One has to take the statement in it's proper context. The minimum Occulus spec [290/970 class] price is named as the limiting factor.Isn't VR more AFR friendly now? Isn't that the way AMD plans on doing it?
VR it seems will have way more uses than gaming. Around here the focus is strictly on the gaming aspect. To me it seem like less gpu power would be needed to render a static image of say the Smithsonian or a persons heart, etc.
GC: You've mentioned the existence of quality content as one of the drivers of VR adoption. What are the other obstacles limiting widespread adoption of VR?
RT: They are two-fold. First of all, it's expensive for a lot of people. Now the people who really want it, just don't care. Quite frankly Oculus could have sold their headsets for a thousand dollars and I'm pretty convinced they still would have sold all of them. Because this first wave of adopters just aren't going to care. But when you get to the second wave of adopters, to people who are not technically savvy, then the price becomes tentative. And right now it's cost-prohibitive for a lot of people.
If you look at the minimum spec for either the Oculus or the HTC, and then you look at how many units of the minimum spec have been sold since their launch, so I'm talking about the Radeon 290 or GeForce GTX 970, according to Jon Peddie Research, the total install base of those parts or better is 7.5 million units. So we're going to have to make it possible to run good quality VR at a much lower price. And I'm confident with Polaris we're going to have a big impact to help that.
28nm Hawaii/Grenada XT
438mm2 - 6200 Million transistors
~14,15 Transistors/mm2
14nm FF has ~2.2 density = ~31,15 Transistor/mm2 (Assuming we keep the same design)
Polaris 11 = 232mm2
So Polaris 11 could have close to ~7200 Million transistors or one Billion transistors more than Hawaii/Grenada
For comparison, Fiji has 8900 Million Transistors
So even at the same clocks, Polaris 11 at 232mm2 could be faster than Hawaii/Grenada XT.
Gate utilization decreases with 14nm contra 28nm. So you may end up with 57% more working transistors and not 120% more.
Also remember parts that dont shrink so well. Like a memory controller.
ehmm what ??
Yea you also have to remember that Hawaii/Grenada has 512bit Memory controller, Polaris 11 will be HBM2 most probable. That will leave more space for more Shaders, TMUs, ROPs, etc.
Yes they did that every time. But the same uarch is the same uarch.
They also said the performance/watt is 2x. A direct shrink will pretty much do that.
I'm coming to the conclusion that 14nm finfet might be equal cost or cheaper per transistor that 28nm. An older 28nm GPU of similar performance might be more expensive to fab than a 14nm one, even more so if architectural improvements are effected, meaning less transistors needed for 'X' performance. It would in effect, cost more to make the cheaper selling GPU. Not a situation you want.That's easy, cost. Someone here said it - efficiency is a new "marketing" term in itself. "Go Green." "Buy Carbon Dollars" or whatever the hell they are called.
Say you have two equal products:
GPU R7 470 (equal performance to current R9 390) @ $150-200
GPU R9 480 (R9 390 + 10%) @ 300-350
the 470 uses 220W, the 480 uses 110W. You will have the savvy buyers (and those cutting edge) opting for 480, but those on a budget would probably get the 470.
And now both users can run VR if the 290 is the base-point they are aiming for.
[All numbers above are made up, just used for my point]
At this point, who knows. My aggressive stance that they need to make more money (now) stems from the fact that NV is going to keep doing their little tricks and AMD is going to need resources to fight back. They've had amazing tech for years, often selling it for pennies compared to NV. If this is their chance to make some money, grow their brand (no longer the budget brand!), I say do what you can and go for it.
However, I also have zero confidence in their marketing teams so who knows if it will even matter if they have any advantages when they can't get joe public to acknowledge it.
This is also here Intels structure win in cost.
Assuming it will use HBM2.
My sense is that this round, AMD is going for the gold in all metrics.
Not to mention that the cost of 14nm FF in the table above is for Q4 2016 (first 3-6 months of production) when 28nm is at its 3-4th year of production.
More I think of it:
It don't explicitly say that Polaris would be the means to make the "quality VR run at a much lower price" the second statement implies Polaris will help.
