I love algae, but it's not the best from an accessibility >>> perspective. >>> >>> I'd like to propose that we use a new default colormap. If we do >>> this, I see two options: >>> >>> * Retain a "branding" by developing a new one either by using the >>> techniques used by matplotlib (or one of the maps they opted not
to
>>> use) or by modifying algae to be more accessible; looking at the >>> response functions, I suspect it would be reasonably possible to >>> modify it. (Modifying algae is my preference.) >>> * Use viridis (which we may then have to ship if we have older >>> versions of matplotlib to support) >>> >>> -Matt >>> _______________________________________________ >>> yt-dev mailing list >>> yt-dev@lists.spacepope.org >>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >> >> _______________________________________________ >> yt-dev mailing list >> yt-dev@lists.spacepope.org >> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org > > > > _______________________________________________ > yt-dev mailing list > yt-dev@lists.spacepope.org > http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >
-- Erik Schnetter schnetter@gmail.com http://www.perimeterinstitute.ca/personal/eschnetter/ _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
I think we should probably put it up for a vote and we should send an e-mail to yt-users about it.
On Thu, Jan 21, 2016 at 12:06 PM, Matthew Turk matthewturk@gmail.com wrote:
Hi folks,
I've put up a comparison image:
http://i.imgur.com/Afxdb0G.jpg
Left is Kacper, middle is me, right is Nathan.
Honestly I think all could go in, but we should pick a default -- whether it's one of these or a different one. Anyone have a strong opinion?
-Matt
On Mon, Jan 18, 2016 at 10:19 AM, B.W. Keller kellerbw@mcmaster.ca wrote:
Wow, all of these look great. I think I like Matt's best for painting our bikeshed, but I would be happy with any of them.
On Mon, Jan 18, 2016 at 11:10 AM, Kacper Kowalik <xarthisius.kk@gmail.com
wrote:
On 01/18/2016 09:45 AM, Matthew Turk wrote:
Hi all,
I've experimented a bit and come up with this:
https://images.hub.yt/u/fido/m/9bbe3cf6-png/
The script:
http://paste.yt-project.org/show/6151/
This was designed with the viscm project, which is awfully cool. What do folks think? I think Kacper and Nathan also experimented with viscm and have some ideas too, so maybe we should put it up for an eventual vote.
This is my experiment:
https://images.hub.yt/u/fido/m/f180a901-png/
Source:
http://paste.yt-project.org/show/6166/
Cheers, Kacper
Also, I would campaign for calling whatever our new colormap turns out to be one of these three things, in increasing order of my preference:
agar kelp kanten
-Matt
On Thu, Jan 14, 2016 at 10:57 AM, Matthew Turk matthewturk@gmail.com
wrote:
Hi Stuart and everyone else,
This is great info. I appreciate everyone's thoughtful replies.
Having both a sequential colormap (which would replace algae) and a diverging colormap, would be awesome. The Paraview devs shipped the new matplotlib ones (like Inferno) in 5.0. I think it would be a fun experiment to see if we can come up with something sufficiently "branded" or different. And then if we can't, fall back on something like Inferno?
-Matt
On Wed, Jan 6, 2016 at 12:48 PM, Levy, Stuart A salevy@illinois.edu
wrote:
There was a fair bit of discussion about colormaps - terrible,
useful,
beautiful - at IEEE Vis last October. The viridis colormap was a
featured
one. So was the traditional rainbow, which lots of info-vis and
perceptual
people piled on to criticize.
Among design criteria for a continuous-valued colormap is whether
it's
"sequential" (like the typical yt colormap, or viridis) or
"diverging".
You'd want a diverging colormap to show signed deviations from a
norm -
where the eye should be caught by places where a value is either
much less
than, or much more than, something in the middle. Is it worth
offering a
typical divergent colormap, as well as a new typical sequential one,
in yt?
Note that among the Stefan van der Walt & Nathaniel Smith writeup ( http://bids.github.io/colormap/ ) on their development of better
cmaps, they
use Nathan Goldbaum's galaxy evolution as a test case for six
(sequential)
examples! =>
http://vorpus.org/%7Enjs/goldbaum-galaxies-all-colormaps.mkv
A neat web site with sample colormaps - aimed at mapping discrete
values on
geographic maps, so not directly applicable but cool - is this, by
Cynthia
Brewer and Mark Harrower at PSU: http://colorbrewer2.org/ It has a library of predesigned cmaps, and lets you sift them by
being
colorblind-safe, photocopy safe, etc. ________________________________ From: yt-dev [yt-dev-bounces@lists.spacepope.org] on behalf of B.W.
Keller
[kellerbw@mcmaster.ca] Sent: Wednesday, January 06, 2016 12:13 To: yt-dev@lists.spacepope.org Subject: Re: [yt-dev] Default colormap
There is a really excellent paper on designing color maps called
"Color
Sequences for Univariate Maps: Theory, Experiments, and Principles"
that you
can get here:
http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq...
If we design a new colormap, this would be a good reference along
with those
scipy resources. I personally would love to have an accessible,
yt-custom
colormap.
On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter schnetter@gmail.com
wrote:
> > I think there are several colourmaps that were created when Viridis > was invented. I personally like Inferno. > > -erik > > On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum <
nathan12343@gmail.com>
> wrote: >> I would also be for coming up with our own colormap. That said, I
think
>> simply modifying algae won't be enough, since it is too
perceptually
>> nonlinear. >> >> On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com
wrote:
>>> >>> I would go for modifying algae. >>> >>>> On Jan 6, 2016, at 11:30 AM, Matthew Turk <matthewturk@gmail.com
>>>> wrote: >>>> >>>> Hi folks, >>>> >>>> For a long time we've used "algae," which was designed by Britton >>>> about eight years ago, as the default colormap. This has been
really
>>>> nice for "branding" yt -- if you see an algae plot, it's probably >>>> (not >>>> definitely) made with yt. But it's also not accessible from a >>>> colorblindness perspective. Stefan van der Walt has been giving
some
>>>> really great talks lately about building a better colormap for >>>> matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU )
which
>>>> culminated in viridis, which is shipping in recent versions of >>>> matplotlib and will become the default. >>>> >>>> In support of this, he built a tool called viscm which can
generate
>>>> reduced versions of colormaps to show what they would be like
with
>>>> varying degrees of insensitivity to color. I've generated
outputs
>>>> from viscm of three of the custom colormaps we ship with yt: >>>> >>>> Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ >>>> Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I
believe
>>>> this is now also shipped with MPL) >>>> Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/ >>>> >>>> I love algae, but it's not the best from an accessibility >>>> perspective. >>>> >>>> I'd like to propose that we use a new default colormap. If we do >>>> this, I see two options: >>>> >>>> * Retain a "branding" by developing a new one either by using the >>>> techniques used by matplotlib (or one of the maps they opted not
to
>>>> use) or by modifying algae to be more accessible; looking at the >>>> response functions, I suspect it would be reasonably possible to >>>> modify it. (Modifying algae is my preference.) >>>> * Use viridis (which we may then have to ship if we have older >>>> versions of matplotlib to support) >>>> >>>> -Matt >>>> _______________________________________________ >>>> yt-dev mailing list >>>> yt-dev@lists.spacepope.org >>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>> >>> _______________________________________________ >>> yt-dev mailing list >>> yt-dev@lists.spacepope.org >>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >> >> >> >> _______________________________________________ >> yt-dev mailing list >> yt-dev@lists.spacepope.org >> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >> > > > > -- > Erik Schnetter schnetter@gmail.com > http://www.perimeterinstitute.ca/personal/eschnetter/ > _______________________________________________ > yt-dev mailing list > yt-dev@lists.spacepope.org > http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
here's a link:
http://eric.ed.gov/?id=EJ1005874
really cool stuff.
On Thu, Jan 21, 2016 at 1:10 PM, Nathan Goldbaum nathan12343@gmail.com wrote:
I think we should probably put it up for a vote and we should send an e-mail to yt-users about it.
On Thu, Jan 21, 2016 at 12:06 PM, Matthew Turk matthewturk@gmail.com wrote:
Hi folks,
I've put up a comparison image:
http://i.imgur.com/Afxdb0G.jpg
Left is Kacper, middle is me, right is Nathan.
Honestly I think all could go in, but we should pick a default -- whether it's one of these or a different one. Anyone have a strong opinion?
-Matt
On Mon, Jan 18, 2016 at 10:19 AM, B.W. Keller kellerbw@mcmaster.ca wrote:
Wow, all of these look great. I think I like Matt's best for painting our bikeshed, but I would be happy with any of them.
On Mon, Jan 18, 2016 at 11:10 AM, Kacper Kowalik < xarthisius.kk@gmail.com> wrote:
On 01/18/2016 09:45 AM, Matthew Turk wrote:
Hi all,
I've experimented a bit and come up with this:
https://images.hub.yt/u/fido/m/9bbe3cf6-png/
The script:
http://paste.yt-project.org/show/6151/
This was designed with the viscm project, which is awfully cool. What do folks think? I think Kacper and Nathan also experimented with viscm and have some ideas too, so maybe we should put it up for an eventual vote.
