HARDWARE: Benchmark test

FYI:

Test Total: 454 Graphics: 178 CPU: 123 I/O: 153

AMD 2000+. Asus A7N8X, 512MB DDR, Geforce 4 Ti 4600, 40Gig 7200 RPM Western Digital HD

Todd

Vary interesting.. I was wondering if anybody else might have run this test???? It does involve some large assembles, in an actual solid works environment. So real numbers.. I was hoping we could get some other people to run it, just to compare some numbers..

Test Number 1 Test Total = 1,040 Graphics = 587 CPU = 263 I/O = 190

This is with a Celeron 2.0Ghz, 1.045Gb ram, Matrox Parhelia 128MB, Win2K SP4, Solidworks 2003 SP4.

These where not really the numbers I was hoping for. As you can see from the graphics time, I'm not even close to the Quadro time of 141 sec. But this is a really advanced graphics card. I was wondering what "real" numbers some others might be seeing??

Thanks Kev>

"Carine & Kevin" schrieb im Newsbeitrag news:9ad2b.1458$ snipped-for-privacy@news20.bellglobal.com...

I have gathered something about 30 results from ou german online forum at

Your results are quiet strange, compared to those at my webpage. Did you do a "standard" or the "SPECacp" test (the SPECacp generate a lot hogher values ...)

HTH, Stefan

I used the SPECacp test.

I didn't understand the difference, what are they doing differently between the two? Is the Standard test the one I should be using?

Todd

Yes I too was doing the SPECacp test.. I will have to re run it using the other for comparison.

Thanks for the heads up on that.

Kevin

"Todd" schrieb im Newsbeitrag news:bidc27$84v65$ snipped-for-privacy@ID-201522.news.uni-berlin.de...

I can't say what is the difference between the two sets in that benchmark, for I didn't observe the process. But I saw a great difference in the results at my own computer (standard 248 - 34-134-80 and SPECacp 451 - 195-129-127).

All the results gathered at my webpage are the "Standard" test. So if you want to compare your results with these you should run the standard test.

HTH, Stefan

Did you do a "standard" or the "SPECacp" test

What is the link to the standard test?

Thanks,

CAD ROB

It is just an option you get when you run the test. There is no seperate link for either test, both are included with the download.

Todd

your 1.7ghz machine would probably benefit from more ram- I think most people run 1gb these days -512mb min

Kevin,

Your 1.7 GHz machine definitely needs more RAM. I believe that the benchmark says that you have to have 512MB. That's why the I/O score is so bad for the faster machine--it's spending all of its time swapping stuff in and out of RAM. The CPU score is bad because it is waiting for stuff to swap in and out.

The Graphics/CPU/I/O split is not at all accurate. A faster processor will almost always cut the time on all of the tests. In fact, all of the tests usually scale pretty closely to the processor speed.

Jerry Steiger Tripod Data Systems

When I upgrade here at work, I'll not be able to go to the fastest processor. Just cant seem to justify it.. The woodworking shop I work at only has 30 people total. 5 in the front office, 6 computes, 2 designers, you get the point. So what I'm asking is in around the 2.4Ghz mark what is the better way to go.

I was thinking about a Asus P4P800VM mother board, P4 - 2.4Ghz (800 FBS),

512 Mb dual channel memory (matched pair), the Matrox card (I know they will not let me loose that), then my 15Gb hard drive from my old system, Win2K Sp4, Solidworks 2003 Sp4.

What should I look for in an AMD self made system to be equal to this. All the reading I've been doing has been in regards of Pentium's.

Kevin

Kevin,

There are a lot of different motherboard manufacturers out there but I have stayed with Asus from my experience. The A7N8X has treated me well, and the latest version has the newest nVidia nForce chipset.

I just quickly priced out a A7N8X with AMD 3000+, 1 Gb Kingston RAM plus the $10 fot them to test it and it ran around $500. Thats for the fastest processor, obviously the slower processors will get more bang for the buck. Just going with an AMD 2700+ saves you about a $100.

Some other important items to look at when upgrading your computer:

1. Make sure the actual case will accept the motherboard. A lot of manufacturers have proprietary layouts that will not allow a new motherboard.

1. Make sure you have a good power supply with at least 350Watts. If you buy one, go with at least 400W just to make sure.

Thats all i can think of right now. Hope that helps.

Todd

Here are my results with the standard test Version 2003.1.6. My system is: Pentium 4 @ 2.8 gHz, 1GB Ram, NVIDIA Quadro4 900 XGL w/128 mB RAM, WinXp Pro SP1, SW2003 SP4.0.

Complete 164 Graphics 27 Proc. 86 I/O 51

WT

Sorry, I dont know of any links for computer stores up there. The one I have purchased from, and you can see if they will ship there, is

They have good prices on thier motherboard combos and some video cards.

Todd

Well, I think i have some of it figured out..

My Celeron only has a L2 cache of 128 Mb, I have a processor at work that is a true P4 2.0Ghz with a 512mb cache.. they have both same front bus speed (400). I'll have to swap them out and see what happens.. Who ever set up my system at the store also had the memory set to the wrong voltage as well (DDR-266).. that might not have helped...

this is the results as current with the Celeron.

Test Total = 395 Graphics = 71 CPU = 259 I/O = 65 Standard test

Car>

This is from my Home PC. It's an overclocked AMD set-up that performs quite well. Even though this system is stable at these speeds,I would not use an overclocked system for work. Mostly this system is for fun, and that just what I am having here, some fun with this benchmark. 8-D

ABIT NF7-S nForce2 motherboard Athlon XP2100+(1,730Mhz) at 2,453Mhz (223x11)

512MB PC3200 TYAN Tachyon G9700 Pro (ATI Radeon 9700 Pro)

Test Averages for 5 test(s). Test Total = 174 Graphics = 24 CPU = 87 I/O = 63

Complete Results:

**********************************************

SolidWorks 2003 Workstation Benchmark User Name : Gene Computer Name: GENE Manufacturer : Custom Built Model : ABTI NF7-S Rev-2

**** Overall Test Results **** Note: All results are in seconds. Lower scores are better. Test Number 1 Test Total = 197 Graphics = 25 CPU = 86 I/O = 86 Test Number 2 Test Total = 168 Graphics = 23 CPU = 87 I/O = 58 Test Number 3 Test Total = 165 Graphics = 24 CPU = 87 I/O = 54 Test Number 4 Test Total = 170 Graphics = 24 CPU = 85 I/O = 61 Test Number 5 Test Total = 168 Graphics = 24 CPU = 90 I/O = 54 Test Averages for 5 test(s). Test Total = 174 Graphics = 24 CPU = 87 I/O = 63