Fractal Design Pop Air RGB Black TG review
Palit GeForce GTX 1630 4GB Dual review
FSP Dagger Pro (850W PSU) review
Razer Leviathan V2 gaming soundbar review
Guru3D NVMe Thermal Test - the heatsink vs. performance
EnGenius ECW220S 2x2 Cloud Access Point review
Alphacool Eisbaer Aurora HPE 360 LCS cooler review
Noctua NH-D12L CPU Cooler Review
Silicon Power XPOWER XS70 1TB NVMe SSD Review
Hyte Y60 chassis review
Nvidia will release mainstream GP106 GPUs later this year
According to several sources the equivalent of the GTX 950 and GTX 960 (we don't know the names yet but let's call them GTX 1050 and 1060 for now) would be released later in the year, an Autumn release meaning the Christmas season.
The chip empowering the product will be a 16nm fabbed Pascal under the logical code-name GP106. The GP106 is used for price competitive and sensitive products that hit a good price/perf ratio.
If we chart things up (and the chart below ids very speculative) we see the following. BTW the GP104-400 for the Ti might very well become a GP100 based SKU, which would make more sense
NVIDIA GeForce | Possible name |
Pascal | Fab | Board | Date | |
GTX Titan X | >> | TBD | GP 100 | 16nm | Reference -> AIB | TBD |
GTX 980Ti | >> | GTX 1080 Ti | GP 104-400 | 16nm | Reference -> AIB | June/July |
GTX 980 | >> | GTX 1080 | GP 104-200 | 16nm | Reference -> AIB | June |
GTX 970 | >> | GTX 1070 | GP 104-150 | 16nm | AIB design | June |
GTX 960 | >> | GTX 1060 | GP 106 | 16nm | AIB design | October |
GTX 950 | >> | GTX 1050 | GP 106 | 16nm | AIB design | October |
So from the looks of it this summer will be all about the successors to the Geforce GTX 970, GTX 980 and GTX 980 Ti.
« Mafia 3 released on 7th of October 2016 · Nvidia will release mainstream GP106 GPUs later this year
· Lian Li Launches 1499 USD DK-04 Desk-type Case »
NVIDIA Will Fully Implement Async Compute Via Driver Support - 09/05/2015 07:07 PM
And they've got Oxide from Ashes of Singularity to confirm that. Oxide’s developer “Kollock” wrote that NVIDIA has not fully implemented yet Async Compute in its driver...
Nvidia Wanted Oxide dev DX12 benchmark to disable certain DX12 Features ? (content updated) - 08/31/2015 12:37 PM
An intersting read was posted by a spokes person of Oxide, basically their DX12 benchmark of the Beta Ashes of the Singularity runs a notch better at specific AMD graphics cards. Is seems Nvidia is ...
Nvidia Wins First round in Battle with Samsung - 04/08/2015 08:42 AM
NVIDIA received a favorable ruling from ITC in the Patent Dispute with Samsung and Qualcomm. The judge presiding over the patent case against Samsung and Qualcomm in the U.S. International Trade Comm...
NVIDIA wanted to show kepler at CES but didn't - 01/13/2012 01:30 PM
With AMD
NVIDIA warns about counterfeit GeForce GPUs - 09/01/2011 10:10 AM
NVIDIA send out a warning to only buy graphics cards from authorized partners as the company has the discovered that substandard GeForce-branded products are being sold on the gray market: "Rese...
Ieldra
Senior Member
Posts: 3490
Joined: 2007-01-27
Senior Member
Posts: 3490
Joined: 2007-01-27
#5261037 Posted on: 04/21/2016 02:31 AM
What do you mean by 2x FP16?
From what I saw this is only Tesla specific, I kind of doubt geforce benefits from such stuff..
that's why i'm waiting
What do you mean by 2x FP16?
From what I saw this is only Tesla specific, I kind of doubt geforce benefits from such stuff..
that's why i'm waiting
abula
Senior Member
Posts: 845
Joined: 2002-01-24
Senior Member
Posts: 845
Joined: 2002-01-24
#5261044 Posted on: 04/21/2016 03:20 AM
I dont think we will see the GTX1080Ti soon, i think the release will be very similar to the GTX780 and GTX980, where their Ti counter parts where released 6-8 months later. My guess is
GTX1080/1070 = June 2016 (Announced on Computex)
GTX1060/1050 = December 2016
GTX1080Ti = February 2017 (probably announced in CES)
I dont think we will see the GTX1080Ti soon, i think the release will be very similar to the GTX780 and GTX980, where their Ti counter parts where released 6-8 months later. My guess is
GTX1080/1070 = June 2016 (Announced on Computex)
GTX1060/1050 = December 2016
GTX1080Ti = February 2017 (probably announced in CES)
ivymike10mt
Senior Member
Posts: 226
Joined: 2015-01-28
Senior Member
Posts: 226
Joined: 2015-01-28
#5263291 Posted on: 04/26/2016 03:54 PM
I dont think we will see the GTX1080Ti soon, i think the release will be very similar to the GTX780 and GTX980, where their Ti counter parts where released 6-8 months later. My guess is
GTX1080/1070 = June 2016 (Announced on Computex)
GTX1060/1050 = December 2016
GTX1080Ti = February 2017 (probably announced in CES)
Indeed..
Maxwell "relase date strategy" was good/profitable for Nvidia.
So they may use similar or same plan for a Pascal.
Also more harder "handle" many GPU lines in same time...
I dont think we will see the GTX1080Ti soon, i think the release will be very similar to the GTX780 and GTX980, where their Ti counter parts where released 6-8 months later. My guess is
GTX1080/1070 = June 2016 (Announced on Computex)
GTX1060/1050 = December 2016
GTX1080Ti = February 2017 (probably announced in CES)
Indeed..
Maxwell "relase date strategy" was good/profitable for Nvidia.
So they may use similar or same plan for a Pascal.
Also more harder "handle" many GPU lines in same time...
-Tj-
Senior Member
Posts: 17490
Joined: 2012-05-18
Senior Member
Posts: 17490
Joined: 2012-05-18
#5263381 Posted on: 04/26/2016 05:58 PM
^
That plan started with Kepler GK104 chip (680gtx), by Fermi GF104/114 it was too weak vs AMD high-end so they couldn't do such pricing..
Imo its crazy, but people will always buy it even if its overpriced..
I would say by nv its only top tier gpus that are worth and will last much longer. I mean so far any of these Gx104/114/204 chips showed its real place
^
That plan started with Kepler GK104 chip (680gtx), by Fermi GF104/114 it was too weak vs AMD high-end so they couldn't do such pricing..
Imo its crazy, but people will always buy it even if its overpriced..
I would say by nv its only top tier gpus that are worth and will last much longer. I mean so far any of these Gx104/114/204 chips showed its real place

Click here to post a comment for this news story on the message forum.
Senior Member
Posts: 17490
Joined: 2012-05-18
What do you mean by 2x FP16?
From what I saw this is only Tesla specific, I kind of doubt geforce benefits from such stuff..