This is my experiment:
https://images.hub.yt/u/fido/m/f180a901-png/
Source:
http://paste.yt-project.org/show/6166/
Cheers, Kacper
Also, I would campaign for calling whatever our new colormap turns out to be one of these three things, in increasing order of my preference:
agar kelp kanten
-Matt
On Thu, Jan 14, 2016 at 10:57 AM, Matthew Turk matthewturk@gmail.com
wrote:
Hi Stuart and everyone else,
This is great info. I appreciate everyone's thoughtful replies.
Having both a sequential colormap (which would replace algae) and a diverging colormap, would be awesome. The Paraview devs shipped the new matplotlib ones (like Inferno) in 5.0. I think it would be a fun experiment to see if we can come up with something sufficiently "branded" or different. And then if we can't, fall back on something like Inferno?
-Matt
On Wed, Jan 6, 2016 at 12:48 PM, Levy, Stuart A salevy@illinois.edu
wrote:
> There was a fair bit of discussion about colormaps - terrible,
useful,
> beautiful - at IEEE Vis last October. The viridis colormap was a
featured
> one. So was the traditional rainbow, which lots of info-vis and
perceptual
> people piled on to criticize. > > Among design criteria for a continuous-valued colormap is whether
it's
> "sequential" (like the typical yt colormap, or viridis) or
"diverging".
> You'd want a diverging colormap to show signed deviations from a
norm -
> where the eye should be caught by places where a value is either
much less
> than, or much more than, something in the middle. Is it worth
offering a
> typical divergent colormap, as well as a new typical sequential
one, in yt?
> > Note that among the Stefan van der Walt & Nathaniel Smith writeup ( > http://bids.github.io/colormap/ ) on their development of better
cmaps, they
> use Nathan Goldbaum's galaxy evolution as a test case for six
(sequential)
> examples! =>
http://vorpus.org/%7Enjs/goldbaum-galaxies-all-colormaps.mkv
> > A neat web site with sample colormaps - aimed at mapping discrete
values on
> geographic maps, so not directly applicable but cool - is this, by
Cynthia
> Brewer and Mark Harrower at PSU: > http://colorbrewer2.org/ > It has a library of predesigned cmaps, and lets you sift them by
being
> colorblind-safe, photocopy safe, etc. > ________________________________ > From: yt-dev [yt-dev-bounces@lists.spacepope.org] on behalf of
B.W. Keller
> [kellerbw@mcmaster.ca] > Sent: Wednesday, January 06, 2016 12:13 > To: yt-dev@lists.spacepope.org > Subject: Re: [yt-dev] Default colormap > > There is a really excellent paper on designing color maps called
"Color
> Sequences for Univariate Maps: Theory, Experiments, and Principles"
that you
> can get here: >
http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq...
> > If we design a new colormap, this would be a good reference along
with those
> scipy resources. I personally would love to have an accessible,
yt-custom
> colormap. > > On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter <
schnetter@gmail.com> wrote:
>> >> I think there are several colourmaps that were created when Viridis >> was invented. I personally like Inferno. >> >> -erik >> >> On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum <
nathan12343@gmail.com>
>> wrote: >>> I would also be for coming up with our own colormap. That said, I
think
>>> simply modifying algae won't be enough, since it is too
perceptually
>>> nonlinear. >>> >>> On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com
wrote:
>>>> >>>> I would go for modifying algae. >>>> >>>>> On Jan 6, 2016, at 11:30 AM, Matthew Turk <
matthewturk@gmail.com>
>>>>> wrote: >>>>> >>>>> Hi folks, >>>>> >>>>> For a long time we've used "algae," which was designed by
Britton
>>>>> about eight years ago, as the default colormap. This has been
really
>>>>> nice for "branding" yt -- if you see an algae plot, it's
probably
>>>>> (not >>>>> definitely) made with yt. But it's also not accessible from a >>>>> colorblindness perspective. Stefan van der Walt has been
giving some
>>>>> really great talks lately about building a better colormap for >>>>> matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU
) which
>>>>> culminated in viridis, which is shipping in recent versions of >>>>> matplotlib and will become the default. >>>>> >>>>> In support of this, he built a tool called viscm which can
generate
>>>>> reduced versions of colormaps to show what they would be like
with
>>>>> varying degrees of insensitivity to color. I've generated
outputs
>>>>> from viscm of three of the custom colormaps we ship with yt: >>>>> >>>>> Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ >>>>> Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I
believe
>>>>> this is now also shipped with MPL) >>>>> Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/ >>>>> >>>>> I love algae, but it's not the best from an accessibility >>>>> perspective. >>>>> >>>>> I'd like to propose that we use a new default colormap. If we
do
>>>>> this, I see two options: >>>>> >>>>> * Retain a "branding" by developing a new one either by using
the
>>>>> techniques used by matplotlib (or one of the maps they opted
not to
>>>>> use) or by modifying algae to be more accessible; looking at the >>>>> response functions, I suspect it would be reasonably possible to >>>>> modify it. (Modifying algae is my preference.) >>>>> * Use viridis (which we may then have to ship if we have older >>>>> versions of matplotlib to support) >>>>> >>>>> -Matt >>>>> _______________________________________________ >>>>> yt-dev mailing list >>>>> yt-dev@lists.spacepope.org >>>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>>> >>>> _______________________________________________ >>>> yt-dev mailing list >>>> yt-dev@lists.spacepope.org >>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>> >>> >>> >>> _______________________________________________ >>> yt-dev mailing list >>> yt-dev@lists.spacepope.org >>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>> >> >> >> >> -- >> Erik Schnetter schnetter@gmail.com >> http://www.perimeterinstitute.ca/personal/eschnetter/ >> _______________________________________________ >> yt-dev mailing list >> yt-dev@lists.spacepope.org >> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org > > > > _______________________________________________ > yt-dev mailing list > yt-dev@lists.spacepope.org > http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
Hi Mike,
Ah, awesome. I will watch that. I do know that Ralf Kaehler designed one for the AMNH's show that was inverted from what we usually think. Sam might have reverse engineered it at some point... it'd be pretty cool to explore that type of thing as well!
-Matt
PS re: the middle colormap, your check is in the mail.
On Thu, Jan 21, 2016 at 12:13 PM, Michael Zingale < michael.zingale@stonybrook.edu> wrote:
here's a link:
http://eric.ed.gov/?id=EJ1005874
really cool stuff.
On Thu, Jan 21, 2016 at 1:10 PM, Nathan Goldbaum nathan12343@gmail.com wrote:
I think we should probably put it up for a vote and we should send an e-mail to yt-users about it.
On Thu, Jan 21, 2016 at 12:06 PM, Matthew Turk matthewturk@gmail.com wrote:
Hi folks,
I've put up a comparison image:
http://i.imgur.com/Afxdb0G.jpg
Left is Kacper, middle is me, right is Nathan.
Honestly I think all could go in, but we should pick a default -- whether it's one of these or a different one. Anyone have a strong opinion?
-Matt
On Mon, Jan 18, 2016 at 10:19 AM, B.W. Keller kellerbw@mcmaster.ca wrote:
Wow, all of these look great. I think I like Matt's best for painting our bikeshed, but I would be happy with any of them.
On Mon, Jan 18, 2016 at 11:10 AM, Kacper Kowalik < xarthisius.kk@gmail.com> wrote:
On 01/18/2016 09:45 AM, Matthew Turk wrote:
Hi all,
I've experimented a bit and come up with this:
https://images.hub.yt/u/fido/m/9bbe3cf6-png/
The script:
http://paste.yt-project.org/show/6151/
This was designed with the viscm project, which is awfully cool.
What
do folks think? I think Kacper and Nathan also experimented with viscm and have some ideas too, so maybe we should put it up for an eventual vote.
This is my experiment:
https://images.hub.yt/u/fido/m/f180a901-png/
Source:
http://paste.yt-project.org/show/6166/
Cheers, Kacper
Also, I would campaign for calling whatever our new colormap turns
out
to be one of these three things, in increasing order of my
preference:
agar kelp kanten
-Matt
On Thu, Jan 14, 2016 at 10:57 AM, Matthew Turk <
matthewturk@gmail.com> wrote:
> Hi Stuart and everyone else, > > This is great info. I appreciate everyone's thoughtful replies. > > Having both a sequential colormap (which would replace algae) and a > diverging colormap, would be awesome. The Paraview devs shipped the > new matplotlib ones (like Inferno) in 5.0. I think it would be a
fun
> experiment to see if we can come up with something sufficiently > "branded" or different. And then if we can't, fall back on
something
> like Inferno? > > -Matt > > On Wed, Jan 6, 2016 at 12:48 PM, Levy, Stuart A <
salevy@illinois.edu> wrote:
>> There was a fair bit of discussion about colormaps - terrible,
useful,
>> beautiful - at IEEE Vis last October. The viridis colormap was a
featured
>> one. So was the traditional rainbow, which lots of info-vis and
perceptual
>> people piled on to criticize. >> >> Among design criteria for a continuous-valued colormap is whether
it's
>> "sequential" (like the typical yt colormap, or viridis) or
"diverging".
