[Video] Vega Radeon Instinct MI25 boasts 12.5TFLOPS / 25TFLOPS performance; AMD unveils Vega for professional use

Details for AMD's upcoming range of Vega powered graphics cards have emerged, with the top dog, the Radeon Instinct MI25 is outperforming Nvidia's Pascal powered Titan X graphics cards handily in certain applications, and crushing Maxwell without any issue.

The MI25 is a vega powered GPU that is having HBM and up to 12.5 TFLOPS of FP32 performance. In FP16, the MI25 puts out 25 TFLOPS. Possibly indicating that Vega will certainly offer improved FP16 support much like how Mark Cerny, the lead PS4 / PS4 Pro architect implied.

The first Vega based graphics processing unit from AMD has finally leaked out for the professional use, reports Videocardz. The MI25 Instinct is the Vega based flagship accelerator, which is targeted at deep learning applications.

The Instinct lineup is showing the MI6, MI8 and MI25 GPU-based accelerators from AMD. The classification of the scheme appears to denote the performance metric of each GPU, with the MI6 clocking in at 5.7 TFLOPs, the MI8 clocking in at 8.2 TFLOPs and while no value has been given for the MI25, using the same idea so that we can estimate it at around 25 TFLOPs, reports WccFtech.

Not many details are given about the Vega based accelerator except that it will have a TDP of less than 300W, 2x Packed Math.

The Vega based product could have a core count of anywhere from 4096 to 6250 with clocks ranging from 1000 mhz to 1600 mhz. Since this is a passively cooled affair, it will not expect to be clocked very high.

AMD has also included slides of pre-configured rack designs, he maximum compute of the Inventec Rack is 3 petaflops of FP16 compute. This number divided by the total amount of GPUs present inside means we are looking at exactly 25 TFLOPs of FP16 per GPU.

AMD is really focusing a lot of the Vega Radeon Instinct MI25 marketing around its FP16 performance that is not surprising given FP16 performance is not only higher but also FP16 is adequately accurate for deep-learning, machine learning and many other HPC tasks.

You can also watch the AMD Vega Radeon Instinct Performance video below.

Tags
Amd
Real Time Analytics