NVIDIA Quadro 4000 by PNY 2GB GDDR5 PCI Express Gen 2 x16 DVI-I DL, Dual DisplayPort and Stereo OpenGL, DirectX, CUDA, and OpenCL Professional Graphics Board, VCQ4000-PB

I bought this card to use mainly with Maya and Mudbox. I had been using a Quadro FX 1700 (512 MB) and I expected better. And I got that, and more. Maya is much snappier working with scenes of 500,000 polygons (63 skinned, textured models). I built the scene using the FX 1700, but it took a noticeable amount of time to select a model or move the camera. With the Quadro 4000 (2 GB) everything moves smoothly. I am sure there are limits on how large a scene you can be manage even with this card, but I haven't tried to create something that breaks it.On my older Quadro card, I had to sacrifice some gaming performance to get decent performance on Maya 2011. Cards like the GT 9800 (and the 325M I have in my laptop) work, but you can tell the difference when you select components (particularly vertices or edges). On a Quadro, you get the selection faster and more accurately, especially when working in the perspective view. On consumer cards, sometimes the wrong thing gets selected, and often there is a defined lag between when you click and when the selection is highlighted.However, on the Quadro 4000, I get great gaming performance, for example running Left 4 Dead 2 at 1920x1080 with high detail with a good frame rate rather than having to reduce the resolution at run lesser detail that I had to do with the old card. I could run games full bore with the GT 9800, but I had performance issues with Maya (even Maya 2009).I am well pleased with the card. My PC had only a 350 Watt supply, this card works well with that, so I didn't have to swap out to a larger power supply. I don't have but a single hard disk in this machine (lots of network storage), so I expect that helps on the power consumption, I did notice that even after an hour of gaming I never once heard a fan switch on high rate, which, while I understand the need for this, I still find annoying.Given the price range, I expect this will become a popular card for those that need serious performance with professional level 3D programs.

Would not recommend for high performance games, but is a great product, and ran the games that I play with ease.Good graphics, and fit well into my tower with very little noise.

Installed this into an HP Elite desktop (Windows 7 64-bit OS). It fit with only about 2 mm clearance to spare at the ends of the board, but it does fit & it works well. I run NX mechanical CAD and the graphics performance is good.The only issue I had, a minor one, was that the installation instructions were not too clear about whether I was to hook up one or two power leads to the board. I interpreted the instructions to say I needed only one power cable hookup, but it turned out I should have hooked up two. I called PNY's technical support line after hours & left a message describing my question. The next morning a technician called me back & clarified that in my case I needed two power hookups. The tech was knowledgeable, helpful & pleasant. Because I had only run the computer briefly with just the one power hookup (to make sure everything powered up OK, and to download the latest driver), no harm was done.This graphics board performs well and I appreciated the timely call-back & helpful tech support.

I work at VFX houses in LA, and I'm not really sold on this product for a number of reasons.- Cost: That's a LOT of money for not a lot of Cores. This has always been true of Quadro series cards, but why? leading to....- Support: Nvidia will tell you you're paying for the benefits of a 1) long product life cycle. (true, these cards have been around FOREVER, hence so few cores) 2) high quality support (didn't really ever find them helpful when we were having problems with our software Blue-screening these cards out) and 3) custom tailored drivers (again, you'd try to discuss the potential of a bug in their drivers and they just were not interested at all) So it's possible that they're telling the truth, but it seems like they're only interested in you if you're company name starts with Autode and ends with dobe. This was offensive given that we had like 75 of these cards and they were failing in product demos for our clients.- Reliability. In our Supermicro 8-12 core workstations, this was the most common component failure, company wide. Most VFX houses you'd think pop render nodes, but we were just always losing these cards, about once a month at the worst point. A huge time hole, given what a PIA returns and exchanges are.- Driver quality. This is a tough area to place blame. Our artists would suffer 3DS max and Maya crashes ALL THE TIME. Sometimes this is the artist's fault for cramming too much into a file, sometimes it's the 3D software... On the other hand, we did find a previous revision of the Nvidia driver that seemed to make the greatest number of our machines the most stable, most of the time. This would imply issues with the driver, in particular the latest driver. This would all be trial and error of course. Another huge time hole. If progression theory in development is to be believed, the safe assumption should be that the most recent non-beta driver release is the most stable, but this is not the case here. Updating can mean more crashing.- Let's take a step back and be fair here for a moment. I don't know what it would be like to run a VFX house on a gaggle of ever changing desktop gaming cards. For all I know it would be an even greater nightmare. For all I know, no maya or 3DS max project would open the same on any two computers. I could be overlooking a much darker and much more insane world, and that's basically their sales pitch. Nvidia will tell you to buy a company's worth of these cards to have consistency on the vague threat of "MORE" problems if you don't.I'd be curious to see that scenario play out with a neutral 3rd party that doesn't profit off the selling of crazy overpriced hardware.