Hello,

> I have to disagree - it says it all about trig functions, that's all. If
> the majority of people want to run ray-tracing programs then you're
> absolutely right of course. My program has no trig in it at all, so the
> benchmarks are completely different.

This particular example, maybe (I haven't studied PoV's internals). Though
of course the FPA11 does not implement trig in hardware, either, it still
uses software to perform the functions (albeit using the hardware's simpler
instructions).

However, evidence as regards the 700+FPA11 and the StrongARM (not the Alpha
or C) was in fact gained in Hong Kong, as I don't have one. We also developed
some code to simulate radar returns from a 3D database, which used no trig
(well I looked up some sin's and cos's at the beginning). This program
exhibited a performance increase of 2.5 times on the 700+FPA11 over the
250MHz StrongARM (bear in mind, faster than a typical one). I can mail you
the source if you're interested.

> Presumably most people *do* just want to make ray traced images rather than
> running hydrocodes etc (or in my case a financial dynamics model), which is
> probably a good thing since the output is usually a lot prettier :-).

;-)

	Yours aye,

	Neil

--
* Neil A Carson
* The Royal Military College of Science, Shrivenham
* e-mail carson@rmcs.cranfield.ac.uk
* Address: Kitchener Mess, RMCS, Shrivenham SN6 8LA. Tel: (0370) 593183