Jeff Mottle Posted August 11, 2009 Share Posted August 11, 2009 Hi all, As some of you may have seen in some other threads NVIDIA and ATI are sending me video cards to evaluate. I will of course be testing the top cards, but I also want to evaluate these against the cards that you would also consider when purchasing a new GPU. Specifically which cards you compare when looking at Quadro vs GeForce and FireGL vs FirePro I have a new v8750 sitting on my desk now, but I can get any other card as well. Please post the cards you would like to see reviewed and which cards you are comparing. Cheers, Jeff Link to comment Share on other sites More sharing options...
rmccoy Posted August 12, 2009 Share Posted August 12, 2009 I would be interested in a comparison between the GeForce gtx 275, gtx 285, gtx 295 cards Link to comment Share on other sites More sharing options...
AJLynn Posted August 12, 2009 Share Posted August 12, 2009 I think comparisons between cards of the same line sound very pedestrian. So are plain old reviews. What I'd like to see are analyses like: -What does SLI/Crossfire do for Max/Autocad/Revit? -nVidia and ATI cards at the same price point (expensive, medium, cheap) -Regular vs. Quadro/FireGL (two perspectives: the "equivalent" cards, e.g. same GPU generation and clock speed of Geforce and Quadro, and same price points) -What can you get Max to do with a very high end card that you can't get it to do with a $100 9800GT? I think you know what I think the answer is but I want somebody to check it Also, any specialized information you can think of that gamer sites don't cover: which card best handles the new Max on screen shader system, do any of these make Revit suck less, does video card choice matter at all to Autocad users (and try that with the i7 down-clocked until it's as slow as the P4's a lot of Autocad users are stuck with) etc. Link to comment Share on other sites More sharing options...
Tommy L Posted August 12, 2009 Share Posted August 12, 2009 I think the relentless march of upgrading needs to be held to account. In a review, I would like to know just exactly why I should trade my 8800gtx/Q6600/XP64 system (which quite frankly costs peanuts and performs like a charm) for a swanky new top of the line item. I think I have an average system for someone who is in the market to upgrade. There are a couple of hot topics right now: OS (xp64 vs W7) 7-8-9series nvidia vs GTX2** Q-series vs I7 are the first things that come to mind. I dont want to see a comparison between the new line items (this is highly publicized anyway), I want to see the practical cost/benefit analysis of existing standards to the new fangled for the small operator. Why the small operator? The little guy (small studio/freelancer/hobbiest) are the guys who care most about the small margins. Just my 2c, Tom. Link to comment Share on other sites More sharing options...
Jeff Mottle Posted August 12, 2009 Author Share Posted August 12, 2009 I think comparisons between cards of the same line sound very pedestrian. So are plain old reviews. What I'd like to see are analyses like: Awesome thanks! There were a few reasons both companies were interested in having me review their cards: 1. They want to know why everyone is buying Geforce/FireGL vs Quadro/FirePro cards and what features, performance/price differential would push the decision to the Quadro/FireGL side. 2. They want "real" reviews as you said and I told them I was not going to run the typical SpecPerf tests that you see on all of the gaming sites. I said I would source real world models that could be used to create real world tests. I'll also be working with ATI and NVIDIA engineers to find out what types of models will put the cards through their paces to really showcase the difference between a gamer card vs a pro card. I also found out that the SpecPerf tests aren't updated as quickly as the card features are updated so a lot of times the results don't really showcase the true performance of a card. 3. No one has done hardware reviews specifically for our industry. This type of feedback is exactly what I need. I'll also be looking for real-world models/scenes to test, so if you have any let me know. I'll be starting the reviews in Sept as I'm going to be out of the office for the next 3 weeks, but will be arranging all of the cards and models for my return. Link to comment Share on other sites More sharing options...
mi75 Posted August 12, 2009 Share Posted August 12, 2009 I look forward to the reviews Jeff I think Tommy L and AJLynn pretty much summed up what I'd also like from these reviews. Mart Link to comment Share on other sites More sharing options...
kippu Posted August 12, 2009 Share Posted August 12, 2009 good news jeff, will wait for your reviews Link to comment Share on other sites More sharing options...
