NVIDIA to Unveil "Pascal" at the 2016 Computex

[btarunr]

NVIDIA is reportedly planning to unveil its next-generation GeForce GTX "Pascal" GPUs at the 2016 Computex show, in Taipei, scheduled for early-June. This unveiling doesn't necessarily mean market availability. SweClockers reports that problems, particularly related to NVIDIA supplier TSMC getting its 16 nm FinFET node up to speed, especially following the recent Taiwan earthquake, could delay market available to late- or even post-Summer. It remains to be seen if the "Pascal" architecture debuts as an all-mighty "GP100" chip, or a smaller, performance-segment "GP104" that will be peddled as enthusiast-segment over being faster than the current big-chip, the GM200. NVIDIA's next generation GeForce nomenclature will also be particularly interesting to look out for, given that the current lineup is already at the GTX 900 series.



View at TechPowerUp Main Site

 

[ZoneDymo]

GTX Olympus

 

[RejZoR]

When is Computex "airing"? Pascal is interesting me. A lot.

 

[Frick]

When is Polaris due?

 

[the54thvoid]

News elsewhere suggested paper launch from Nvidia in early April and first release may actually be mobility parts. Strong chance is we might not see top tier desktop till Q4.

 

[FordGT90Concept]

When is Polaris due?

Before September.

 

[RejZoR]

Wasn't GTX 900 series launched in September/October timeframe? Could be around the same...

 

[AsRock]

When is Computex "airing"? Pascal is interesting me. A lot.

Think we all getting eager AMD or nVidia side tbh, even more with the nm shrink and all.

 

[medi01]

For Polaris, AMD said "mid 2016", whatever that means.

PS
Heh, and it will be Samsung 14nm (AMD) vs TSMC 16nm (NV) and what I've heard so far, at least for Apple's chips, TSMC was superior, despite being "bigger".

 

[rtwjunkie]

This seems to go right along with what I've been saying for 6 months: don't expect Pascal before the last few months of this year. Which means, likely just like with Maxwell, the more affordable mainstream cards will not be released until Jan/Feb, just like 960.

It's still a long time to wait for anyone that feels they need an upgrade right now.

 

[AsRock]

That make me even more happy, means i get more value out of my 290X already had it over 2 years now, wow don't time fly sheesh.

 

[EarthDog]

Wasn't GTX 900 series launched in September/October timeframe? Could be around the same...

970 and 980 were 9/2014. 980ti was 6/2015...

 

[medi01]

970 and 980 were 9/2014. 980ti was 6/2015...

980Ti was a bump up from 980 to spoil Fury launch.

PS
Yeah, or knockdown from Titan, but the point is, anti-Fury move.

 

[rtwjunkie]

980Ti was a bump up from 980 to spoil Fury launch.

By bump up, you mean the release date? That is correct. If you mean a bump up of the 980 card, that is incorrect. 980Ti is cut down Titan, not bumped up 980. 980Ti is a GM200 chip, 980 is a GM204.

 

[FordGT90Concept]

It's still a long time to wait for anyone that feels they need an upgrade right now.

But the jump in performance is huge and comes at lower power draw. IMHO, it's worth waiting. I already plan on selling my R9 390 and getting a Polaris card for those very reasons.

 

[nickbaldwin86]

2018 yet..?.. I want Volta.

 

[PP Mguire]

So we won't see Volta until late 2018 even though before it was scheduled for 2017. Highly disappointing. Guess I'm getting water blocks for Quakecon after all.

 

[RejZoR]

Volta is just a tiny bump from Pascal. Pascal however will be a huge bump to the Maxwell 2. My GTX 980 is fast and all, but I somehow miss AMD. I guess having several generations of AMD's finest Radeons left a mark on me. If only they weren't so god damn late with Fury cards, I'd probably be rocking one today...

 

[PP Mguire]

Volta is just a tiny bump from Pascal. Pascal however will be a huge bump to the Maxwell 2. My GTX 980 is fast and all, but I somehow miss AMD. I guess having several generations of AMD's finest Radeons left a mark on me. If only they weren't so god damn late with Fury cards, I'd probably be rocking one today...

Volta will be refined with less power draw, but the fact here remains I'll be on 4k uncapable GPUs for 2 years before I get an upgrade. This rustles my jimmies so bad and I hope AMD comes out with something this year that kicks ass.

 

[FordGT90Concept]

Polaris...does...kick...ass...it will probably debut before Pascal too.

 

[rruff]

Wasn't GTX 900 series launched in September/October timeframe? Could be around the same...

Maxwell was was launched in Jan-Feb of that year with the 750. If Nvidia is focusing on mobility, the first desktop cards may be the 750 replacements.

 

[PP Mguire]

Polaris...does...kick...ass...it will probably debut before Pascal too.

Yea, in AMD figures the power consumption does, but I'm talking actual performance. I really don't want to wait until 2017 for a GPU upgrade.

 

[rtwjunkie]

the first desktop cards may be the 750 replacements.

That role has been laid out for the 950SE. Since they announced it so late in Maxwell's life as the 750 replacement, my guess is it will fill that slot long into the Pascal cycle.

 

[TheGuruStud]

For Polaris, AMD said "mid 2016", whatever that means.

PS
Heh, and it will be Samsung 14nm (AMD) vs TSMC 16nm (NV) and what I've heard so far, at least for Apple's chips, TSMC was superior, despite being "bigger".

It wasn't much of a difference and TSMC is notoriously crap on yields. A tiny arm chip is simple compared to a monolithic GPU. This is where TSMC always falters. Plus, Sammy has no doubt been refining their process.

 

[HumanSmoke]

It wasn't much of a difference and TSMC is notoriously crap on yields. A tiny arm chip is simple compared to a monolithic GPU. This is where TSMC always falters. Plus, Sammy has no doubt been refining their process.

Where do you come up with this stuff? Xilinx is already shipping it's Zynq Ultrascale+ MPSoC's made on TSMC's 16nm FF+ process. Considering it ships up to 1506 pin package (comparable to an upper mainstream GPU or performance APU) I don't think it qualifies as "a tiny arm chip"

Spoiler: Package Dimensions

If Samsung are supposedly so far ahead, and I haven't seen any definitive proof that they are - either with yields or process ( 16nmFF vs 14nmLPE was definitely a TSMC win), it makes you wonder why TSMC secured over two-thirds of Apple's A9 business, and are looking increasingly likely to be sole supplier of the A10.

If a past history of bad yields is an indication of things going forward, AMD start ordering in industrial quantites of Xanax for the Zen ramp given Globalfoundries abysmal past record