Be Quiet! Pure Power 12 M - 850W ATX 3.0 PSU review
Corsair H170i Elite Capellix XT review
Forspoken: PC performance graphics benchmarks
ASRock Z790 Taichi review
The Callisto Protocol: PC graphics benchmarks
G.Skill TridentZ 5 RGB 6800 MHz CL34 DDR5 review
Be Quiet! Dark Power 13 - 1000W PSU Review
Palit GeForce RTX 4080 GamingPRO OC review
Core i9 13900K DDR5 7200 MHz (+memory scaling) review
Seasonic Prime Titanium TX-1300 (1300W PSU) review
NVIDIA partners halt GeForce GTX 970, 980, 980Ti production and GPU Pascal codenames surface
In anticipation of upcoming Pascal products in the coming month or two, at least that's what hwbattle is reporting this morning. As we get closer to the actual launch more and more tiny little details surface, combined bringing a bigger picture.
Can you smell it in the air ? :sniff::sniff: - yeah that's the smell of 16nm GPUs coming closer and closer.
It looks like that Nvidia will indeed name the upcoming GTX series the GeForce GXT 1070, 1080 and as it seems now perhaps even a 1080 Ti. The GPU naming now surfaced as well. Logic however would assume a Ti release later in the year opposed to everything being released at once in the summer.
- Equivalent to GTX 980 Ti will use a GP 104-400 GPU
- Equivalent to GTX 980 will use a GP 104-200 GPU
- Equivalent to GTX 970 will use a GP 104-150 GPU
As you can see that still leaves open GP100 (big pascal) for later in the year with likely a Titan based naming, ...
| Maxwell | Pascal | Board | Date | |
| NVIDIA Geforce GTX 980Ti | >> | GP 104-400 | Reference -> AIB | early June |
| NVIDIA Geforce GTX 980 | >> | GP 104-200 | Reference -> AIB | early June |
| NVIDIA Geforce GTX 970 | >> | GP 104-150 | AIB design | mid June |
Nvidia Pascal Consumer card announced during Computex - 04/11/2016 02:48 PM
But they will launch later ... Nvidia is to announce its consumer grade Pascal graphics cards at Computex 2016 from May 31-June 4, graphics card manufacturers include Asus, Gigabyte and MSI. They wou...
NVIDIA Pascal GP104 Die Photo - 04/11/2016 10:39 AM
Last week Nvidia announced the Tesla P100 data-center GPU and if you looked closely at the photo's you could already clearly see that big Pascal 15B transistor GPU being used. A new photo this time...
Judge rules that Samsung and Qualcomm do not violate Nvidia patents - 10/12/2015 09:10 AM
According to a judge ruling from the International Trade Commission, Samsung and Qualcomm do not violate Nvidia patents. Nvidia released this news themselves on their blog. Nvidia will appeal this r...
AMD and Nvidia prep for next-gen DirectX 12 - 03/24/2015 11:19 AM
Several new sets of slides from AMD and Nvidia as well surfaced on the webs through slideshare, the decks show what the companies have been up to in regards to DirectX 12, and they do look pretty exc...
Nvidia PhysX Source Code Now Available Free - 03/06/2015 09:03 AM
NVIDIA today put more than a decade of research, development and investment in gaming physics into the hands of game developers – by offering free source code for NVIDIA PhysX on GitHub. Thi...
sverek
Senior Member
Posts: 6070
Joined: 2011-01-02
Senior Member
Posts: 6070
Joined: 2011-01-02
#5257338 Posted on: 04/13/2016 09:04 AM
If they already halt production, does it mean Pascal will perform way better than current arch that it won't make sense to have those anymore or its regular routine?
If they already halt production, does it mean Pascal will perform way better than current arch that it won't make sense to have those anymore or its regular routine?
fantaskarsef
Senior Member
Posts: 14169
Joined: 2014-07-21
Senior Member
Posts: 14169
Joined: 2014-07-21
#5257343 Posted on: 04/13/2016 09:22 AM
Somehow I'm not sure the Ti will be based on the GP104. Because it would make the '80' card a cut down chip... does that work with production costs?