>> You'd want a diverging colormap to show signed deviations from a
norm -
>> where the eye should be caught by places where a value is either
much less
>> than, or much more than, something in the middle. Is it worth
offering a
>> typical divergent colormap, as well as a new typical sequential
one, in yt?
>> >> Note that among the Stefan van der Walt & Nathaniel Smith writeup ( >> http://bids.github.io/colormap/ ) on their development of better
cmaps, they
>> use Nathan Goldbaum's galaxy evolution as a test case for six
(sequential)
>> examples! =>
http://vorpus.org/%7Enjs/goldbaum-galaxies-all-colormaps.mkv
>> >> A neat web site with sample colormaps - aimed at mapping discrete
values on
>> geographic maps, so not directly applicable but cool - is this, by
Cynthia
>> Brewer and Mark Harrower at PSU: >> http://colorbrewer2.org/ >> It has a library of predesigned cmaps, and lets you sift them by
being
>> colorblind-safe, photocopy safe, etc. >> ________________________________ >> From: yt-dev [yt-dev-bounces@lists.spacepope.org] on behalf of
B.W. Keller
>> [kellerbw@mcmaster.ca] >> Sent: Wednesday, January 06, 2016 12:13 >> To: yt-dev@lists.spacepope.org >> Subject: Re: [yt-dev] Default colormap >> >> There is a really excellent paper on designing color maps called
"Color
>> Sequences for Univariate Maps: Theory, Experiments, and
Principles" that you
>> can get here: >>
http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq...
>> >> If we design a new colormap, this would be a good reference along
with those
>> scipy resources. I personally would love to have an accessible,
yt-custom
>> colormap. >> >> On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter <
schnetter@gmail.com> wrote:
>>> >>> I think there are several colourmaps that were created when
Viridis
>>> was invented. I personally like Inferno. >>> >>> -erik >>> >>> On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum <
nathan12343@gmail.com>
>>> wrote: >>>> I would also be for coming up with our own colormap. That said,
I think
>>>> simply modifying algae won't be enough, since it is too
perceptually
>>>> nonlinear. >>>> >>>> On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com
wrote:
>>>>> >>>>> I would go for modifying algae. >>>>> >>>>>> On Jan 6, 2016, at 11:30 AM, Matthew Turk <
matthewturk@gmail.com>
>>>>>> wrote: >>>>>> >>>>>> Hi folks, >>>>>> >>>>>> For a long time we've used "algae," which was designed by
Britton
>>>>>> about eight years ago, as the default colormap. This has been
really
>>>>>> nice for "branding" yt -- if you see an algae plot, it's
probably
>>>>>> (not >>>>>> definitely) made with yt. But it's also not accessible from a >>>>>> colorblindness perspective. Stefan van der Walt has been
giving some
>>>>>> really great talks lately about building a better colormap for >>>>>> matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU
) which
>>>>>> culminated in viridis, which is shipping in recent versions of >>>>>> matplotlib and will become the default. >>>>>> >>>>>> In support of this, he built a tool called viscm which can
generate
>>>>>> reduced versions of colormaps to show what they would be like
with
>>>>>> varying degrees of insensitivity to color. I've generated
outputs
>>>>>> from viscm of three of the custom colormaps we ship with yt: >>>>>> >>>>>> Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ >>>>>> Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I
believe
>>>>>> this is now also shipped with MPL) >>>>>> Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/ >>>>>> >>>>>> I love algae, but it's not the best from an accessibility >>>>>> perspective. >>>>>> >>>>>> I'd like to propose that we use a new default colormap. If we
do
>>>>>> this, I see two options: >>>>>> >>>>>> * Retain a "branding" by developing a new one either by using
the
>>>>>> techniques used by matplotlib (or one of the maps they opted
not to
>>>>>> use) or by modifying algae to be more accessible; looking at
the
>>>>>> response functions, I suspect it would be reasonably possible
to
>>>>>> modify it. (Modifying algae is my preference.) >>>>>> * Use viridis (which we may then have to ship if we have older >>>>>> versions of matplotlib to support) >>>>>> >>>>>> -Matt >>>>>> _______________________________________________ >>>>>> yt-dev mailing list >>>>>> yt-dev@lists.spacepope.org >>>>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>>>> >>>>> _______________________________________________ >>>>> yt-dev mailing list >>>>> yt-dev@lists.spacepope.org >>>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>>> >>>> >>>> >>>> _______________________________________________ >>>> yt-dev mailing list >>>> yt-dev@lists.spacepope.org >>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>>> >>> >>> >>> >>> -- >>> Erik Schnetter schnetter@gmail.com >>> http://www.perimeterinstitute.ca/personal/eschnetter/ >>> _______________________________________________ >>> yt-dev mailing list >>> yt-dev@lists.spacepope.org >>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >> >> >> >> _______________________________________________ >> yt-dev mailing list >> yt-dev@lists.spacepope.org >> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >> _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
-- Michael Zingale Associate Professor
Dept. of Physics & Astronomy • Stony Brook University • Stony Brook, NY 11794-3800 *phone*: 631-632-8225 *e-mail*: Michael.Zingale@stonybrook.edu *web*: http://www.astro.sunysb.edu/mzingale github: http://github.com/zingale
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
+1 for right
-Kenza
On Thu, Jan 21, 2016 at 12:00 PM Matthew Turk matthewturk@gmail.com wrote:
Hi Mike,
Ah, awesome. I will watch that. I do know that Ralf Kaehler designed one for the AMNH's show that was inverted from what we usually think. Sam might have reverse engineered it at some point... it'd be pretty cool to explore that type of thing as well!
-Matt
PS re: the middle colormap, your check is in the mail.
On Thu, Jan 21, 2016 at 12:13 PM, Michael Zingale < michael.zingale@stonybrook.edu> wrote:
here's a link:
http://eric.ed.gov/?id=EJ1005874
really cool stuff.
On Thu, Jan 21, 2016 at 1:10 PM, Nathan Goldbaum nathan12343@gmail.com wrote:
I think we should probably put it up for a vote and we should send an e-mail to yt-users about it.
On Thu, Jan 21, 2016 at 12:06 PM, Matthew Turk matthewturk@gmail.com wrote:
Hi folks,
I've put up a comparison image:
http://i.imgur.com/Afxdb0G.jpg
Left is Kacper, middle is me, right is Nathan.
Honestly I think all could go in, but we should pick a default -- whether it's one of these or a different one. Anyone have a strong opinion?
-Matt
On Mon, Jan 18, 2016 at 10:19 AM, B.W. Keller kellerbw@mcmaster.ca wrote:
Wow, all of these look great. I think I like Matt's best for painting our bikeshed, but I would be happy with any of them.
On Mon, Jan 18, 2016 at 11:10 AM, Kacper Kowalik < xarthisius.kk@gmail.com> wrote:
On 01/18/2016 09:45 AM, Matthew Turk wrote: > Hi all, > > I've experimented a bit and come up with this: > > https://images.hub.yt/u/fido/m/9bbe3cf6-png/ > > The script: > > http://paste.yt-project.org/show/6151/ > > This was designed with the viscm project, which is awfully cool. What > do folks think? I think Kacper and Nathan also experimented with > viscm and have some ideas too, so maybe we should put it up for an > eventual vote.
