Nvidia officially announces Quadro M6000

Based on the same GM200 Maxwell GPU

Following the launch of the new GTX Titan X, based on new GM200 Maxwell GPU, Nvidia has now announced the new professional graphics card based on the same GPU, the Quadro M6000.

This is actually the first time that Nvidia launches a Quadro graphics card so close to the launch of the Geforce counterpart, providing professional users much earlier access to now the fastest Quadro graphics card, the Quadro M6000. As it is based on the same GPU as the GTX Titan X, the Maxwell GM200, the Quadro M6000 packs 3072 CUDA cores, has a maximum GPU Boost clock of 1140MHz and push 7 TFLOPs of single precision compute performance. When compared to the previously available Kepler-based K6000, the new Quadro M6000 offers twice the pixel throughput and also brings quite a few new features.




Both the previously available Quadro K6000 and the new Quadro M6000 feature 12GB of GDDR5 memory paired up with a 384-bit memory interface, but with higher 6.6GHz clock on the M6000, this one offers higher 317GB/s memory bandwidth. The GM200 also brought GPU Boost to the Quadro lineup, something that should significantly help with performance in certain scenarios.

While we have significantly more CUDA cores and all other features coming from Maxwell GPU, the Quadro M6000 has a TDP of 250W, just 25W over the previously available Quadro K6000. Despite the raise TDP, the Quadro M6000 still needs a single 8-pin PCI-Express power connector. The performance gain should be roughly somewhere around 50 percent depending on the scenario.

The new Quadro M6000 is also capable of driving four 4K displays, while previous Quadro graphics cards were limited to two, all thanks to the updated display controller. It also comes with the latest NVENC video encoder.




Nvidia gave a couple of new Quadro M6000 graphics card to Lucasfilm, which, according to Lutz Latta, Principal Engineer at Lucasfilm "saw overall gains of 55% in heavy a compute and memory access ray-tracing application using layered shadow maps."

While it has a rather limited double precision compute performance due to its native FP64 rate of 1/32 FP32, Nvidia claims that this will not be a problem and that Quadro M6000 will definitely outperform its predecessors.

According to Nvidia, the new Quadro M6000 flagship should be available soon in both complete systems and as individual graphics cards. Unfortunately, Nvidia did not reveal any details regarding the price but expect it to go over US $5000, just as it was the case with previous Quadro flagship graphics cards.



Source: Nvidia.com.


News by Luca Rocchi and Marc Büchel - German Translation by Paul Görnhardt - Italian Translation by Francesco Daghini


Previous article - Next article
comments powered by Disqus
Nvidia officially announces Quadro M6000 - NVIDIA - News - ocaholic