Just as a guess, the GP100 ('Titan') has 3840 shader processors, I guess the GP104-400 (biggest one, 'x80 Ti') will have something close to ~3000 shaders. GP104-200 ('x80') will be cut down to ~2500 shaders then? But does it make much sense to have the 'x80' as a cut down chip? It would mean that even the GP104-150 ('x70') would be a full GP104 chip in production, then laster cut? Wouldn't that mean that it unessecarily increased the 'x80's / 'x70's chip production cost?
I thought the 980Ti would only make sense because it was a cut down TitanX, making some use of faulty chips they can't put onto the top tier gaming cards / low rank professional cards.
Or am I seeing it all wrong, or doesn't it make much difference in production costs?
Somehow I'm not sure the Ti will be based on the GP104. Because it would make the '80' card a cut down chip... does that work with production costs?
Just as a guess, the GP100 ('Titan') has 3840 shader processors, I guess the GP104-400 (biggest one, 'x80 Ti') will have something close to ~3000 shaders. GP104-200 ('x80') will be cut down to ~2500 shaders then? But does it make much sense to have the 'x80' as a cut down chip? It would mean that even the GP104-150 ('x70') would be a full GP104 chip in production, then laster cut? Wouldn't that mean that it unessecarily increased the 'x80's / 'x70's chip production cost?
I thought the 980Ti would only make sense because it was a cut down TitanX, making some use of faulty chips they can't put onto the top tier gaming cards / low rank professional cards.
Or am I seeing it all wrong, or doesn't it make much difference in production costs?
Koniakki
Senior Member
Posts: 2843
Joined: 2009-09-15
Senior Member
Posts: 2843
Joined: 2009-09-15
#5257347 Posted on: 04/13/2016 09:28 AM
I seems the time for having a Ti which was based on a Titan with a few "compromises" comes to an end. I hope I'm wrong.
I would have love a 1080Ti based on Titan again with a few things cut down and 12GB HBM2. Wishful thinking I guess.
We will know soon if the performance is there and it justifies their decision. I would have like some GP100.
Btw, this is all based or speculations/rumors so far obviously.
This. Also we can assume if its based on GP104 then no HBM2 for the Ti.
I seems the time for having a Ti which was based on a Titan with a few "compromises" comes to an end. I hope I'm wrong.
I would have love a 1080Ti based on Titan again with a few things cut down and 12GB HBM2. Wishful thinking I guess.
We will know soon if the performance is there and it justifies their decision. I would have like some GP100.
Btw, this is all based or speculations/rumors so far obviously.
...I thought the 980Ti would only make sense because it was a cut down TitanX, making some use of faulty chips they can't put onto the top tier gaming cards / low rank professional cards....
This. Also we can assume if its based on GP104 then no HBM2 for the Ti.
k3vst3r
Senior Member
Posts: 3675
Joined: 2009-01-03
Senior Member
Posts: 3675
Joined: 2009-01-03
#5257349 Posted on: 04/13/2016 09:34 AM
Equivalent to GTX 980 Ti will use a GP 104-400 GPU should be 1080
Equivalent to GTX 980 will use a GP 104-200 GPU should be 1070
Equivalent to GTX 970 will use a GP 104-150 GPU should be 1060
GP 104 won't be using HBM2
GP 100 probably 6 to 12 month away yet, with plenty of HBM2 ready for production runs.
GP 100 is probably the Ti and Titan naming range.
GP 100 is 610mm so yield rates on brand new process easy talking 6 month minimum from now before they can get yields to an acceptable production rate.
Equivalent to GTX 980 Ti will use a GP 104-400 GPU should be 1080
Equivalent to GTX 980 will use a GP 104-200 GPU should be 1070
Equivalent to GTX 970 will use a GP 104-150 GPU should be 1060
GP 104 won't be using HBM2
GP 100 probably 6 to 12 month away yet, with plenty of HBM2 ready for production runs.
GP 100 is probably the Ti and Titan naming range.
GP 100 is 610mm so yield rates on brand new process easy talking 6 month minimum from now before they can get yields to an acceptable production rate.
Click here to post a comment for this news story on the message forum.
Senior Member
Posts: 586
Joined: 2008-06-20
I can't believe we are finally seeing the last days of 28nm GPUs! Hurray!
(I am not saying they are bad, not at all, but we have been stuck with 28nm for too long!)