This is my experiment:
https://images.hub.yt/u/fido/m/f180a901-png/
Source:
http://paste.yt-project.org/show/6166/
Cheers, Kacper
> > Also, I would campaign for calling whatever our new colormap turns out > to be one of these three things, in increasing order of my preference: > > agar > kelp > kanten > > -Matt > > On Thu, Jan 14, 2016 at 10:57 AM, Matthew Turk < matthewturk@gmail.com> wrote: >> Hi Stuart and everyone else, >> >> This is great info. I appreciate everyone's thoughtful replies. >> >> Having both a sequential colormap (which would replace algae) and a >> diverging colormap, would be awesome. The Paraview devs shipped the >> new matplotlib ones (like Inferno) in 5.0. I think it would be a fun >> experiment to see if we can come up with something sufficiently >> "branded" or different. And then if we can't, fall back on something >> like Inferno? >> >> -Matt >> >> On Wed, Jan 6, 2016 at 12:48 PM, Levy, Stuart A < salevy@illinois.edu> wrote: >>> There was a fair bit of discussion about colormaps - terrible, useful, >>> beautiful - at IEEE Vis last October. The viridis colormap was a featured >>> one. So was the traditional rainbow, which lots of info-vis and perceptual >>> people piled on to criticize. >>> >>> Among design criteria for a continuous-valued colormap is whether it's >>> "sequential" (like the typical yt colormap, or viridis) or "diverging". >>> You'd want a diverging colormap to show signed deviations from a norm - >>> where the eye should be caught by places where a value is either much less >>> than, or much more than, something in the middle. Is it worth offering a >>> typical divergent colormap, as well as a new typical sequential one, in yt? >>> >>> Note that among the Stefan van der Walt & Nathaniel Smith writeup ( >>> http://bids.github.io/colormap/ ) on their development of better cmaps, they >>> use Nathan Goldbaum's galaxy evolution as a test case for six (sequential) >>> examples! => http://vorpus.org/%7Enjs/goldbaum-galaxies-all-colormaps.mkv >>> >>> A neat web site with sample colormaps - aimed at mapping discrete values on >>> geographic maps, so not directly applicable but cool - is this, by Cynthia >>> Brewer and Mark Harrower at PSU: >>> http://colorbrewer2.org/ >>> It has a library of predesigned cmaps, and lets you sift them by being >>> colorblind-safe, photocopy safe, etc. >>> ________________________________ >>> From: yt-dev [yt-dev-bounces@lists.spacepope.org] on behalf of B.W. Keller >>> [kellerbw@mcmaster.ca] >>> Sent: Wednesday, January 06, 2016 12:13 >>> To: yt-dev@lists.spacepope.org >>> Subject: Re: [yt-dev] Default colormap >>> >>> There is a really excellent paper on designing color maps called "Color >>> Sequences for Univariate Maps: Theory, Experiments, and Principles" that you >>> can get here: >>> http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq... >>> >>> If we design a new colormap, this would be a good reference along with those >>> scipy resources. I personally would love to have an accessible, yt-custom >>> colormap. >>> >>> On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter < schnetter@gmail.com> wrote: >>>> >>>> I think there are several colourmaps that were created when Viridis >>>> was invented. I personally like Inferno. >>>> >>>> -erik >>>> >>>> On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum < nathan12343@gmail.com> >>>> wrote: >>>>> I would also be for coming up with our own colormap. That said, I think >>>>> simply modifying algae won't be enough, since it is too perceptually >>>>> nonlinear. >>>>> >>>>> On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com wrote: >>>>>> >>>>>> I would go for modifying algae. >>>>>> >>>>>>> On Jan 6, 2016, at 11:30 AM, Matthew Turk < matthewturk@gmail.com> >>>>>>> wrote: >>>>>>> >>>>>>> Hi folks, >>>>>>> >>>>>>> For a long time we've used "algae," which was designed by Britton >>>>>>> about eight years ago, as the default colormap. This has been really >>>>>>> nice for "branding" yt -- if you see an algae plot, it's probably >>>>>>> (not >>>>>>> definitely) made with yt. But it's also not accessible from a >>>>>>> colorblindness perspective. Stefan van der Walt has been giving some >>>>>>> really great talks lately about building a better colormap for >>>>>>> matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU ) which >>>>>>> culminated in viridis, which is shipping in recent versions of >>>>>>> matplotlib and will become the default. >>>>>>> >>>>>>> In support of this, he built a tool called viscm which can generate >>>>>>> reduced versions of colormaps to show what they would be like with >>>>>>> varying degrees of insensitivity to color. I've generated outputs >>>>>>> from viscm of three of the custom colormaps we ship with yt: >>>>>>> >>>>>>> Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ >>>>>>> Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I believe >>>>>>> this is now also shipped with MPL) >>>>>>> Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/ >>>>>>> >>>>>>> I love algae, but it's not the best from an accessibility >>>>>>> perspective. >>>>>>> >>>>>>> I'd like to propose that we use a new default colormap. If we do >>>>>>> this, I see two options: >>>>>>> >>>>>>> * Retain a "branding" by developing a new one either by using the >>>>>>> techniques used by matplotlib (or one of the maps they opted not to >>>>>>> use) or by modifying algae to be more accessible; looking at the >>>>>>> response functions, I suspect it would be reasonably possible to >>>>>>> modify it. (Modifying algae is my preference.) >>>>>>> * Use viridis (which we may then have to ship if we have older >>>>>>> versions of matplotlib to support) >>>>>>> >>>>>>> -Matt >>>>>>> _______________________________________________ >>>>>>> yt-dev mailing list >>>>>>> yt-dev@lists.spacepope.org >>>>>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>>>>> >>>>>> _______________________________________________ >>>>>> yt-dev mailing list >>>>>> yt-dev@lists.spacepope.org >>>>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>>>> >>>>> >>>>> >>>>> _______________________________________________ >>>>> yt-dev mailing list >>>>> yt-dev@lists.spacepope.org >>>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>>>> >>>> >>>> >>>> >>>> -- >>>> Erik Schnetter schnetter@gmail.com >>>> http://www.perimeterinstitute.ca/personal/eschnetter/ >>>> _______________________________________________ >>>> yt-dev mailing list >>>> yt-dev@lists.spacepope.org >>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>> >>> >>> >>> _______________________________________________ >>> yt-dev mailing list >>> yt-dev@lists.spacepope.org >>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>> > _______________________________________________ > yt-dev mailing list > yt-dev@lists.spacepope.org > http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
-- Michael Zingale Associate Professor
Dept. of Physics & Astronomy • Stony Brook University • Stony Brook, NY 11794-3800 *phone*: 631-632-8225 *e-mail*: Michael.Zingale@stonybrook.edu *web*: http://www.astro.sunysb.edu/mzingale github: http://github.com/zingale
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
Sounds good, but before we do that, I think we should have a suitable waiting period for others to propose them. I'll send an email out to yt-users asking for ideas on it.
On Thu, Jan 21, 2016 at 12:10 PM, Nathan Goldbaum nathan12343@gmail.com wrote:
I think we should probably put it up for a vote and we should send an e-mail to yt-users about it.
On Thu, Jan 21, 2016 at 12:06 PM, Matthew Turk matthewturk@gmail.com wrote:
Hi folks,
I've put up a comparison image:
http://i.imgur.com/Afxdb0G.jpg
Left is Kacper, middle is me, right is Nathan.
Honestly I think all could go in, but we should pick a default -- whether it's one of these or a different one. Anyone have a strong opinion?
-Matt
On Mon, Jan 18, 2016 at 10:19 AM, B.W. Keller kellerbw@mcmaster.ca wrote:
Wow, all of these look great. I think I like Matt's best for painting our bikeshed, but I would be happy with any of them.
On Mon, Jan 18, 2016 at 11:10 AM, Kacper Kowalik < xarthisius.kk@gmail.com> wrote:
On 01/18/2016 09:45 AM, Matthew Turk wrote:
Hi all,
I've experimented a bit and come up with this:
https://images.hub.yt/u/fido/m/9bbe3cf6-png/
The script:
http://paste.yt-project.org/show/6151/
This was designed with the viscm project, which is awfully cool. What do folks think? I think Kacper and Nathan also experimented with viscm and have some ideas too, so maybe we should put it up for an eventual vote.
This is my experiment:
https://images.hub.yt/u/fido/m/f180a901-png/
Source:
http://paste.yt-project.org/show/6166/
Cheers, Kacper
Also, I would campaign for calling whatever our new colormap turns out to be one of these three things, in increasing order of my preference:
agar kelp kanten
-Matt
On Thu, Jan 14, 2016 at 10:57 AM, Matthew Turk matthewturk@gmail.com
wrote:
Hi Stuart and everyone else,
This is great info. I appreciate everyone's thoughtful replies.
Having both a sequential colormap (which would replace algae) and a diverging colormap, would be awesome. The Paraview devs shipped the new matplotlib ones (like Inferno) in 5.0. I think it would be a fun experiment to see if we can come up with something sufficiently "branded" or different. And then if we can't, fall back on something like Inferno?
-Matt
On Wed, Jan 6, 2016 at 12:48 PM, Levy, Stuart A salevy@illinois.edu
wrote:
> There was a fair bit of discussion about colormaps - terrible,
useful,
> beautiful - at IEEE Vis last October. The viridis colormap was a
featured
> one. So was the traditional rainbow, which lots of info-vis and
perceptual
> people piled on to criticize. > > Among design criteria for a continuous-valued colormap is whether
it's
> "sequential" (like the typical yt colormap, or viridis) or
"diverging".
> You'd want a diverging colormap to show signed deviations from a
norm -
> where the eye should be caught by places where a value is either
much less
> than, or much more than, something in the middle. Is it worth
offering a
> typical divergent colormap, as well as a new typical sequential
one, in yt?
> > Note that among the Stefan van der Walt & Nathaniel Smith writeup ( > http://bids.github.io/colormap/ ) on their development of better
cmaps, they
> use Nathan Goldbaum's galaxy evolution as a test case for six
(sequential)
> examples! =>
http://vorpus.org/%7Enjs/goldbaum-galaxies-all-colormaps.mkv
> > A neat web site with sample colormaps - aimed at mapping discrete
values on
> geographic maps, so not directly applicable but cool - is this, by
Cynthia
> Brewer and Mark Harrower at PSU: > http://colorbrewer2.org/ > It has a library of predesigned cmaps, and lets you sift them by
being
> colorblind-safe, photocopy safe, etc. > ________________________________ > From: yt-dev [yt-dev-bounces@lists.spacepope.org] on behalf of
B.W. Keller
> [kellerbw@mcmaster.ca] > Sent: Wednesday, January 06, 2016 12:13 > To: yt-dev@lists.spacepope.org > Subject: Re: [yt-dev] Default colormap > > There is a really excellent paper on designing color maps called
"Color
> Sequences for Univariate Maps: Theory, Experiments, and Principles"
that you
> can get here: >
http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq...
