The Nvidia P6000 and P5000 see the Pascal platform making its way from gamer graphics to cards for 3D and VFX artists, and video editors.
There's a new king in the world of graphics cards, as Nvidia has used the Siggraph 2016 conference to launch its new flagship model, the Quadro P6000 (above) – alongside the moderately-less-powerful P5000. These high-end boards contrast with AMD's new low-cost approach with a focus on buildinf mid-range boards in the new Radeon Pro WX range including the Radio Pro WX 7100.
Nvidia's Quadro P6000 and P5000 graphics cards are the first pro-grade models to be based on the company's Pascal architecture, which follows the Maxwell platform used by last year's Quadro M6000. Read our Nvidia Quadro M6000 review.
Pascal boasts loads of nerdy-sounding improvements: a 16nm FinFET, better simultaneous multi-projection, dynamic load balancing and support for GDDR5X RAM. But what do these actually mean, and how do they make the new graphics cards more powerful?
The 16nm FinFET refers to the size of the transistors used to crunch the numbers behind both 3D scenes and other intensive tasks you might use your graphics card to do – like video effects, simulations or even some Photoshop filters. These are down from Maxwell's 20nm (or 0.000016mm down from 0.000020mm, if you prefer it in units you can actually see).
Smaller transistors means that Nvidia can pack more onto the board, so more number-crunching can be done at once – so you end up with a more powerful graphics card.
The P6000 has 3,840 CUDA cores – essentially mini-processors built from these transistors – up from the M6000's 3,072. The P5000 (below) has 2,560 cores, up from the M5000's 2,048.
Nvidia says that Pascal is twice as powerful at simultaneous multi-projection – ie outputting two video signals at once – than Maxwell. This is useful when you're outputting different images for each eye – eg for 3D cinema or VR.
Dynamic load balancing is what the boards use to decide how much of its power to each of the multiple tasks you might ask it to do – as the scene you're previewing or composite or edit you're looking at could include 3D elements (including a camera, lights, models and the like) and non-3D elements that the card is also processing (video, 2D effects, etc). The card's driver works out what needs the processing power most and directs it proportionally.
GDDR5X RAM can pass information in an out twice as fast as the GDDR5 used by Maxwell – avoiding bottlenecks that could slow down performance. The P6000 has 24GB of RAM, up from the M6000's 12GB. The P5000 has 16GB, up from the M5000's 8GB.
Both cards support the DisplayPort 1.4 output standard, which means they can output to screens at resolutions up to 7,680x4,320 (aka 33-megapixel or twice the resoltion of 5K screens like that offered by the Apple's top-of-the-line iMac).
Both Pascal-based Quadro graphics cards will be out in September.
New Iray plugins
Nvidia also used Siggraph to announce new versions of its Iray photorealistic rendering engine, which adds new features including generating renders specifically for virtual reality (including spherical panoramas and lightfields.
The new Iray plugin for Autodesk 3ds Max is out next week, with plugins for Maya and Cinema 4D "to follow".
Nvidia takes back Mental Ray for Maya
While Nvidia owns the Mental Ray rendering engine, in the past the Maya plugin for it has been provided by Autodesk. Nvidia is taking development, sales and support of/for this back in-house - essentially so it can develop it faster than before. A new Mental Ray for Maya plugin is due from Nvidia in September, with support for a faster, easier-to-use global illumination rendering system called GI-Next.