Please register or login. There are 0 registered and 1101 anonymous users currently online. Current bandwidth usage: 326.30 kbit/s December 10 - 10:50pm EST 
Hardware Analysis
      
Forums Product Prices
  Contents 
 
 

  Latest Topics 
 

More >>
 

    
 
 

  You Are Here: 
 
/ Forums / Other Hardware /
 

  Give us feedback! 760 MP vs. Xeon DP 
 
 Author 
 Date Written 
 Tools 
Sander Sassen Jun 16, 2001, 08:49am EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List Replies: 20 - Views: 1968
Hello everyone,

Just wanted to let you know that we're working on a review of the 760 MP dual Athlon MP platform vs. the dual Xeon DP. I'm interested to find out what benchmarks you'd like us to run or what other tests you'd like to see performed on either system.

Just post your replies in this thread, and please add some motivation why you think a specific benchmark is worth looking into. Furthermore if there's other stuff you've been missing in other reviews but is of key interest to you, let us know and we'll do our best to get that into the review.

Looking forward to your reply!

Thanks and have a good weekend!

Sander Sassen

CEO, Hardware Analysis
Email: ssassen@hardwareanalysis.com
Visit us at: http://www.hardwareanalysis.com


Sander Sassen
Editor in Chief - Hardware Analysis
ssassen@hardwareanalysis.com
Want to enjoy fewer advertisements and more features? Click here to become a Hardware Analysis registered user.
Justin Jun 18, 2001, 03:20am EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> [No Subject]
I would say ALOT of real world tests, like 3DStudio Max, and others like that

Robert Kropiewnicki Jun 18, 2001, 10:39am EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Benchmarking
Hmmmm, wish I knew which benchmarks outside of Sandra and 3DMark tested what.....

I do have one suggestion for you, if you can. See if you can include a dual T-Bird setup in the testing as well. I, for one, would be interested in seeing how much of a difference the hardware prefetch (among other enhancements) helps the Palomino based MP.

Here's one thing I can suggest......if there are tests out there that you're aware of that utilize SSE and SSE2, use them among the tests (will help show how much of a difference the included SSE set makes in MP), but make sure you point out which ones are using them.

I think the important thing in all this is to be very clear as to what each benchmark you are using is actually testing and why you're using that particular benchmark (outside of Sandra which seems to be the starting point of most benchmarks).

Make sure all BIOS revisions, drivers, etc, are the latest and greatest OFFICIAL version. Detail the revision levels of everything so we all have a clear idea of what is actually being benchmarked. Try to keep everything as consistent as possible wherever possible.

Dan Mepham Jun 18, 2001, 10:08pm EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> [No Subject]
Are you guys more interested in, for example, server-type benchmarks (ie, stuff real adopters of a Tyan Thunder K7 are likely to use), or in stuff that may be more relevant to YOU (I'm assuming most of you don't run hardcore server apps..)...?

Or both? :)

Dan Mepham

Editor in Chief, Hardware Analysis
Email : dmepham@hardwareanalysis.com
Visit us at : http://www.hardwareanalysis.com

Dan Mepham
keboman Jun 18, 2001, 10:37pm EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> [No Subject]
Q3 Benchs!

AHAHAHA

just kidding


Robert Kropiewnicki Jun 19, 2001, 11:50am EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Server benchmarks, mmmmmmmmmmm
Dan, Sander, et al,

Server benchmarks and workstation type benchmarks should definitely be included. Regardless of whether most people here would be looking at it for a home rig (haven't been here long enough to get a feel for the readers), both of these platforms are aimed at the workstation and 2-way server environment. Thus, any benchmarks you could do to show the power of these babies in the environment they were aimed at would be much appreciated.

That said, I would still suggest running Sandra at the very least since that's one of the all purpose benchmarks most seem to relate to and can get an idea of how much more performance you get vs. a uni-processor setup.

Randall Kennedy Jun 19, 2001, 06:04pm EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Don't Forget OfficeBench 2001!
Sander,

As you already know, OfficeBench 2001 delivered some rather compelling results in AnandTech's recent 760MP coverage. Not only did our tool provide effective differentiation of single processor systems (Athlon MP 1.2GHz was nearly 57% faster than Xeon 1.7GHz), it also demonstrated why dual-processors are a must have for the serious knowledge worker (dual-Athlon MP 1.2GHz was over 93% faster than the fastest single-CPU system).

We have the most compelling, real-world, SMP-aware benchmark out there. I encourage you to use the eval copy we provided to Hardware Analysis and to let your readers know what you find.

Regards,

Randall C. Kennedy
Director of Research
Competitive Systems Analysis
http://www.csaresearch.com

keboman Jun 19, 2001, 06:20pm EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> [No Subject]
How about some SETI scores, Folding@home, etc.
Whilst I don't use them, they have gained in popularity and would help to differentiate your article from the crowd.

Kebo

Justin Jun 19, 2001, 10:15pm EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Tests
Sander,

I would like to see some really good tests but i am not expecting you to use a FireGL video card to see how well 3DStudio Max will run but what i would really like to see is how well and what programs ( I also mean OS tests) that will use the MultiProcessor enviroment. Just as a sugjestion try seeing how long it will take you to render out a full screen in 3DStudio Max, then try Auto CAD then Photoshop 6.0 or whatever you happen to have, I would really like to put on of these dual systems together for myself

Rich Hayes Jun 21, 2001, 08:44pm EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> [No Subject]
One thing that seems important to me as an SMP fan, but has got little to none interest in reviews is the point that SMP systems may very well be the popular high end gaming system of choice by the end of this year.

With the new Microsoft OS being based on the NT core for both home and business user Win98 will no longer be the king of video games,
also... now that both types of users will be using basically the same OS
you might as well pick up one extra CPU and the professional version
of the OS.