> > If we design a new colormap, this would be a good reference along
with those
> scipy resources. I personally would love to have an accessible,
yt-custom
> colormap. > > On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter <
schnetter@gmail.com> wrote:
>> >> I think there are several colourmaps that were created when Viridis >> was invented. I personally like Inferno. >> >> -erik >> >> On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum <
nathan12343@gmail.com>
>> wrote: >>> I would also be for coming up with our own colormap. That said, I
think
>>> simply modifying algae won't be enough, since it is too
perceptually
>>> nonlinear. >>> >>> On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com
wrote:
>>>> >>>> I would go for modifying algae. >>>> >>>>> On Jan 6, 2016, at 11:30 AM, Matthew Turk <
matthewturk@gmail.com>
>>>>> wrote: >>>>> >>>>> Hi folks, >>>>> >>>>> For a long time we've used "algae," which was designed by
Britton
>>>>> about eight years ago, as the default colormap. This has been
really
>>>>> nice for "branding" yt -- if you see an algae plot, it's
probably
>>>>> (not >>>>> definitely) made with yt. But it's also not accessible from a >>>>> colorblindness perspective. Stefan van der Walt has been
giving some
>>>>> really great talks lately about building a better colormap for >>>>> matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU
) which
>>>>> culminated in viridis, which is shipping in recent versions of >>>>> matplotlib and will become the default. >>>>> >>>>> In support of this, he built a tool called viscm which can
generate
>>>>> reduced versions of colormaps to show what they would be like
with
>>>>> varying degrees of insensitivity to color. I've generated
outputs
>>>>> from viscm of three of the custom colormaps we ship with yt: >>>>> >>>>> Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ >>>>> Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I
believe
>>>>> this is now also shipped with MPL) >>>>> Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/ >>>>> >>>>> I love algae, but it's not the best from an accessibility >>>>> perspective. >>>>> >>>>> I'd like to propose that we use a new default colormap. If we
do
>>>>> this, I see two options: >>>>> >>>>> * Retain a "branding" by developing a new one either by using
the
>>>>> techniques used by matplotlib (or one of the maps they opted
not to
>>>>> use) or by modifying algae to be more accessible; looking at the >>>>> response functions, I suspect it would be reasonably possible to >>>>> modify it. (Modifying algae is my preference.) >>>>> * Use viridis (which we may then have to ship if we have older >>>>> versions of matplotlib to support) >>>>> >>>>> -Matt >>>>> _______________________________________________ >>>>> yt-dev mailing list >>>>> yt-dev@lists.spacepope.org >>>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>>> >>>> _______________________________________________ >>>> yt-dev mailing list >>>> yt-dev@lists.spacepope.org >>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>> >>> >>> >>> _______________________________________________ >>> yt-dev mailing list >>> yt-dev@lists.spacepope.org >>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>> >> >> >> >> -- >> Erik Schnetter schnetter@gmail.com >> http://www.perimeterinstitute.ca/personal/eschnetter/ >> _______________________________________________ >> yt-dev mailing list >> yt-dev@lists.spacepope.org >> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org > > > > _______________________________________________ > yt-dev mailing list > yt-dev@lists.spacepope.org > http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
I'd like to propose this colormap:
http://i.imgur.com/nKFZWSC.png
Script here:
http://paste.yt-project.org/show/6194/
Cameron
On Thu, Jan 21, 2016 at 11:05 AM, Matthew Turk matthewturk@gmail.com wrote:
Sounds good, but before we do that, I think we should have a suitable waiting period for others to propose them. I'll send an email out to yt-users asking for ideas on it.
On Thu, Jan 21, 2016 at 12:10 PM, Nathan Goldbaum nathan12343@gmail.com wrote:
I think we should probably put it up for a vote and we should send an e-mail to yt-users about it.
On Thu, Jan 21, 2016 at 12:06 PM, Matthew Turk matthewturk@gmail.com wrote:
Hi folks,
I've put up a comparison image:
http://i.imgur.com/Afxdb0G.jpg
Left is Kacper, middle is me, right is Nathan.
Honestly I think all could go in, but we should pick a default -- whether it's one of these or a different one. Anyone have a strong opinion?
-Matt
On Mon, Jan 18, 2016 at 10:19 AM, B.W. Keller kellerbw@mcmaster.ca wrote:
Wow, all of these look great. I think I like Matt's best for painting our bikeshed, but I would be happy with any of them.
On Mon, Jan 18, 2016 at 11:10 AM, Kacper Kowalik < xarthisius.kk@gmail.com> wrote:
On 01/18/2016 09:45 AM, Matthew Turk wrote:
Hi all,
I've experimented a bit and come up with this:
https://images.hub.yt/u/fido/m/9bbe3cf6-png/
The script:
http://paste.yt-project.org/show/6151/
This was designed with the viscm project, which is awfully cool.
What
do folks think? I think Kacper and Nathan also experimented with viscm and have some ideas too, so maybe we should put it up for an eventual vote.
This is my experiment:
https://images.hub.yt/u/fido/m/f180a901-png/
Source:
http://paste.yt-project.org/show/6166/
Cheers, Kacper
Also, I would campaign for calling whatever our new colormap turns
out
to be one of these three things, in increasing order of my
preference:
agar kelp kanten
-Matt
On Thu, Jan 14, 2016 at 10:57 AM, Matthew Turk <
matthewturk@gmail.com> wrote:
> Hi Stuart and everyone else, > > This is great info. I appreciate everyone's thoughtful replies. > > Having both a sequential colormap (which would replace algae) and a > diverging colormap, would be awesome. The Paraview devs shipped the > new matplotlib ones (like Inferno) in 5.0. I think it would be a
fun
> experiment to see if we can come up with something sufficiently > "branded" or different. And then if we can't, fall back on
something
> like Inferno? > > -Matt > > On Wed, Jan 6, 2016 at 12:48 PM, Levy, Stuart A <
salevy@illinois.edu> wrote:
>> There was a fair bit of discussion about colormaps - terrible,
useful,
>> beautiful - at IEEE Vis last October. The viridis colormap was a
featured
>> one. So was the traditional rainbow, which lots of info-vis and
perceptual
>> people piled on to criticize. >> >> Among design criteria for a continuous-valued colormap is whether
it's
>> "sequential" (like the typical yt colormap, or viridis) or
"diverging".
>> You'd want a diverging colormap to show signed deviations from a
norm -
>> where the eye should be caught by places where a value is either
much less
>> than, or much more than, something in the middle. Is it worth
offering a
>> typical divergent colormap, as well as a new typical sequential
one, in yt?
>> >> Note that among the Stefan van der Walt & Nathaniel Smith writeup ( >> http://bids.github.io/colormap/ ) on their development of better
cmaps, they
>> use Nathan Goldbaum's galaxy evolution as a test case for six
(sequential)
>> examples! =>
http://vorpus.org/%7Enjs/goldbaum-galaxies-all-colormaps.mkv
>> >> A neat web site with sample colormaps - aimed at mapping discrete
values on
>> geographic maps, so not directly applicable but cool - is this, by
Cynthia
>> Brewer and Mark Harrower at PSU: >> http://colorbrewer2.org/ >> It has a library of predesigned cmaps, and lets you sift them by
being
>> colorblind-safe, photocopy safe, etc. >> ________________________________ >> From: yt-dev [yt-dev-bounces@lists.spacepope.org] on behalf of
B.W. Keller
>> [kellerbw@mcmaster.ca] >> Sent: Wednesday, January 06, 2016 12:13 >> To: yt-dev@lists.spacepope.org >> Subject: Re: [yt-dev] Default colormap >> >> There is a really excellent paper on designing color maps called
"Color
>> Sequences for Univariate Maps: Theory, Experiments, and
Principles" that you
>> can get here: >>
http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq...
