GeForce GTX 1070 Ti Launch Date October 26th + Specs and Details
So we've shed a few words already on the upcoming GeForce GTX 1070 Ti. We've already confirmed the product through our industry sources, they will not say the final name just yet but we know for fact that the GPU is a GP104-300.
That GPU codename is a confirmed fact. The product also is referred to as "Ti" within the companies I spoke to. Earlier on my sources also confirmed a release end of October, which matches up with the Gigabyte teaser earlier on. Now I expected that to be October the 28th, but that being a saturday is a no-go as Nvidia traditionally released their products on a Tuesday or Thursday (check it .. a true fact). Nordic hardware now states it confirmed a launch date of October the 26th, which is ... a Thursday :) So I'd say that is a very solid assumption slah confirmed fact on their end.
Earlier on MyDrivers already posted information of specifications that might be spot on. The GPU has 20 Shader clusters, just one would be disabled totalling towards 2432, which is just 128 shader procs lower than that of the GTX 1080. The clock speeds of the GTX 1070 Ti would be higher than the GTX 1070 as well at 1607 MHz and a 1683 MHz GPU Boost woth 8.00 GHz (GDDR5-effective) memory clock. The TDP would be 180W, similar to the GTX 1080. If I perform some reverse math on this, the product would offer just over 8 TFLOPS in perf. Personally I think it'll be priced at 399 USD.
If we chart it all up, here we go & thanks to SH SOTN for this spot:
| ||||||
---|---|---|---|---|---|---|
Reference GeForce | Titan X | GTX 1080 | GTX 1070 Ti | GTX 1070 | GTX 1060 | GTX 980 Ti |
(2016 edition) | ||||||
GPU | GP102-400-A1 | GP104-400-A1 | GP104-300-A1 | GP104-200-A1 | GP106-400-A1 | GM200 |
Architecture | Pascal | Pascal | Pascal | Pascal | Pascal | Maxwell |
Transistor count | 12 Billion | 7.2 Billion | 7.2 Billion | 7.2 Billion | 4.4 Billion | 8 Billion |
Fabrication Node | TSMC 16 nm | TSMC 16 nm | TSMC 16 nm | TSMC 16 nm | TSMC 16 nm | TSMC 28 nm |
CUDA Cores | 3,584 | 2,560 | 2,432 | 1,920 | 1,280 | 2,816 |
SMMs / SMXs | 28 | 20 | 19 | 15 | 10 | 22 |
ROPs | 96 | 64 | 64 | 64 | 48 | 96 |
GPU Clock Core | 1,417 MHz | 1,607 MHz | 1,607 MHz | 1,506 MHz | 1,506 MHz | 1,002 MHz |
GPU Boost clock | 1,531 MHz | 1,733 MHz | 1,683 MHz | 1,683 MHz | 1,709 MHz | 1,076 MHz |
Memory Clock | 2,500 MHz | 2,500 MHz | 2,000 MHz | 2,000 MHz | 2,000 MHz | 1,753 MHz |
Memory Size | 12 GB | 8 GB | 8 GB | 8 GB | 3 GB / 6 GB | 6 GB |
Memory Bus | 384-bit | 256-bit | 256-bit | 256-bit | 192-bit | 384-bit |
Memory Bandwidth | 480 GB/s | 320 GB/s | 256 GB/s | 256 GB/s | 192 GB/s | 337 GB/s |
FP Performance | 11.0 TFLOPS | 9.0 TFLOPS | 8.17 TFLOPS | 6.45 TFLOPS | 4.61 TFLOPS | 6.4 TFLOPS |
GPU Thermal Threshold | 94 Degrees C | 94 Degrees C | 94 Degrees C | 94 Degrees C | 94 Degrees C | 91 Degrees C |
TDP | 250 Watts | 180 Watts | 180 Watts | 150 Watts | 120 Watts | 250 Watts |
Launch MSRP ref | $1200 | $499 | $399 | $379/$449 | $249/$299 | $699 |
Mind you that the specs and details mentioned are still based on speculation mainly. But I have a hunch they are pretty precise or at the very least pretty darn close.
Expert Oriented GeForce GTX 1060 OC Graphics Card - 09/25/2017 08:05 AM
Asia based Expert Oriented announced a new GeForce GTX 1060 OC in its line-up. Codenamed GF-GTX1060-E6GB/OC2/DF, this VR-ready graphics card is packed with 1280 CUDA Cores, a 192-bit memory interface,...