I know many argue that the gaming performance increase is minimum..
you must take into account more than just FPS. Overall gaming
on an SMP system is more fluid and responsive, also if you check the task manager you will find that Win2K divides the load between the CPUs very well.
There are arguments for both sides of this story, BUT when WinXP
is the standard OS there is no doubt that two CPUs will run better
than one on the same OS.
This should also make SMP much more attractive for other types of home users as well.


Anyway.. what I am getting at is.. heck yeah I want to see gaming benchmarks.

Also you may wish to consider mentioning the WinXP release in your review. After all.. anyone planning on purchasing such a nice high end system should be pleased to know that the system will (*should*)
be much more compatible with games and other software that was traditionally tweaked towards Win98 users.

Well.. good luck with the review... I KNOW you are having a blast
playing with all that great hardware. :)


Rich

futureman Jun 26, 2001, 11:43am EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> [No Subject]
How about running your tests with a full slate of apps open in the background representative of those used in a typical workstation environment along side your regular tests. Using a 'clean' system is all fine and dandy, but the average user has a bunch of stuff going on at any moment.

I think that by doing this you will get a better overall picture of how the 2 systems compare.

Maybe something like a CAD session with 3dstudioMax and ICQ open while benching photoshop...

NickName Jun 26, 2001, 02:31pm EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> That sounds good.
I also think that you should try it with about 30-40 IE windows open. That takes up emense ammounts of physical memory and virtual memory. That would be good.

DaveO Jul 02, 2001, 06:03am EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Those little bridges
Don't forget to fix those little bridges AMD so thoughtfully cut for you ;) I know you weren't gonna forget about that... :D


--------------------------------------
What does this button do?
Brendan Harnett Jul 02, 2001, 01:27pm EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> How about some realworld apps, workstation style?
I'd love to see some stopwatch testing of the high-end workstation apps likely to be run on a machine based on Athlon MP. I don't think it will have any serious server adoption (even if it was the best option out there (which we don't know for sure) IT folks aren't going to touch anything AMD with a barge pole). I don't think the current Athlon MP will be attractive to gamers and hobbyists, but I know graphics professionals who have already oredered their Athlon MP based systems!

I'd love to see as many of the following as you can swing:

Softimage
Maya
Lightscape
Form Z
3DSMax
Rhino
Microstation
SolidWorks


And maybe some video stuff too like...

Premier
Aftereffects

And some Visual stuff:

Photoshop
Indesign
Freehand
Illustrator

These kinds of things don't need to be 'official' benchmarks. The tasks that many of these programs perform can take hours (or months), and simply holding a stopwatch to them can be a very good indication of performance, since 2 seconds is nothing to a 4 hour render.

This is what I think would be most informative,

Hope to see this article!

-Brendan


Brendan Harnett Jul 06, 2001, 03:43pm EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> So what do you think?
I want to see the article!

Rene Heindl Jul 23, 2001, 05:28pm EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> dual Athlon vs dual Xeon
I would be very interested in that article. As I am a CAD-user I would be eager to see how these machines perform in pro/Engineer or CATIA. Also it would be interesting to determine, if there is a performance gain when compared to a non dual system (such as single PIII, P4 or Athlon).
I am looking forward to the mentioned article. Further I'd like to congratulate you, this seems to be a rather useful forum....so go on like that!

Greetings from Austria

Rene

Rene Heindl Jul 30, 2001, 05:09pm EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> various professional graphics cards
It also would be interesting to see, how the mentioned systems perform their tests with various graphics cards for the Cad-Market (Elsa Gloria III, ATI FireGL2, 3D-Labs and so on). I have heard that the Graphic card Manufacturers differently take advantage of the processor and chipset-architectures, would be interesting to see if this prooves to be right.

Greetings

Rene

John Petersen Aug 01, 2001, 12:45am EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Athlon MP and Xeon MP Benchmark
As already stated by another posting, I suggest that you use setiathome as a benchmark. We benchmark all hour systems with the screen off. Currently across 96 data sets our dual 1.2 GHz Athlon system is running at 5.8 hours per analysis set. We have not checked it with the screen on but with our dual PII and PIII systems did not show a speed hit with the screen on and so may not be a problem but this goes untried.

John

John Petersen Aug 01, 2001, 12:46am EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Athlon MP and Xeon MP Benchmark
As already stated by another posting, I suggest that you use setiathome as a benchmark. We benchmark all hour systems with the screen off. Currently across 96 data sets our dual 1.2 GHz Athlon system is running at 5.8 hours per analysis set. We have not checked it with the screen on but with our dual PII and PIII systems did not show a speed hit with the screen on and so may not be a problem but this goes untried.

John

John Petersen Aug 01, 2001, 12:46am EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> Athlon MP and Xeon MP Benchmark
As already stated by another posting, I suggest that you use setiathome as a benchmark. We benchmark all hour systems with the screen off. Currently across 96 data sets our dual 1.2 GHz Athlon system is running at 5.8 hours per analysis set. We have not checked it with the screen on but with our dual PII and PIII systems did not show a speed hit with the screen on and so may not be a problem but this goes untried.

John

Julie Pelletier Aug 12, 2001, 10:13am EDT Reply - Quote - Report Abuse
Private Message - Add to Buddy List  
>> dual athlon vs dual xeon
where I can find the benchmark test result you do ?
I look to buy one of the 2 config.

Thanks

Julie


Write a Reply >>


 

    
 
 

  Topic Tools 
 
RSS UpdatesRSS Updates
 

  Related Articles 
 
 

  Newsletter 
 
A weekly newsletter featuring an editorial and a roundup of the latest articles, news and other interesting topics.

Please enter your email address below and click Subscribe.