On Mar 3, 3:57 pm, Jeff Gilchrist <jeff.gilchr...@gmail.com> wrote:
> On Tue, Mar 3, 2009 at 10:32 AM, Cactus <rieman...@googlemail.com> wrote:
> > My machine is a Dell M65 portable as follows:
>
> > Intel Mobile Core 2 Duo T7400 running at 2.16GHz
> > Merom at 65nm, family 6, model F, stepping 6
>
> > L1 2 x 32 KBytes data
> > L1 2 x 3 2KBytes instructions
> > L2 4096 KBytes
>
> Hmm, a Mobile Core 2 might also contains some difference as well.
> Here are the details of the 65nm and 45nm desktop Core2 versions I
> have.
>
> Intel Core 2 Duo E6600 (MMX, SSE, SSE2, SSE3, SSSE3, EM64T)
> Conroe at 65nm, family 6, model F, stepping 6
> L1 Data 2 x 32 KBytes
> L1 Instruc 2 x 32 KBytes
> L2 4096 KBytes
>
> Intel Core 2 Quad Q9550 (MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, EM64T)
> Yorkfield at 45nm, family 6, model 7, stepping 7
> L1 Data 4 x 32 KBytes
> L1 Instruc 4 x 32 KBytes
> L2 2 x 6144 KBytes
>
> Major difference difference between 65nm and 45nm Core2 is the
> addition of SSE4.1 instructions, plus I think they were able to
> optimize some things to improve cycle times.
>
> The FSB speeds are also different between Xeon, Desktop ,and
> especially Mobile Core2's which would also have an impact on the
> benchmark scores would it not? Your mobile FSB is 667 MHz, the E6600
> runs at 1066 MHz, and the Q9550 runs at 1333 MHz.
Yes the FSB is different and I assume this will impact on stuff that
does not run in the cache. It would certainly be useful to have
Windows and linux results on the same machine.
One interesting idea is that MSVC *.lib libraries are supposed to be
interchangeable with gcc *.a libraries. So if I can get a mpir.a
libraty compiled for Core2, I may actually be able to run this in
Windows. It won't support stack unwinding but that doesn't matter too
much for testing.
Right now it would make sense to get as many benchmarks as we can.
Mariah put her mpirbench code in the files area on this site and I put
my python code there as well. I can let you have my windows build
project for mpirbench if that would help.
Brian
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"mpir-devel" group.
To post to this group, send email to mpir-devel@googlegroups.com
To unsubscribe from this group, send email to
mpir-devel+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/mpir-devel?hl=en
-~----------~----~----~----~------~----~------~--~---