Crazy Homeless Guy Posted August 12, 2009 Share Posted August 12, 2009 (edited) A 3. No one has done hardware reviews specifically for our industry. If you are looking for a specific test.... I would like to see on how different cards reacts to the high object count with heavy multi/sub-object materials that come with large scale projects imported into Max from Revit. As BIM trickles to all arch firms, this will be an issue for everyone. Edited August 12, 2009 by Crazy Homeless Guy Link to comment Share on other sites More sharing options...
luckyluc Posted August 12, 2009 Share Posted August 12, 2009 This sounds interesting Jeff, I would also like to see the difference between the GTX top cards and the Rad. HD4870 and HD4890 topcards in MAX. Link to comment Share on other sites More sharing options...
MegaPixel Posted August 12, 2009 Share Posted August 12, 2009 Awesome thanks! There were a few reasons both companies were interested in having me review their cards: 1. They want to know why everyone is buying Geforce/FireGL vs Quadro/FirePro cards and what features, performance/price differential would push the decision to the Quadro/FireGL side. I have yet to be convinced that the Quadros performance compared to the Geforce is noticable enough to pay all that extra money for it. I put personal money on a Quadro years ago for a MAX workstation and was amazed to see that my Wife's low end geforce card at the time handled my projects soo much better. I wouldn't mind seeing some reviews or dicussions for or against either product. Thanks - MP Link to comment Share on other sites More sharing options...
mattclinch Posted August 12, 2009 Share Posted August 12, 2009 performance benefits of MAXtreme drivers with Quadro cards...? Link to comment Share on other sites More sharing options...
SandmanNinja Posted August 12, 2009 Share Posted August 12, 2009 performance benefits of MAXtreme drivers with Quadro cards...? +1 Noone in the computer industry benchmarks that. It would be great to see a high performance video card WITH optimised video drivers for the software that we use. How will you be testing the video cards? Will you move the mouse around the viewport quickly and note the fps? Will it be a light/texture/DISPLACEMENT map/poly heavy scene in 3ds? If so, would we be able to download the test file ourselves? "Say, if I pop $200 on a new video card, I know my video performance will quadruple" Link to comment Share on other sites More sharing options...
cadmunkey Posted August 12, 2009 Share Posted August 12, 2009 Awesome thanks! There were a few reasons both companies were interested in having me review their cards: 1. They want to know why everyone is buying Geforce/FireGL vs Quadro/FirePro cards and what features, performance/price differential would push the decision to the Quadro/FireGL side. I hope they are joking, maybe its the extreme price difference for a worse spec'd card that puts us off? Link to comment Share on other sites More sharing options...
tec Posted August 12, 2009 Share Posted August 12, 2009 Hi Jeff I am interested in comparing nvidia FX580 vs FX1800 in MAX & Revit. Is there a big difference in performance with standard models ( Thanks Link to comment Share on other sites More sharing options...
Jeff Mottle Posted August 12, 2009 Author Share Posted August 12, 2009 I hope they are joking, maybe its the extreme price difference for a worse spec'd card that puts us off? Yeah, the general consensus seems to be that a GeForce performs better per dollar, so I'm curious to know if there is something we are all doing wrong that we are not getting magnitudes of speed increase. I specifically mentioned this to the execs I spoke to and I think they were all quite surprised, so hopefully I can create a definitive test and ask their engineers how to get the performance out of the quadros cards that make up that price difference. Link to comment Share on other sites More sharing options...
