Is the GP102 Titan going to support the professional drivers? If not, then there have been no tables turned. This is just another Titan, without DP power, more suitable for gaming.
Is the GP102 Titan going to support the professional drivers? If not, then there have been no tables turned. This is just another Titan, without DP power, more suitable for gaming.Update today from Nvidia to Anandtech confirming GP102, die size of 471mm2, neutured FP64. So basically GP100 physically stripped of the HPC stuff like NVlink, FP64 etc etc. If I remember correctly, that is about the size of G80 on 90nm
Is the GP102 Titan going to support the professional drivers? If not, then there have been no tables turned. This is just another Titan, without DP power, more suitable for gaming.
I don't think there will be a 1080 Ti. Aftermarket GP104s already hit out of the box what you'd expect from a stock GP102-as-1080 Ti.
Next stop is Volta GV104, IMHO.
Rumours disagree with you. As per rumours, the next step is a GP100 based Titan with HBM2 in 2017, and obviously a Ti version would also follow. The only reason why Nvidia may even be considering that, is probably cause Volta may be later in 2018.
Rumours disagree with you. As per rumours, the next step is a GP100 based Titan with HBM2 in 2017, and obviously a Ti version would also follow. The only reason why Nvidia may even be considering that, is probably cause Volta may be later in 2018.
The worst part is, by the time we actually get a real, full blown Pascal chip, its life will be so short because at that point mid range Volta will only be a few months away, tarnishing that flagship pride that most buyers long for yet again.
The only place where I have seen this "rumor" is WCCFTech, and they made it up, can guarantee you that.
GP102 was designed separately from GP100 because GP100 isn't a gaming-focused architecture.
30% larger die, more expensive HBM2, only for a memory bandwidth gain? I mean the margins on Titan brand are massive, but might as well price the GP102 there anyway and get even more profit.
Unless they do both. $1500 GP102 Titan Black, $2000 GP100 Titan YOLO.
Pascal has been out for only a couple months and we only have the mid range chips. Pascal is going to be milked to absolute death. We will likely see 4 or 5 flagships before its all said and done. In just a few months we have had 1 flagship with a flagship price, and another one is being released on the 2nd of August. They will all be expensive as hell and none of them will have flagship status longevity because the milking process requires a new flagship every several months or so.
The worst part is, by the time we actually get a real, full blown Pascal chip, its life will be so short because at that point mid range Volta will only be a few months away, tarnishing that flagship pride that most buyers long for yet again.
Then your only salvation is to give up PC gaming. Seriously, if these things irk you this deeply, why would you even consider bothering with them anymore?
I mean, isn't there an option for you? Is there no GPU out there that you would be happy with? The profgress of chips is pretty fast, so of course the last one in any series will be the shortest life. Naturally.
What,you want to bet me on it too? You can eat that Nvidia flavored catfood when they release a GP100 Geforce card.
What,you want to bet me on it too? You can eat that Nvidia flavored catfood when they release a GP100 Geforce card.
I'll bet you a copy of Call of Duty 2017 that GP100 never gets released as a GeForce card.
It seals the deal for me; I'll either get a used GTX 980 or a 4gb RX480 now and hold off until the refreshes hit or Volta/Navi comes out. The prospect of GP104 with 12ghz G5X sounds interesting.... but the prices would need to be radically adjusted from what they are now. And who knows, maybe Polaris truly was the stop gap and Vega will be the actual leap forward we were hoping AMD would deliver. (probably not, but here's to hoping).
Also why did you sell your 980?Anything wrong with it?I won't spend more than $600 next gen, and I won't buy unless I get stock performance 2x better perf than the 980 at 980's TDP.
I will go with the first company that can match all 3 criteria.
Do you remember the price of a 6800GT? I sure do. Bought one for myself at the first day of availability at COMPUSA. 399.00. That was a 2nd tier card. Thee was the ULTRA above it. Somewhere above 500 bucks wasn't it?
At any rate, GTX 680 was 499 at launch. 780Ti was how much?
Look, you can complain about pricing all the live long day, but in the end, there will be a product that suits your price and performance. I don't see a problem. It all depends on what you wish to spend and what you're happy with. Being unhappy because one cannot afford the highest tier is a waste of time. Otherwise, everyone would be driving mercedes when a kia would do.
Just buy what you want to at a price you're able to live with. That's all there is too it. Mourning over new pricing is a waste of time when there isn't anything anyone can do about it EXCEPT the actual act of mourning. Gets nothing done.
There's no logic for GP100 as a gaming card. FP64 1:2 ratio means half the CUDA CORES on the chip are useless in games. All they do is consume power. Tesla P100 is a 300W beast, running a cut down chip. Think about that for a bit and understand that there's no advantage for GP100 vs GP102 for gaming.
Zilch. Same effective FP32 performance, enough bandwidth with memory compression.
To put GP100 as a gaming card would be a downgrade from GP102 as you get significantly worse perf/w and added cost of HBM2 for no gains.
30% larger die, more expensive HBM2, only for a memory bandwidth gain?
HBM uses less power (around half as much as GDDR5) as well which leaves you with extra juice to throw at the GPU. It also takes up a lot less area, which makes for overall physically smaller cards like the Fury Nano a possibility.
I'm excited to see NV adopting it as it does have implications beyond just having more memory bandwidth, which even itself isn't a bad thing.
There's no logic for GP100 as a gaming card. FP64 1:2 ratio means half the CUDA CORES on the chip are useless in games. All they do is consume power. Tesla P100 is a 300W beast, running a cut down chip. Think about that for a bit and understand that there's no advantage for GP100 vs GP102 for gaming.
Zilch. Same effective FP32 performance, enough bandwidth with memory compression.
To put GP100 as a gaming card would be a downgrade from GP102 as you get significantly worse perf/w and added cost of HBM2 for no gains.
GP100 is 600mm2, with HBM2's space savings already. It's also 300W, again, with HBM2's power savings.
This is why its such a detriment to have 1:2 FP64 CC on the chip for gaming, it's simply bloat on a level that has meant NV has had to design a new gaming flagship in GP102 without that FP64.
GP100 is 600mm2, with HBM2's space savings already. It's also 300W, again, with HBM2's power savings.
This is why its such a detriment to have 1:2 FP64 CC on the chip for gaming, it's simply bloat on a level that has meant NV has had to design a new gaming flagship in GP102 without that FP64.