The NVIDIA GTX 1070 specs are official
The numbers are in and they’re impressive. NVIDIA is throwing down a ridiculous amount of power per watt with the latest iteration of GPUs, and the GTX 1070 is no exception.
The GTX 1070 shares the same base GPU core as the beefier (and pricier) GTX 1080, but it’s a slightly “hobbled” version of it, running with 15 out of 20 stream processors enabled. This begs the question whether it will be possible to unlock those stream processors with custom firmware or not…
- 16 nm GP104 silicon “GP104-200-A1” GPU
- 1920 CUDA cores
- 15 out of 20 streaming multiprocessors enabled
- 120 TMUs
- 64 ROPs
- 256-bit GDDR5 memory, 8 GB standard memory amount
- Max GPU Boost frequency 1600 MHz
- 6.75 TFLOP/s single-precision floating point performance
- 150W TDP, single 8-pin PCIe power connector
- 3x DisplayPort 1.4, 1x HDMI 2.0b
- 2-way SLI with SLI HB bridge support
With a MSRP of $379 USD this card is set to be faster than the existing Titan X which is still commanding a price tag of over $1,000 USD, all while using far less power. If that isn’t a game changer, then I don’t know what is!
This places the GTX 1070 squarely in the enviable position of being the mainstream choice for most gamers, and likely positions it to be the best selling and thereby most lucrative card NVIDIA will release this hardware cycle.
This card is on my “want” list, and it should probably be on yours, too!
It officially arrives on June 10th, so get your pre-orders in now or wait like me and probably be up the creek trying to find one… ah well, the life of a late-early-adopter.