>> >> If we design a new colormap, this would be a good reference along
with those
>> scipy resources. I personally would love to have an accessible,
yt-custom
>> colormap. >> >> On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter <
schnetter@gmail.com> wrote:
>>> >>> I think there are several colourmaps that were created when
Viridis
>>> was invented. I personally like Inferno. >>> >>> -erik >>> >>> On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum <
nathan12343@gmail.com>
>>> wrote: >>>> I would also be for coming up with our own colormap. That said,
I think
>>>> simply modifying algae won't be enough, since it is too
perceptually
>>>> nonlinear. >>>> >>>> On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com
wrote:
>>>>> >>>>> I would go for modifying algae. >>>>> >>>>>> On Jan 6, 2016, at 11:30 AM, Matthew Turk <
matthewturk@gmail.com>
>>>>>> wrote: >>>>>> >>>>>> Hi folks, >>>>>> >>>>>> For a long time we've used "algae," which was designed by
Britton
>>>>>> about eight years ago, as the default colormap. This has been
really
>>>>>> nice for "branding" yt -- if you see an algae plot, it's
probably
>>>>>> (not >>>>>> definitely) made with yt. But it's also not accessible from a >>>>>> colorblindness perspective. Stefan van der Walt has been
giving some
>>>>>> really great talks lately about building a better colormap for >>>>>> matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU
) which
>>>>>> culminated in viridis, which is shipping in recent versions of >>>>>> matplotlib and will become the default. >>>>>> >>>>>> In support of this, he built a tool called viscm which can
generate
>>>>>> reduced versions of colormaps to show what they would be like
with
>>>>>> varying degrees of insensitivity to color. I've generated
outputs
>>>>>> from viscm of three of the custom colormaps we ship with yt: >>>>>> >>>>>> Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ >>>>>> Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I
believe
>>>>>> this is now also shipped with MPL) >>>>>> Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/ >>>>>> >>>>>> I love algae, but it's not the best from an accessibility >>>>>> perspective. >>>>>> >>>>>> I'd like to propose that we use a new default colormap. If we
do
>>>>>> this, I see two options: >>>>>> >>>>>> * Retain a "branding" by developing a new one either by using
the
>>>>>> techniques used by matplotlib (or one of the maps they opted
not to
>>>>>> use) or by modifying algae to be more accessible; looking at
the
>>>>>> response functions, I suspect it would be reasonably possible
to
>>>>>> modify it. (Modifying algae is my preference.) >>>>>> * Use viridis (which we may then have to ship if we have older >>>>>> versions of matplotlib to support) >>>>>> >>>>>> -Matt >>>>>> _______________________________________________ >>>>>> yt-dev mailing list >>>>>> yt-dev@lists.spacepope.org >>>>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>>>> >>>>> _______________________________________________ >>>>> yt-dev mailing list >>>>> yt-dev@lists.spacepope.org >>>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>>> >>>> >>>> >>>> _______________________________________________ >>>> yt-dev mailing list >>>> yt-dev@lists.spacepope.org >>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>>> >>> >>> >>> >>> -- >>> Erik Schnetter schnetter@gmail.com >>> http://www.perimeterinstitute.ca/personal/eschnetter/ >>> _______________________________________________ >>> yt-dev mailing list >>> yt-dev@lists.spacepope.org >>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >> >> >> >> _______________________________________________ >> yt-dev mailing list >> yt-dev@lists.spacepope.org >> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >> _______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
Thanks, Cameron.
Unless we get any other submissions, I've got a Google Form ready to go to ask for preferences on this that I'll send out tomorrow morning. It includes the montages and color space diagrams; they can all be viewed here: https://drive.google.com/open?id=0B1DaRHnSGW8Za0U4cEFTTzVPcVE
-Matt
On Wed, Jan 27, 2016 at 10:22 PM, Cameron Hummels chummels@gmail.com wrote:
I'd like to propose this colormap:
http://i.imgur.com/nKFZWSC.png
Script here:
http://paste.yt-project.org/show/6194/
Cameron
On Thu, Jan 21, 2016 at 11:05 AM, Matthew Turk matthewturk@gmail.com wrote:
Sounds good, but before we do that, I think we should have a suitable waiting period for others to propose them. I'll send an email out to yt-users asking for ideas on it.
On Thu, Jan 21, 2016 at 12:10 PM, Nathan Goldbaum nathan12343@gmail.com wrote:
I think we should probably put it up for a vote and we should send an e-mail to yt-users about it.
On Thu, Jan 21, 2016 at 12:06 PM, Matthew Turk matthewturk@gmail.com wrote:
Hi folks,
I've put up a comparison image:
http://i.imgur.com/Afxdb0G.jpg
Left is Kacper, middle is me, right is Nathan.
Honestly I think all could go in, but we should pick a default -- whether it's one of these or a different one. Anyone have a strong opinion?
-Matt
On Mon, Jan 18, 2016 at 10:19 AM, B.W. Keller kellerbw@mcmaster.ca wrote:
Wow, all of these look great. I think I like Matt's best for painting our bikeshed, but I would be happy with any of them.
On Mon, Jan 18, 2016 at 11:10 AM, Kacper Kowalik < xarthisius.kk@gmail.com> wrote:
On 01/18/2016 09:45 AM, Matthew Turk wrote: > Hi all, > > I've experimented a bit and come up with this: > > https://images.hub.yt/u/fido/m/9bbe3cf6-png/ > > The script: > > http://paste.yt-project.org/show/6151/ > > This was designed with the viscm project, which is awfully cool. What > do folks think? I think Kacper and Nathan also experimented with > viscm and have some ideas too, so maybe we should put it up for an > eventual vote.
This is my experiment:
https://images.hub.yt/u/fido/m/f180a901-png/
Source:
http://paste.yt-project.org/show/6166/
Cheers, Kacper
> > Also, I would campaign for calling whatever our new colormap turns out > to be one of these three things, in increasing order of my preference: > > agar > kelp > kanten > > -Matt > > On Thu, Jan 14, 2016 at 10:57 AM, Matthew Turk < matthewturk@gmail.com> wrote: >> Hi Stuart and everyone else, >> >> This is great info. I appreciate everyone's thoughtful replies. >> >> Having both a sequential colormap (which would replace algae) and a >> diverging colormap, would be awesome. The Paraview devs shipped the >> new matplotlib ones (like Inferno) in 5.0. I think it would be a fun >> experiment to see if we can come up with something sufficiently >> "branded" or different. And then if we can't, fall back on something >> like Inferno? >> >> -Matt >> >> On Wed, Jan 6, 2016 at 12:48 PM, Levy, Stuart A < salevy@illinois.edu> wrote: >>> There was a fair bit of discussion about colormaps - terrible, useful, >>> beautiful - at IEEE Vis last October. The viridis colormap was a featured >>> one. So was the traditional rainbow, which lots of info-vis and perceptual >>> people piled on to criticize. >>> >>> Among design criteria for a continuous-valued colormap is whether it's >>> "sequential" (like the typical yt colormap, or viridis) or "diverging". >>> You'd want a diverging colormap to show signed deviations from a norm - >>> where the eye should be caught by places where a value is either much less >>> than, or much more than, something in the middle. Is it worth offering a >>> typical divergent colormap, as well as a new typical sequential one, in yt? >>> >>> Note that among the Stefan van der Walt & Nathaniel Smith writeup ( >>> http://bids.github.io/colormap/ ) on their development of better cmaps, they >>> use Nathan Goldbaum's galaxy evolution as a test case for six (sequential) >>> examples! => http://vorpus.org/%7Enjs/goldbaum-galaxies-all-colormaps.mkv >>> >>> A neat web site with sample colormaps - aimed at mapping discrete values on >>> geographic maps, so not directly applicable but cool - is this, by Cynthia >>> Brewer and Mark Harrower at PSU: >>> http://colorbrewer2.org/ >>> It has a library of predesigned cmaps, and lets you sift them by being >>> colorblind-safe, photocopy safe, etc. >>> ________________________________ >>> From: yt-dev [yt-dev-bounces@lists.spacepope.org] on behalf of B.W. Keller >>> [kellerbw@mcmaster.ca] >>> Sent: Wednesday, January 06, 2016 12:13 >>> To: yt-dev@lists.spacepope.org >>> Subject: Re: [yt-dev] Default colormap >>> >>> There is a really excellent paper on designing color maps called "Color >>> Sequences for Univariate Maps: Theory, Experiments, and Principles" that you >>> can get here: >>> http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq... >>> >>> If we design a new colormap, this would be a good reference along with those >>> scipy resources. I personally would love to have an accessible, yt-custom >>> colormap. >>> >>> On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter < schnetter@gmail.com> wrote: >>>> >>>> I think there are several colourmaps that were created when Viridis >>>> was invented. I personally like Inferno. >>>> >>>> -erik >>>> >>>> On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum < nathan12343@gmail.com> >>>> wrote: >>>>> I would also be for coming up with our own colormap. That said, I think >>>>> simply modifying algae won't be enough, since it is too perceptually >>>>> nonlinear. >>>>> >>>>> On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com wrote: >>>>>> >>>>>> I would go for modifying algae. >>>>>> >>>>>>> On Jan 6, 2016, at 11:30 AM, Matthew Turk < matthewturk@gmail.com> >>>>>>> wrote: >>>>>>> >>>>>>> Hi folks, >>>>>>> >>>>>>> For a long time we've used "algae," which was designed by Britton >>>>>>> about eight years ago, as the default colormap. This has been really >>>>>>> nice for "branding" yt -- if you see an algae plot, it's probably >>>>>>> (not >>>>>>> definitely) made with yt. But it's also not accessible from a >>>>>>> colorblindness perspective. Stefan van der Walt has been giving some >>>>>>> really great talks lately about building a better colormap for >>>>>>> matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU ) which >>>>>>> culminated in viridis, which is shipping in recent versions of >>>>>>> matplotlib and will become the default. >>>>>>> >>>>>>> In support of this, he built a tool called viscm which can generate >>>>>>> reduced versions of colormaps to show what they would be like with >>>>>>> varying degrees of insensitivity to color. I've generated outputs >>>>>>> from viscm of three of the custom colormaps we ship with yt: >>>>>>> >>>>>>> Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ >>>>>>> Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I believe >>>>>>> this is now also shipped with MPL) >>>>>>> Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/ >>>>>>> >>>>>>> I love algae, but it's not the best from an accessibility >>>>>>> perspective. >>>>>>> >>>>>>> I'd like to propose that we use a new default colormap. If we do >>>>>>> this, I see two options: >>>>>>> >>>>>>> * Retain a "branding" by developing a new one either by using the >>>>>>> techniques used by matplotlib (or one of the maps they opted not to >>>>>>> use) or by modifying algae to be more accessible; looking at the >>>>>>> response functions, I suspect it would be reasonably possible to >>>>>>> modify it. (Modifying algae is my preference.) >>>>>>> * Use viridis (which we may then have to ship if we have older >>>>>>> versions of matplotlib to support) >>>>>>> >>>>>>> -Matt >>>>>>> _______________________________________________ >>>>>>> yt-dev mailing list >>>>>>> yt-dev@lists.spacepope.org >>>>>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>>>>> >>>>>> _______________________________________________ >>>>>> yt-dev mailing list >>>>>> yt-dev@lists.spacepope.org >>>>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>>>> >>>>> >>>>> >>>>> _______________________________________________ >>>>> yt-dev mailing list >>>>> yt-dev@lists.spacepope.org >>>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>>>> >>>> >>>> >>>> >>>> -- >>>> Erik Schnetter schnetter@gmail.com >>>> http://www.perimeterinstitute.ca/personal/eschnetter/ >>>> _______________________________________________ >>>> yt-dev mailing list >>>> yt-dev@lists.spacepope.org >>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>> >>> >>> >>> _______________________________________________ >>> yt-dev mailing list >>> yt-dev@lists.spacepope.org >>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>> > _______________________________________________ > yt-dev mailing list > yt-dev@lists.spacepope.org > http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
-- Cameron Hummels NSF Postdoctoral Fellow Department of Astronomy California Institute of Technology http://chummels.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
I like middle. (and this is not based on who's it is -- but you shouldn't have told us the names)
Btw, this reminds me that I saw a very nice talk at the AAPT meeting by a UCSC graduate (Zoe ... last name escapes me). She did studies on using movies in teaching (and used a lot of Joel Primack's stuff), and worked with a range of students, and one of the conclusions is that students understood simulations of dark matter a lot better when the dark matter was colored dark instead of in bright colors. Seems obvious, in hindsight, but most of the movies you see online use bright colors / whites for high dark matter density. Her talk was very interesting, and it is nice to see someone thinking about how the color choices we make in movies affects how people learn from them.
