Guy Baruch wrote:
>
>
> Oded Arbel wrote:
>
>> Hi list.
>> ...
>> First I've run the OpenSSL speed test on several machines I have
>> access to, and summerized the 512b and 4096b keys verifys/sec. to
>> make a long story short, the AMD Duron 700MHz I'm running at home had
>> 7202,202 respectivly, while the P4 1.5GHz I'm running at work showed
>> 6910,191 respectivly.
>
>
> what is sample size ? what is standard deviation ? what are
> significant digits ?
Who cares - I'm not doing an exact sciense, just trying to get the feel
for it. I did two or three runs on each CPU and they all came about the
same (not more then 1-2 precent each side) so I just took the first result.
>> I was thinking to myself - I know that Pentium 4 uses a radicly
>> different processor architecture then P3 and other x86 compatibles
>> and needs to have programs recompiled to it for the best performance
>> (IIRC I've read it somewhere).
>> I tried to recompile OpenSSL for my Penitum 4 - apparently it wasn't
>> that easy.
>
>
> what compiler version ? did you try intel's compiler ?
Gcc 3.2. sorry I didn't mention it earlier. I didn't try intel's
compiler, but if it's not super easy as a drop in replacement for gcc
then I'm not going to do it. I wanted to know how my processor rates
using standard software and whether re-compilation would give much
better results considering the work that needs to be done. since I like
to keep my system uptodate on everything, recompiling everything using
something more difficult the --rebuilding SRPMs will take too much time
to be of any use.
Where can I get that compiler ? I though that gcc has support for
pentium4 optimizations - the man page certainly suggests that.
>> I just went ahead and ran 'openssl speed' on the binary produced in
>> the build directory. interestingly enough, the results were even
>> lower (!) then the previous benchmark on the same machine.
>
> again, could be measurement error .
I _know_ its a measurment error - but still I would expect some
improvment and not statisticly the same result.
>> TW - It's a P4 with SDRAM, while the Duron is sporting DDRs. not that
>> I think it matters, as the OpenSSL speedtest is about raw processing
>> power and not memory bandwidth - a lowley 72pin DRAM can handle the
>> memory requirements of that benchmark.
>
> compile it statically, send it to a friend with RDRAM P4 , test
> hypothesis ...
> (also IIRC there are utilities to measure cache miss ratio, STW for one)
That would be interesting - any one on the list has a RDRAM P4 and would
like to help me test a little ?
--
Oded
::..
(This feature enables rc to export functions using here documents into the
environment; the author does not expect users to find this feature useful.)
-- from the rc(1) man page
=================================================================
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
