[Numpy-discussion] ANN: Scipy 0.15.0 release

Matthew Brett matthew.brett at gmail.com
Mon Jan 12 07:04:33 EST 2015


Hi,

On Mon, Jan 12, 2015 at 7:13 AM, Ralf Gommers <ralf.gommers at gmail.com> wrote:
>
>
> On Mon, Jan 12, 2015 at 4:50 AM, cjw <cjw at ncf.ca> wrote:
>>
>> Paul,
>>
>> Wot, no AMD64?
>
>
> Colin, this is well known from previous scipy and numpy releases. It's due
> to not having a freely available 64-bit compiler chain available at the
> moment with which we can build official binaries. You can get 64-bit Windows
> installers of most scientific Python distributions (like Anaconda, Enthough
> Canopy, WinPython) and for just Scipy from the site of Christoph Gohlke.
>
> Your tone on this list is not appreciated by the way, it borders on
> trolling. If you have a serious question to which you really don't know the
> answer, please pose it in a less disrespectful way.

Ralf - honestly I think it's best not to use the term 'trolling' under
any circumstances - it can be a heavy weapon [1], although it's
perfectly reasonable to say to Colin that you would find it helpful if
he used a different tone.  In this case, I couldn't be sure whether
Colin meant his email to be light-hearted or not.

Colin - just to add to Ralf's reply on the 64-bit issue - here are a few links:

* Stackoverflow answer with some references [2]
* Numpy mailing list question about 64-bit installers in 2011 [3]
* Another discussion I started in 2013 [4]
* Commentary on problems for Numpy etc on Windows [5]
* Our current best hope : Carl Kleffner's mingw-w64 build chain [6]

If Carl K is listening here - Carl - what's the current best way to help?

Cheers,

Matthew

[1] http://nipyworld.blogspot.co.uk/2012/06/define-troll.html
[2] http://stackoverflow.com/questions/11200137/installing-numpy-on-64bit-windows-7-with-python-2-7-3
[3] http://comments.gmane.org/gmane.comp.python.numeric.general/42118
[4] http://mail.scipy.org/pipermail/numpy-discussion/2013-February/065339.html
[5] https://github.com/numpy/numpy/wiki/Numerical-software-on-Windows
[6] https://github.com/numpy/numpy/wiki/Mingw-static-toolchain



More information about the NumPy-Discussion mailing list