On Thu, Jan 21, 2016 at 1:06 PM, Matthew Turk matthewturk@gmail.com wrote:
Hi folks,
I've put up a comparison image:
http://i.imgur.com/Afxdb0G.jpg
Left is Kacper, middle is me, right is Nathan.
Honestly I think all could go in, but we should pick a default -- whether it's one of these or a different one. Anyone have a strong opinion?
-Matt
On Mon, Jan 18, 2016 at 10:19 AM, B.W. Keller kellerbw@mcmaster.ca wrote:
Wow, all of these look great. I think I like Matt's best for painting our bikeshed, but I would be happy with any of them.
On Mon, Jan 18, 2016 at 11:10 AM, Kacper Kowalik <xarthisius.kk@gmail.com
wrote:
On 01/18/2016 09:45 AM, Matthew Turk wrote:
Hi all,
I've experimented a bit and come up with this:
https://images.hub.yt/u/fido/m/9bbe3cf6-png/
The script:
http://paste.yt-project.org/show/6151/
This was designed with the viscm project, which is awfully cool. What do folks think? I think Kacper and Nathan also experimented with viscm and have some ideas too, so maybe we should put it up for an eventual vote.
This is my experiment:
https://images.hub.yt/u/fido/m/f180a901-png/
Source:
http://paste.yt-project.org/show/6166/
Cheers, Kacper
Also, I would campaign for calling whatever our new colormap turns out to be one of these three things, in increasing order of my preference:
agar kelp kanten
-Matt
On Thu, Jan 14, 2016 at 10:57 AM, Matthew Turk matthewturk@gmail.com
wrote:
Hi Stuart and everyone else,
This is great info. I appreciate everyone's thoughtful replies.
Having both a sequential colormap (which would replace algae) and a diverging colormap, would be awesome. The Paraview devs shipped the new matplotlib ones (like Inferno) in 5.0. I think it would be a fun experiment to see if we can come up with something sufficiently "branded" or different. And then if we can't, fall back on something like Inferno?
-Matt
On Wed, Jan 6, 2016 at 12:48 PM, Levy, Stuart A salevy@illinois.edu
wrote:
There was a fair bit of discussion about colormaps - terrible,
useful,
beautiful - at IEEE Vis last October. The viridis colormap was a
featured
one. So was the traditional rainbow, which lots of info-vis and
perceptual
people piled on to criticize.
Among design criteria for a continuous-valued colormap is whether
it's
"sequential" (like the typical yt colormap, or viridis) or
"diverging".
You'd want a diverging colormap to show signed deviations from a
norm -
where the eye should be caught by places where a value is either
much less
than, or much more than, something in the middle. Is it worth
offering a
typical divergent colormap, as well as a new typical sequential one,
in yt?
Note that among the Stefan van der Walt & Nathaniel Smith writeup ( http://bids.github.io/colormap/ ) on their development of better
cmaps, they
use Nathan Goldbaum's galaxy evolution as a test case for six
(sequential)
examples! =>
http://vorpus.org/%7Enjs/goldbaum-galaxies-all-colormaps.mkv
A neat web site with sample colormaps - aimed at mapping discrete
values on
geographic maps, so not directly applicable but cool - is this, by
Cynthia
Brewer and Mark Harrower at PSU: http://colorbrewer2.org/ It has a library of predesigned cmaps, and lets you sift them by
being
colorblind-safe, photocopy safe, etc. ________________________________ From: yt-dev [yt-dev-bounces@lists.spacepope.org] on behalf of B.W.
Keller
[kellerbw@mcmaster.ca] Sent: Wednesday, January 06, 2016 12:13 To: yt-dev@lists.spacepope.org Subject: Re: [yt-dev] Default colormap
There is a really excellent paper on designing color maps called
"Color
Sequences for Univariate Maps: Theory, Experiments, and Principles"
that you
can get here:
http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq...
If we design a new colormap, this would be a good reference along
with those
scipy resources. I personally would love to have an accessible,
yt-custom
colormap.
On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter schnetter@gmail.com
wrote:
> > I think there are several colourmaps that were created when Viridis > was invented. I personally like Inferno. > > -erik > > On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum <
nathan12343@gmail.com>
> wrote: >> I would also be for coming up with our own colormap. That said, I
think
>> simply modifying algae won't be enough, since it is too
perceptually
>> nonlinear. >> >> On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com
wrote:
>>> >>> I would go for modifying algae. >>> >>>> On Jan 6, 2016, at 11:30 AM, Matthew Turk <matthewturk@gmail.com
>>>> wrote: >>>> >>>> Hi folks, >>>> >>>> For a long time we've used "algae," which was designed by Britton >>>> about eight years ago, as the default colormap. This has been
really
>>>> nice for "branding" yt -- if you see an algae plot, it's probably >>>> (not >>>> definitely) made with yt. But it's also not accessible from a >>>> colorblindness perspective. Stefan van der Walt has been giving
some
>>>> really great talks lately about building a better colormap for >>>> matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU )
which
>>>> culminated in viridis, which is shipping in recent versions of >>>> matplotlib and will become the default. >>>> >>>> In support of this, he built a tool called viscm which can
generate
>>>> reduced versions of colormaps to show what they would be like
with
>>>> varying degrees of insensitivity to color. I've generated
outputs
>>>> from viscm of three of the custom colormaps we ship with yt: >>>> >>>> Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ >>>> Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I
believe
>>>> this is now also shipped with MPL) >>>> Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/ >>>> >>>> I love algae, but it's not the best from an accessibility >>>> perspective. >>>> >>>> I'd like to propose that we use a new default colormap. If we do >>>> this, I see two options: >>>> >>>> * Retain a "branding" by developing a new one either by using the >>>> techniques used by matplotlib (or one of the maps they opted not
to
>>>> use) or by modifying algae to be more accessible; looking at the >>>> response functions, I suspect it would be reasonably possible to >>>> modify it. (Modifying algae is my preference.) >>>> * Use viridis (which we may then have to ship if we have older >>>> versions of matplotlib to support) >>>> >>>> -Matt >>>> _______________________________________________ >>>> yt-dev mailing list >>>> yt-dev@lists.spacepope.org >>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>> >>> _______________________________________________ >>> yt-dev mailing list >>> yt-dev@lists.spacepope.org >>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >> >> >> >> _______________________________________________ >> yt-dev mailing list >> yt-dev@lists.spacepope.org >> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >> > > > > -- > Erik Schnetter schnetter@gmail.com > http://www.perimeterinstitute.ca/personal/eschnetter/ > _______________________________________________ > yt-dev mailing list > yt-dev@lists.spacepope.org > http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
On 01/21/2016 12:12 PM, Michael Zingale wrote:
I like middle. (and this is not based on who's it is -- but you shouldn't have told us the names)
Btw, this reminds me that I saw a very nice talk at the AAPT meeting by a UCSC graduate (Zoe ... last name escapes me). She did studies on using movies in teaching (and used a lot of Joel Primack's stuff), and worked with a range of students, and one of the conclusions is that students understood simulations of dark matter a lot better when the dark matter was colored dark instead of in bright colors. Seems obvious, in hindsight, but most of the movies you see online use bright colors / whites for high dark matter density. Her talk was very interesting, and it is nice to see someone thinking about how the color choices we make in movies affects how people learn from them.