Perhaps by replacing the top tier price point and thus allowing AMD to sell older GPUs that can run "quality VR at a much lower price." Polaris would maintain their margins while last gen spreads VR goodness to the masses.
http://wccftech.com/amd-polaris-architecture-vr-minimum-spec/#ixzz40ACbH2BW"
"AMD has just completed the shrink to 14 nanometer [with Polaris Architecture]. What this means is, and this is where it comes home to everyone in this room, is that we can produce GPUs that will run the minimum spec of VR at a lower cost, in larger volume, consuming less power and running faster. "
Roy Taylor's statement was that AMD are going to bring Polaris products which are faster than R9 290X and cheaper and more power efficient and in much higher volume (again indicating the price points they intend to hit). I don't know how such a simple statement can be so misinterpreted.
290 is the AMD VR ready. Not the 290X.
I would expect around 390X performance at 125-150W.
290 is the AMD VR ready. Not the 290X.
I would expect around 390X performance at 125-150W.
If AMD is only able to pull off 390x performance at 125w TDP, they're going to get smoked by Nvidia's 14nm considering the GTX 980 is already at around 165w TDP.
Honestly, if AMD is unable to deliver Fury performance @ 125W TDP, it's going to be a huge failure (pending pricing, of course).
If AMD is only able to pull off 390x performance at 125w TDP, they're going to get smoked by Nvidia's 14nm considering the GTX 980 is already at around 165w TDP.
Honestly, if AMD is unable to deliver Fury performance @ 125W TDP, it's going to be a huge failure (pending pricing, of course).
If AMD is only able to pull off 390x performance at 125w TDP, they're going to get smoked by Nvidia's 14nm considering the GTX 980 is already at around 165w TDP.
Honestly, if AMD is unable to deliver Fury performance @ 125W TDP, it's going to be a huge failure (pending pricing, of course).
exactly. Nvidia will have a card faster than GTX 980 at 100w tdp. I think GP104 with 4096 cc and GDDR5X will be >2x faster than gtx 980 at 180w.
When AMD R9 series is undervolted and downclocked a bit, we get massive efficiency gains. It seems we are forgetting how they clocked past the "sweet spot" to compete in absolute terms with Nvidia. New designs allowing dynamic voltages will extract efficiency lost in the present series. As a reminder Nano running with 980 for all those missing the forest.exactly. Nvidia will have a GP106 card faster than GTX 980 at 100w tdp. I think GP104 with 4096 cc and GDDR5X will be >2x faster than gtx 980 at 180w.
When AMD R9 series is undervolted and downclocked a bit, we get massive efficiency gains. It seems we are forgetting how they clocked past the "sweet spot" to compete in absolute terms with Nvidia. New designs allowing dynamic voltages will extract efficiency lost in the present series. As a reminder Nano running with 980 for all those missing the forest.
This is also here Intels structure win in cost.
Assuming it will use HBM2.
Totally irrelevant except to the 1% interested in such things.Comparing Fiji with GM204 is stupid. Underclocking and undervolting a 50% larger GPU to match the power efficiency of a smaller GPU is nothing great. Fiji must be compared to GM200 at its released clocks. If AMD want to compete they need to match Nvidia in terms of perf/sp vs perf/cc, perf/watt, perf/transistor and perf/sq mm. Thats when they can really compete otherwise Nvidia will be always calling the shots.
This confirms what I posted earlier.Please stop posting this outdated chart. It's misleading. Note the date: 2013. As of mid-2014, there were already numerous refinements to the 20nm process and its 16nm FinFET variant, and Qualcomm was reporting that it was "more cost effective compared to 28nm HKMG processes". Again, that was a year and a half ago, so there's been another 18 months since then to refine the foundry processes, increase yields, and reduce costs. Yet you continue to post some consultant's estimate - an educated guess at best - from way back in 2013.
And I see no evidence that Intel is substantially ahead at 14nm. They have yet to produce any mass-market chip that is even as large as TSMC's A9X. Broadwell-E may not have been officially delayed, but I don't believe for a moment that a mid-2016 release is what Intel originally wanted.
AMD has already officially confirmed that some Polaris SKUs will be using HBM.
This confirms what I posted earlier.
It will cost more to produce a GPU of any given performance on 28nm than on 14nm.
This should stop any further suggestions that AMD will reuse present designs. They will have to sell at a discount to the new gen because of power consumed and will also cost more to produce. No one is that incompetent.
Folks, the 2 GPU belief needs to be revisited.