Test Configuration
by
Uwe Scheffel
Tom's Hardware offers benchmark test results for the following graphics cards:
| ATI | Nvidia |
|---|---|
| ATI FirePro V8700 | Nvidia Quadro FX 5600 |
| ATI FireGL V8650 | Nvidia Quadro FX 4800 |
| ATI FireGL V7700 | Nvidia Quadro FX 4600 |
| ATI FireGL V7600 | Nvidia Quadro FX 1700 |
| ATI FireGL V7300 | Nvidia Quadro FX 5500 |
| ATI FireGL V7200 | Nvidia Quadro FX 4500 |
| ATI FireGL V7100 | Nvidia Quadro FX 3500 |
| ATI FireGL V5600 | Nvidia Quadro FX 570 |
| ATI FireGL V5200 | Nvidia Quadro FX 370 |
| ATI FireGL V3600 | Nvidia Quadro FX 1500 |
| CPU | Intel Core 2 Extreme QX6800 (2.93 GHz) |
|---|---|
| Motherboard | Asus P5W64 WS Pro (rev 1.01, BIOS 0802) with Intel 975X Chipset |
| System Memory (RAM) | 4 GB (4 x 1 GB) Corsair XMS2-6400 (CM2X10246400-C3) with CL 4.0-4-4-12 |
| Optical drive | Samsung SH-D163A , SATA150 |
| Power supply | Zalman, ATX 2.01, 510 W |
| Hard disk | Western Digital WD1500ADFD 150 GB |
| Operating System | Windows XP Service Pack 3 |
|---|---|
| 3D API Update | DirectX 9.0c |
| Intel Chipset drivers | Version 8.3.1.1009 |
| ATI FireGL driver | Catalyst 8.453.1-081008 |
| Nvidia Quadro driver | Quadro 182.08 + Maxtreme 11 |
| SPEC Benchmarks settings | Application settings according to SPEC Project Group Rules, driver using application optimizations if available |
| SPEC Benchmarks run | SpecViewperf 10.0 SPECapc 3ds Max 9 (3D Studio Max) SPECapc Solidworks 2007 SPECapc Maya 6.5 |
79
Comments
Related Stories :
- Nvidia Quadro FX 4800: Workstation Graphics At Its Finest?
- Nvidia Quadro FX 4800 Pro Goodness
- Nvidia Unveils New Quadro FX 5800
- PNY Releases Quadro FX 5600 SDI Card With 1.5 GB Of Video Memory
- Workstation-Shootout: ATi FireGL V7600 vs. Nvidia Quadro FX 4600
- Nvidia Quadro FX 3500 graphics standard in new Dell Precision workstation
- Nvidia announces Quadro FX 3450/4500 workstation graphics chips
- OpenGL: ATi FireGL X2-256t and NVIDIA Quadro FX 1100
why do i feel like when everyone compares workstation cards to gaming ones they get it wrong. a 4800 Fx will performs 99% like a 260GTX and if you softmod it to a Quadro than you have the same effect the other way around. really you are paying for driver support. i much rather just pay for the card.
They really mean it when they say great support. I once got a custom driver made specificly for my system overnight for a glitch I had. It saved me from loosing a client and a few thousand dollars for that one specific gig...
why do i feel like when everyone compares workstation cards to gaming ones they get it wrong. a 4800 Fx will performs 99% like a 260GTX and if you softmod it to a Quadro than you have the same effect the other way around. really you are paying for driver support. i much rather just pay for the card.
I think the comparison to the gaming card came from readers in past workstation card stories requesting such comparisons.
Great article, I appreciate the benches comparing the GTX 280 on workstation apps. I'll spend my money on gaming cards and leave it to corporations to purchase workstations...
I think the comparison to the gaming card came from readers in past workstation card stories requesting such comparisons.
this is challanging the consumers intelect with all things on the table ... actulay is the same GPU chip but performs so differently because of few modifications ... wonder how much this thing will keep up from nvidia and amd ... makeing their customers stupid so obvious ... i mean it is the same fukin engine at heart why sell it so overpriced ?
Looking at the results, I cannot understand how you can wholeheartedly recommend FX 4800 over cheaper FirePro V8700. Quadro benchmark results do not seem "convincing" to me since differences are quite small in most cases. The recommendation has to be based on type of work/application someone is using.
The naming is getting confusing again... FX4800 HD4650 HD4850 ...
that nvs 295 sounds interesting...
I'm getting tired of NVIDIA's crap: "....but our cards have CUDA support". Enough marketing! I think someone who's willing to buy a card because they want to program on the GPU MUST know that both vendors have a SDK for stream programming and it's actually the SAME thing. I've tried them both (FireStream and CUDA) and there are very little differences between them. If they wanna brag about 3rd party apps...well how many are they? 2? 3? Just wait until OpenCL (sdk and cl) is finally released and maybe then we'll see more applications in this GPGPU area and maybe they'll stop with this "oh but we have CUDA" thing.
I would like to see them use riva tuner and to tell the drivers it is not a geforce and see what kind of bench marking they get then. Or the other way around. Turn the workstation card into a gaming card. How about throwing a game on a workstaion card. See how it handles it.
I don't think a business will select a card based on how it does overrall in these benchmarks but how they do in individual benchmarks. I think there is a good reason to go with the FireStream instead if you are using something like Maya predominantly.
I think you do know that they are getting an extremely responsive support team which comes when you pay the premium price of Quadro, right?
fayskittles wrote:
How about throwing a game on a workstaion card. See how it handles it.
ya I agree. I want to see the score too. Crysis, 3D mark Vantage etc.
Does the card support SLI ? Or does the programs like Maya and 3DS support SLI ?
Thanks for the article. I wonder if there's a chance Nvidia will send you one of their Tesla solutions for review...
I think the comparison to the gaming card came from readers in past workstation card stories requesting such comparisons.
Eh, it would have been more interesting if it was softmodded. I know the GT200's don't take well to being softmodded (appearntly it is possible), but I doubt anyone thought the performance would be the same unmodified.
empstar, I'm with you on that one. Why only a one way comparo. Let's see the workstation card on the game front.
but will it play Crysis?
I always h