I think colormaps for general audience should be different by default, e.g those silly astronomers tend to use blue for hot and red for cold, even though every bathroom faucet proves them wrong...
Cheers, Kacper
On Thu, Jan 21, 2016 at 1:06 PM, Matthew Turk matthewturk@gmail.com wrote:
Hi folks,
I've put up a comparison image:
http://i.imgur.com/Afxdb0G.jpg
Left is Kacper, middle is me, right is Nathan.
Honestly I think all could go in, but we should pick a default -- whether it's one of these or a different one. Anyone have a strong opinion?
-Matt
On Mon, Jan 18, 2016 at 10:19 AM, B.W. Keller kellerbw@mcmaster.ca wrote:
Wow, all of these look great. I think I like Matt's best for painting our bikeshed, but I would be happy with any of them.
On Mon, Jan 18, 2016 at 11:10 AM, Kacper Kowalik <xarthisius.kk@gmail.com
wrote:
On 01/18/2016 09:45 AM, Matthew Turk wrote:
Hi all,
I've experimented a bit and come up with this:
https://images.hub.yt/u/fido/m/9bbe3cf6-png/
The script:
http://paste.yt-project.org/show/6151/
This was designed with the viscm project, which is awfully cool. What do folks think? I think Kacper and Nathan also experimented with viscm and have some ideas too, so maybe we should put it up for an eventual vote.
This is my experiment:
https://images.hub.yt/u/fido/m/f180a901-png/
Source:
http://paste.yt-project.org/show/6166/
Cheers, Kacper
Also, I would campaign for calling whatever our new colormap turns out to be one of these three things, in increasing order of my preference:
agar kelp kanten
-Matt
On Thu, Jan 14, 2016 at 10:57 AM, Matthew Turk matthewturk@gmail.com
wrote:
Hi Stuart and everyone else,
This is great info. I appreciate everyone's thoughtful replies.
Having both a sequential colormap (which would replace algae) and a diverging colormap, would be awesome. The Paraview devs shipped the new matplotlib ones (like Inferno) in 5.0. I think it would be a fun experiment to see if we can come up with something sufficiently "branded" or different. And then if we can't, fall back on something like Inferno?
-Matt
On Wed, Jan 6, 2016 at 12:48 PM, Levy, Stuart A salevy@illinois.edu
wrote:
> There was a fair bit of discussion about colormaps - terrible,
useful,
> beautiful - at IEEE Vis last October. The viridis colormap was a
featured
> one. So was the traditional rainbow, which lots of info-vis and
perceptual
> people piled on to criticize. > > Among design criteria for a continuous-valued colormap is whether
it's
> "sequential" (like the typical yt colormap, or viridis) or
"diverging".
> You'd want a diverging colormap to show signed deviations from a
norm -
> where the eye should be caught by places where a value is either
much less
> than, or much more than, something in the middle. Is it worth
offering a
> typical divergent colormap, as well as a new typical sequential one,
in yt?
> > Note that among the Stefan van der Walt & Nathaniel Smith writeup ( > http://bids.github.io/colormap/ ) on their development of better
cmaps, they
> use Nathan Goldbaum's galaxy evolution as a test case for six
(sequential)
> examples! =>
http://vorpus.org/%7Enjs/goldbaum-galaxies-all-colormaps.mkv
> > A neat web site with sample colormaps - aimed at mapping discrete
values on
> geographic maps, so not directly applicable but cool - is this, by
Cynthia
> Brewer and Mark Harrower at PSU: > http://colorbrewer2.org/ > It has a library of predesigned cmaps, and lets you sift them by
being
> colorblind-safe, photocopy safe, etc. > ________________________________ > From: yt-dev [yt-dev-bounces@lists.spacepope.org] on behalf of B.W.
Keller
> [kellerbw@mcmaster.ca] > Sent: Wednesday, January 06, 2016 12:13 > To: yt-dev@lists.spacepope.org > Subject: Re: [yt-dev] Default colormap > > There is a really excellent paper on designing color maps called
"Color
> Sequences for Univariate Maps: Theory, Experiments, and Principles"
that you
> can get here: >
http://ccom.unh.edu/sites/default/files/publications/Ware_1988_CGA_Color_seq...
> > If we design a new colormap, this would be a good reference along
with those
> scipy resources. I personally would love to have an accessible,
yt-custom
> colormap. > > On Wed, Jan 6, 2016 at 11:50 AM, Erik Schnetter schnetter@gmail.com
wrote:
>> >> I think there are several colourmaps that were created when Viridis >> was invented. I personally like Inferno. >> >> -erik >> >> On Wed, Jan 6, 2016 at 11:34 AM, Nathan Goldbaum <
nathan12343@gmail.com>
>> wrote: >>> I would also be for coming up with our own colormap. That said, I
think
>>> simply modifying algae won't be enough, since it is too
perceptually
>>> nonlinear. >>> >>> On Wed, Jan 6, 2016 at 10:32 AM, John ZuHone jzuhone@gmail.com
wrote:
>>>> >>>> I would go for modifying algae. >>>> >>>>> On Jan 6, 2016, at 11:30 AM, Matthew Turk <matthewturk@gmail.com
>>>>> wrote: >>>>> >>>>> Hi folks, >>>>> >>>>> For a long time we've used "algae," which was designed by Britton >>>>> about eight years ago, as the default colormap. This has been
really
>>>>> nice for "branding" yt -- if you see an algae plot, it's probably >>>>> (not >>>>> definitely) made with yt. But it's also not accessible from a >>>>> colorblindness perspective. Stefan van der Walt has been giving
some
>>>>> really great talks lately about building a better colormap for >>>>> matplotlib (e.g., https://www.youtube.com/watch?v=xAoljeRJ3lU )
which
>>>>> culminated in viridis, which is shipping in recent versions of >>>>> matplotlib and will become the default. >>>>> >>>>> In support of this, he built a tool called viscm which can
generate
>>>>> reduced versions of colormaps to show what they would be like
with
>>>>> varying degrees of insensitivity to color. I've generated
outputs
>>>>> from viscm of three of the custom colormaps we ship with yt: >>>>> >>>>> Algae: https://images.hub.yt/u/fido/m/d275d5e1-png/ >>>>> Cubehelix: https://images.hub.yt/u/fido/m/8e698928-png/ (I
believe
>>>>> this is now also shipped with MPL) >>>>> Kamae: https://images.hub.yt/u/fido/m/e0e40efa-png/ >>>>> >>>>> I love algae, but it's not the best from an accessibility >>>>> perspective. >>>>> >>>>> I'd like to propose that we use a new default colormap. If we do >>>>> this, I see two options: >>>>> >>>>> * Retain a "branding" by developing a new one either by using the >>>>> techniques used by matplotlib (or one of the maps they opted not
to
>>>>> use) or by modifying algae to be more accessible; looking at the >>>>> response functions, I suspect it would be reasonably possible to >>>>> modify it. (Modifying algae is my preference.) >>>>> * Use viridis (which we may then have to ship if we have older >>>>> versions of matplotlib to support) >>>>> >>>>> -Matt >>>>> _______________________________________________ >>>>> yt-dev mailing list >>>>> yt-dev@lists.spacepope.org >>>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>>> >>>> _______________________________________________ >>>> yt-dev mailing list >>>> yt-dev@lists.spacepope.org >>>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>> >>> >>> >>> _______________________________________________ >>> yt-dev mailing list >>> yt-dev@lists.spacepope.org >>> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >>> >> >> >> >> -- >> Erik Schnetter schnetter@gmail.com >> http://www.perimeterinstitute.ca/personal/eschnetter/ >> _______________________________________________ >> yt-dev mailing list >> yt-dev@lists.spacepope.org >> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org > > > > _______________________________________________ > yt-dev mailing list > yt-dev@lists.spacepope.org > http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org >
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
_______________________________________________ yt-dev mailing list yt-dev@lists.spacepope.org http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
-
B.W. Keller -
Cameron Hummels -
Erik Schnetter -
John ZuHone -
Kacper Kowalik -
Kenza Arraki -
Levy, Stuart A -
Matthew Turk -
Michael Zingale -
Nathan Goldbaum