Watch NVIDIA’s Computex 2017 Keynote Livestream Here
Rumors emerged last week that NVIDIA would be debuting a new enthusiast Volta Titan graphics card at computex this week after photos of the alleged new card leaked out. The company officially announced today via twitter that it’s going to be holding a keynote headlined by President & CEO Jensen Huang at 8:30 PM PST 29th of May.
Related NVIDIA GTX 1170 Alleged Benchmark Leaked, Faster Than 1080 Ti
NVIDIA’s Mysterious Volta Titan Enthusiast Graphics Card
The photograph above was allegedly leaked by an NVIDIA intern last week. It shows a black graphics card with gold accents and “TITAN” engraved on the side. The card has 8+6pin PCIE power connectors and notably features an NVLNK interface, something unusual for a TItan graphics card and is only available on NVIDIA’s professional Tesla and Quadro accelerators.
The photographed “TITAN” card has the same aesthetic look and color scheme of NVIDIA’s recently announced Volta based Tesla accelerators. This has led many to believe that this card will likely join NVIDIA’s new roster of Volta products as a workstation and/or enthusiast gaming card. According to the rumors surrounding this mysterious TITAN card NVIDIA is expected to introduce it at computex.
NVIDIA’s graphics card fitted with NVLINK interfaces have so far all featured HBM2 rather than GDDR memory. This in addition to the fact that GDDR6, which is set to power gaming oriented Volta GeForce graphics cards, isn’t currently available and won’t be until next year indicates that this card features HBM2. Whether this card will in fact make its debut tomorrow or not remains to be seen, we’ll know soon enough. In the mean time take these rumors with a grain of salt.
| Tesla Product | Tesla K40 | Tesla M40 | Tesla P100 | Tesla V100 |
|---|---|---|---|---|
| GPU | GK110 (Kepler) | GM200 (Maxwell) | GP100 (Pascal) | GV100 (Volta) |
| SMs | 15 | 24 | 56 | 80 |
| TPCs | 15 | 24 | 28 | 40 |
| FP32 Cores / SM | 192 | 128 | 64 | 64 |
| FP32 Cores / GPU | 2880 | 3072 | 3584 | 5120 |
| FP64 Cores / SM | 64 | 4 | 32 | 32 |
| FP64 Cores / GPU | 960 | 96 | 1792 | 2560 |
| Tensor Cores / SM | NA | NA | NA | 8 |
| Tensor Cores / GPU | NA | NA | NA | 640 |
| GPU Boost Clock | 810/875 MHz | 1114 MHz | 1480 MHz | 1455 MHz |
| Peak FP32 TFLOP/s* | 5.04 | 6.8 | 10.6 | 15 |
| Peak FP64 TFLOP/s* | 1.68 | 2.1 | 5.3 | 7.5 |
| Peak Tensor Core TFLOP/s* | NA | NA | NA | 120 |
| Texture Units | 240 | 192 | 224 | 320 |
| Memory Interface | 384-bit GDDR5 | 384-bit GDDR5 | 4096-bit HBM2 | 4096-bit HBM2 |
| Memory Size | Up to 12 GB | Up to 24 GB | 16 GB | 16 GB |
| L2 Cache Size | 1536 KB | 3072 KB | 4096 KB | 6144 KB |
| Shared Memory Size / SM | 16 KB/32 KB/48 KB | 96 KB | 64 KB | Configurable up to 96 KB |
| Register File Size / SM | 256 KB | 256 KB | 256 KB | 256KB |
| Register File Size / GPU | 3840 KB | 6144 KB | 14336 KB | 20480 KB |
| TDP | 235 Watts | 250 Watts | 300 Watts | 300 Watts |
| Transistors | 7.1 billion | 8 billion | 15.3 billion | 21.1 billion |
| GPU Die Size | 551 mm² | 601 mm² | 610 mm² | 815 mm² |
| Manufacturing Process | 28 nm | 28 nm | 16 nm FinFET+ | 12 nm FFN |