New Model MSI GeForce GTX 1080 Ti TRIO Gaming X with Triple Fans Spotted - 09/22/2017 09:13 AM
MSi will be releasing a new GeForce GTX 1080 Ti, the new model might be aimed at the Asia market solely. Photos appeared by Hermitage Akibahara showing a sizable card that from the looks of it is ev...
EVGA Launches GeForce GTX 1080 Ti FTW3 ELITE Edition with 12Gbit/s GDDR5X - 09/22/2017 08:07 AM
EVGA's GeForce GTX 1080 Ti FTW3 ELITE cards are available, these have 12 GHz of GDDR5X memory, giving it 528 GB/s of memory bandwidth. These cards are available with either the ELITE Black or White...
GeForce GTX 1070 Ti Specs and Details - Based on GP104-300 - 09/22/2017 08:03 AM
More specs of the upcoming GeForce GTX 1070 Ti are surfacing online. You have seen the rumors and teaser yesterday already. This post will shed some more information. ...
Download: GeForce 385.69 WHQL driver - 09/21/2017 02:30 PM
Download the Nvidia GeForce GameReady 385.69 WHQL driver as released by NVIDIA. This driver adds support for Project Cars 2, Call of Duty: WWII open beta, Total War: WARHAMMER II, Forza Motorsport 7,...
Senior Member
Posts: 9786
Joined: 2011-09-21
good move for nvidia.
by striking at the consumer "sweet spot" msrp, they extend their time on the clock working out the bugs on the next gen, whether that's volta or not. and, not coincidentally, doing a bit of a screw you to amd.
pascal's here for a bit boys.
nvidia can afford to buy time by doing a pascal refresh but it will have to deliver an answer before navi - that is the point the technology scale tilts to amd. nvidia really has to have their scalability game on point by that time.
I just don't get where this confidence in AMD R&D in their graphics department comes from. They need to prove they can deliver something before I will expect anything from them.
Senior Member
Posts: 2397
Joined: 2003-12-15
Why?
What other options are there?
Certainly not vega.
I love that everything has to be about some stupid brand war on G3D these days

No it's a bad purchase because it's a near end of life product, a year and a half ago, wonderful. And it's not like we're talking about yet another titan for the no sense, money doesn't matter "whales" that'll just buy whatever as long as its the fastest thing when they're buying it. Anyway, reminds me of my best pal who kept asking for advice for GPU, and he bought a 970 even tho I kept telling him to wait. Two months before the 1xxx cards came out, his wife tells me he's still mad about wasting that money on a 970.
Senior Member
Posts: 11396
Joined: 2010-12-27
I love that everything has to be about some stupid brand war on G3D these days

No it's a bad purchase because it's a near end of life product, a year and a half ago, wonderful. And it's not like we're talking about yet another titan for the no sense, money doesn't matter "whales" that'll just buy whatever as long as its the fastest thing when they're buying it. Anyway, reminds me of my best pal who kept asking for advice for GPU, and he bought a 970 even tho I kept telling him to wait. Two months before the 1xxx cards came out, his wife tells me he's still mad about wasting that money on a 970.
EoL? What basis did you make that assumption off of?
Volta is going to be out after GDDR6 mass production starts.
You're looking at Q2 2018 for an equivalent of GV104.
We aren't talking about a month or two, 9 months is a long time.
Are you going to tell new system builders to stick with IGP until volta release?
Senior Member
Posts: 1217
Joined: 2017-08-18
well i can see where you'd have that attitude.
what i'm talking about are the physical properties of the gpu die.
for example, volta is at the maximum size of any chip as it encompasses virtually the entire wafer...in order to actually sell them, they have to be sorted with certain sections disabled because of variations in yield...and those disabled sections are not consistent except for being at the periphery...which takes time and money and thus is not ready for primetime.
like ryzen, vega is scalable. instead of "zeppelin" modules, you'd have the equivalent.
meaning at 7nm, they could configure a gpu with multiple times the performance currently available at a far higher rate of yield, lowering costs, power, and temps with increased frequency.
that is navi.
that scares nvidia (justly) as they race to come up with an alternative.
Senior Member
Posts: 474
Joined: 2001-05-02
good move for nvidia.
by striking at the consumer "sweet spot" msrp, they extend their time on the clock working out the bugs on the next gen, whether that's volta or not. and, not coincidentally, doing a bit of a screw you to amd.
pascal's here for a bit boys.
nvidia can afford to buy time by doing a pascal refresh but it will have to deliver an answer before navi - that is the point the technology scale tilts to amd. nvidia really has to have their scalability game on point by that time.
Yup and my beat it is ever bit as fast Vega 56 if just hair bit faster but with out the jackup price