[ Answering the following: ]
> I ran some code to check these functions before I made my
> statement and when I used the sinf and cosf functions the test
> program I was using showed no improvement. So does this mean that
> my libm is is broken?
> Always possible, but very unlikely. You can look at
> s_sinf.o and k_sinf.o in libm using objdump -d. You'll
> find that all fp ops are done in single precision (muls,
> lds, etc) so, fundamentally, the routines are certainly
> not broken (at least not in the way you suggest ;-).
Naively, one would expect that, to compute the sine of some
argument in single precision would require *less* fp operations than
to get it correct up to double precision.
E.g., because almost all calculations _we_ perform are single
precision, on my NeXT I could always get away with a simple 200 line
library of exp, log, etc. - coded by someone of our local nuclear
physics lab - that simply cut some corners here and there. Those
routines were, on average, twice as fast as the ones provided by
NeXT.
> On the Alpha, there is only two reasons why single
> precision might give improve execution time:
> (a) Your code is memory bandwidth limited.
Yep, saw exactly that when we got an early Alpha on loan in January
'93 - the execution time difference between the same model, same
input, when running 32-bit calculations or [the same number of]
64-bit ones was of the order of 60 %, i.e. it took 1.6 times as much
time to pull the (twice larger) data stream through.
Hope this helps,
Toon.
-- Toon Moene (toon@moene.indiv.nluug.nl) Saturnushof 14, 3738 XG Maartensdijk, The Netherlands Phone: +31 346 214290; Fax: +31 346 214286 URL: http://www.knmi.nl/hirlam
-- To unsubscribe: send e-mail to axp-list-request@redhat.com with 'unsubscribe' as the subject. Do not send it to axp-list@redhat.com
Copyright © 1995-1997 Red Hat Software. Legal notices