• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

GTX 1080Ti specs leaked

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Janus67

Benching Team Leader
Joined
May 29, 2005
apparently those numbers were posted to WCCFtech like a month ago
 

deathman20

High Speed Premium Senior
Joined
Aug 5, 2002
Mmmmmmmmmm tasty.
Maybe it will be time for me to just say forget it and pay the price for one of these nicer cards and be set for a few years.
 

Scout

Member
Joined
Sep 9, 2011
Fast but GDDR5 come on Nvidia we all see your earnings put that cash into something interesting aka HBM2.
 

deathman20

High Speed Premium Senior
Joined
Aug 5, 2002
Good point Scout. Totally forgot about HBM2, thought it would of been on one of these cards by now. Guess maybe they are waiting for the next chip to do that :(
 

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
Or..........maybe NVIDIA isn't reaching TOO far ahead into the future and only using what it needs. There are so few people on 4K and doing things with GPUs that require so much memory bandwidth, why not stick with GDDR5X? Besides, all HBM did for Fury X over 980Ti was close the gap a few % from 1080p to 4K.
 

deathman20

High Speed Premium Senior
Joined
Aug 5, 2002
True memory bandwidth isn't always everything but... just give us the good stuff :)
You'd think they'd potentially want to upgrade to HBM because it saves board space as well, which will give them even more money potentially. Depends on the offset costs going from GDDR to HBM which is more expensive.
 

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
You get the same thing with gddr5 as you do with hbm2. Again, this only really matters at high resolutions at this time. :)
 

JrClocker

AKA: JrMiyagi
Joined
Sep 25, 2015
Here's another article...not sure how trustworthy this source is:

http://wccftech.com/nvidia-gtx-1080-ti-launch-january/

7.1% less CUDA cores than Titan X Pascal, but an 8.0% higher boost clock. All Pascal seems to hit the same max overclock range (2,000 MHz or so...mine does 1950 MHz on air)...so it would perform slightly less than the Titan X Pascal, but at a much cheaper price.

But, the rumor also has that the board partners will be able to do their own thing.
 

Robert17

Premium Member
Joined
Feb 1, 2011
Location
Republic of Texas
Perhaps NVidia is simply checking all the boxes, making sure that they produce products at multi-performance levels, utilizing various components/drivers/firmware, filling the market with options, appealing to every possible market niche. Or not. Maybe they want to further the vertical manufacturing scheme to include HBM2 once they can supply their own. Or maybe that they are also playing with PCM related components and these toys are fun for now while R&D is busy with VR, driverless cars, and quantum pet food. Either way, we win.:cool:
 

EarthDog

Gulper Nozzle Co-Owner
Joined
Dec 15, 2008
Location
Buckeyes!
Here's another article...not sure how trustworthy this source is:

http://wccftech.com/nvidia-gtx-1080-ti-launch-january/

7.1% less CUDA cores than Titan X Pascal, but an 8.0% higher boost clock. All Pascal seems to hit the same max overclock range (2,000 MHz or so...mine does 1950 MHz on air)...so it would perform slightly less than the Titan X Pascal, but at a much cheaper price.

But, the rumor also has that the board partners will be able to do their own thing.
Rinse... repeat with these high end "Ti" cards. Nothing new. :)