The review is on

[juin]

With a suse Linux 64 bit for X86-64 (champion by IBM)

[-peep-] french

 

[juin]

Toms expect a 30% yield this is bad

[-peep-] french

 

[juin]

Too bad the review was short and not finish.

Wockstation bad performer.Server apps they have win were AXP 3000+ was allready winning Suse was never been use with a intel cpu why not put intel CPU with Windows 2003 or Redhat9 that use intel compiler and have a version for I2.

[-peep-] french

 

[imgod2u]

"Floating point and SIMD operations (SSE, SSE2) profit from 64-bit processing. A 64-bit processor can natively calculate the important 64-bit floating point format ("double precision" - precise up to 15 decimal places) and is therefore faster - this is the main reason why 64-bit processors take the lead in the floating point benchmarks."

I'm not too sure of what he meant by this but the move to "64-bit computing" has nothing to do with FP precision. FP precision is already 64-bit in precision even on "32-bit processors". Perhaps that wasn't what he meant but the whole page was dedicated to the difference between "32-bit processors" and "64-bit processors" and that's clearly not one of the differences.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.

 

[reever2]

Hey i got a better idea juin, how about put Opteron in a 64bit OS and using non-intel compilers when doing workstation benchies, oh wait everything has to go intels way doesnt it?

 

[imgod2u]

How about actually finding 64-bit OS on the workstation level and software that does? And btw, for their Spec benchmarks, AMD themselves chose to use Intel's compilers. Why? Because they're good compilers and provide better performance than Microsoft's, IBM's or GCC. Imagine, Intel's compilers helping the Athlons.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.

 

[Schmide]

Who here thought for sure they were going to pull of the HSF on the Opterons just to see what was going to happen? I mean they yanked the fan.

The review shows that the Opteron needs a 64 bit recompile to show its true colors. Extra registers mean extra performance.

P4/Xeon has a major advantage due to its high FSB and clock. For SSE related operations Intel is still king.

Opteron is a math monster though.

Dichromatic for your viewing plesure...

 

[imgod2u]

I think the server-level performance we're seeing with Opteron can be attributed to the dedicated memory pocket for each processor and the scalar hypertransport network. Traditional SQL server transactions can be very memory-dependent.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.

 

[juin]

Just look how AMD benchmark they own and you will get it.

[-peep-] french

 

[juin]

My SQL ??? i have not see 1 TPC-C result with a Linux and SQL for a server chip yet except from Tomshardware.Most use 9i or DB2

[-peep-] french

 

[juin]

There were a benchmark you can compare 32 bit to 64 bit.32 bit is faster most of the time.Let wait for Aces benchmark.

Kelledin care to make comment.

[-peep-] french

 

[juin]

Sun does support Opteron at the end.


http://www.theinquirer.net/?article=9085

This is more important that the launch itself.

[-peep-] french

 

[8procstooslow]

The Lightwave and 3dmax benchmarks are interesting.
Upto now the Athlon was not a good a good performer due to the P4 optimizations (I assume SSE2). However now the opteron is very close in terms of performance, which probably means SSE2 is being used. Given the 1.2Ghz clockspeed difference, this point to a very efficient implementation of SSE2. Give the opteron a couple of extra Mhz and Lightwave will fly

 

[sabbath1]

I'd say the server performance of the Opteron is supreme! Finally, AMD has beat Intel in that field when regarding performance! With the only exception of the Itanium 2...

Also, the Opteron finally has integrated thermal protection, no more worries about burning AMD CPU's.
As well as a heatspreader to protect the CPU from being crushed by it's heatsink.

Overall a seemingly fantastic product. AMD has done it!

 

[castle]

That's a great "fair" comparison!

see my new thread for more info.<P ID="edit"><FONT SIZE=-1><EM>Edited by castle on 04/22/03 01:31 PM.</EM></FONT></P>