AJLynn Posted August 12, 2009 Share Posted August 12, 2009 I don't know how much software you want to be trying here, but I know there are specific benefits to running a workstation card in OpenGL intensive software like Maya and Alias Studio that you might be able to address. (Of course, I don't know how many people here use either of those.) In Max it doesn't matter much because Max wants you to use Direct3D, and Cinema4D... I'm not sure about that one. On the one hand it seems to be perfectly happy on gamer cards. On the other hand my desktop with a measly FireGL 5200 beats the tar out of other machines I've tried in Cinebench. Link to comment Share on other sites More sharing options...
cadmunkey Posted August 12, 2009 Share Posted August 12, 2009 Yeah, the general consensus seems to be that a GeForce performs better per dollar, so I'm curious to know if there is something we are all doing wrong that we are not getting magnitudes of speed increase. I specifically mentioned this to the execs I spoke to and I think they were all quite surprised, so hopefully I can create a definitive test and ask their engineers how to get the performance out of the quadros cards that make up that price difference. That would be great Jeff if you are able to get that kind of feedback, nice one. Link to comment Share on other sites More sharing options...
Devin Johnston Posted August 12, 2009 Share Posted August 12, 2009 I'm also very interested to see how the GeForce and Quadro cards compare to each other, I know I've had both and the Quadro cards seem to be more stable and powerful but I've never tested it. In my current system I have a Quadro FX 3700 which was an upgrade from the Fx 1700 I previously had, I can't say I've seen any massive difference in performance and it's certainty not enough to justify the price difference. Link to comment Share on other sites More sharing options...
johr Posted August 12, 2009 Share Posted August 12, 2009 Indeed it's a great idee, some scores are really needed for the most of us not in the professional(ultra expensive) stuff. I'm really interested to see also the unofficial stuff related to soft-moding those budget/modest gamer cards (200-300 price tag ). Link to comment Share on other sites More sharing options...
Devin Johnston Posted August 12, 2009 Share Posted August 12, 2009 We also seem to be on the edge of a real time rendering revolution that's revolving around these graphics cards, it would be great if you could somehow evaluate them based on their possible performance as it relates to these new functions. Link to comment Share on other sites More sharing options...
trajic Posted August 12, 2009 Share Posted August 12, 2009 It would be nice to get your hands on Geforce and Quadro cards of similar specs and test them with GPU based RT in the future! Link to comment Share on other sites More sharing options...
Zdravko Barisic Posted August 12, 2009 Share Posted August 12, 2009 Jeff Please play this test on your graphic cards, also others can do the same, so we can compare the results, it's a real world 3DSMAX test. Real scenes, not SPECviewperf. Here's the test http://rapidshare.com/files/3108339/bmark4.rar.html Just start the script and look at the TXT file, if script does not start the first time, try few time, it works on any max. THANK YOU! Link to comment Share on other sites More sharing options...
R.S Posted August 12, 2009 Share Posted August 12, 2009 Its worth mentioning that Quadro & FireGL cards are designed to display on big panels (dual 2560x1600). Most benchmarks typically run at 1280x1024 and as such the performance benefit of the workstation cards is untapped. The whole point of workstation cards is to power the viewport line anti-aliasing, shading/lighting & textures with minimal redraw and at huge resolutions. In the typical arch/viz environment it's less of a concern as big displays are less common (I personally only use a 1680 x 1050 panel). To test the cards i would suggest two setups, one typical panel (1080x1920) and dual panels with 2560x1600max each. For the benchmark simply setup a flythrough cam in a scene, run it in the viewport, record the FPS with something like FRAPS, repeat 5 times & average it. I have a feeling the performance of the workstation cards will only be apparent on the hi-res test but who knows, after all thats the point of this test. Link to comment Share on other sites More sharing options...
rmccoy Posted August 13, 2009 Share Posted August 13, 2009 I would be really interested in seeing the performance gains you could get with a tesla card. I think this will really only matter once vray rt gpu is released. would it actually be worth investing in one of these cards or just getting a different higher end card. Link to comment Share on other sites More sharing options...
mattclinch Posted August 13, 2009 Share Posted August 13, 2009 To test the cards i would suggest two setups, one typical panel (1080x1920) and dual panels with 2560x1600max each. For the benchmark simply setup a flythrough cam in a scene, run it in the viewport, record the FPS with something like FRAPS, repeat 5 times & average it. +1 I'd also recommend a single panel running 2560x1600 which is what I run. Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now