<br><br><div class="gmail_quote">On Sat, Sep 29, 2012 at 9:07 AM, Paul Moore <span dir="ltr"><<a href="mailto:p.f.moore@gmail.com" target="_blank">p.f.moore@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="im">On 29 September 2012 10:17, Stefan Krah <<a href="mailto:stefan@bytereef.org">stefan@bytereef.org</a>> wrote:<br>
> Tim Delaney <<a href="mailto:timothy.c.delaney@gmail.com">timothy.c.delaney@gmail.com</a>> wrote:<br>
>> If those numbers are similar in other benchmarks, would it be accurate and/or<br>
>> reasonable to include a statement along the lines of:<br>
>><br>
>> "comparable to float performance - usually no more than 3x for calculations<br>
>> within the range of numbers covered by float"<br>
><br>
> For numerical programs, 1.4x (9 digits) to 3x (19 digits) slower would be<br>
> accurate. On Windows the difference is even less.<br>
><br>
> For output formatting, cdecimal is faster than float (at least it was when<br>
> I posted a benchmark a couple of months ago).<br>
<br>
</div>To me, this means that the key point is that for the casual user,<br>
float is no longer the "obvious" choice. You'd choose float for the<br>
convenience of a built in type, and Decimal for the more natural<br>
rounding and precision semantics. If you are sufficiently interested<br>
in performance for it to matter, you're no longer a "casual" user. (Up<br>
until now, I'd have said use float unless your need for the better<br>
behaviour justifies the performance loss - that's no longer the case)<br></blockquote><div> </div><div>Does this mean we want to re-open the discussion about decimal constants? Last time this came up I think we decided that we wanted to wait for cdecimal (which is obviously here) and work out how to handle contexts, the syntax, etc.</div>
</div>