Nvidia release their first PCIe HBM2 powered GPU to the market

Nvidia has officially released their first PCIe HBM2 powered GPU, officially beating AMD to the market with the new memory standard.
The Nvidia P100  is Nvidia’s first HBM2 PCIe 3 GPU to be officially released, though this GPU will only come as a Nvidia Tesla model which is designed for workstation use and not gaming. This means that AMD will likely become the first company to release the first a gaming oriented GPU to use HBM2 memory.
Nvidia’s P100 GPU will also come with several different models, releasing with 16GB and 12GB variants that come with memory bandwidths of 732GB/s and 549 GB/s respectively. This means that there will be versions that use four 4GB HBM2 chips while there will be other versions with three active 4GB HBM2 memory chips.


With this new GPU Nvidia are showcasing how scalable their new P100 GPUs can be when used with NV-link, which allows systems to scale to up to 8 P100 GPUs in a single system.

Sadly this new PCIe Tesla P100 will be limited to PCIe bandwidth, which will limit the number of GPUs that can be used on a single server and the bandwidth that is available to them for inter-GPU communications.