They have same bandwidth.But yeah hawaii have more rops and TMU.But in 1080p rops will have little impact.
They have same bandwidth.But yeah hawaii have more rops and TMU.But in 1080p rops will have little impact.Why compare only TFs for gaming when ROPs are halved, TMUs are less and memory bandwidth almost half and yet Polaris is faster.
As i have said before, you have to compare against R9 390 at the same clocks vs RX480 and see how much faster Polaris is in gaming.
Comparing only TFLOPs for gaming is wrong.
What are you on about? Who underclocked 390X ram, AMD and/or Techpowerup?omg how many times i must say to you that they underclocked 390x to match rx480?
390x 512bit with ram at 2000
rx480 256bit with ram at 4000
Thay have same 256GB/s bandwidth.This is 3rd time i am telling this to you.
Unfortunately, it is currently hardly possible to pinpoint the advantages of Polaris in games. Instead, you can only do this roughly - as in this section. For example, the Radeon R9 390X can be easily upgraded to a theoretically similar level of performance as the Radeon RX 480. By minimizing the clock rate to 1035 MHz, the theoretical shader and texture performance of both graphics cards is the same. A memory clock of only 2,000 MHz also results in an identical memory bandwidth
Where the problem arises is the inability to make a functioning ngg path in vega after so many years of designing new arch.
That surely is a no bonus result
Sad true is that AMD is stuck on 4x shader engines/64rops/256TMU/4096SP.That is pretty unbalanced setup.
I have written before in another thread there is a potential opening here. But, a lot of stars would have to align. Too many IMO.
AMD is simply too far behind, and while there has been a lot of complaing, NVidia is not really standing still on performance on regular games with RTX.
I will be totally shocked if AMD released a gaming GPU that caught 2080Ti on non Ray traced games. I don't think there is more than a 1% chance of that happening.
2060/2050 likely won't have RT cores, so they will be small die pure conventional performance. So no opportunity there.
That leaves 2080/2070. AMD might in theory catch 2080/2070 if the planets align, but they probably won't undercut NVidia on price, just like they didn't with Vega.
So there is opportunity, but likely nothing will come of it.
There was a statement but it didnt explain why.Was there actually ever a statement why it doesn't work? is the hardware broken or just no software support (driver or game devs) ?
I've been thinking, and I'm convinced that 1% chance could actually be much higher.
It's true, Nvidia isn't standing still on regular performance with RTX, but it's also not dedicated the whole silicon to regular performance so, for example, you aren't getting ~750 mm2 worth of regular performance with 2080 Ti, or ~500 mm2 with 2080 - you're getting less.
Even if AMD's new architecture is less efficient than Nvidia's, the fact that it'd be built on 7nm, rather than 12nm, should compensate for that, and so a ~500 mm2 AMD on 7nm might be able to match 2080 Ti on 12nm in regular games. Sure, that would only last until Nvidia releases its 7nm products, but it might be enough time to swift the landscape, seeing as how AMD might be able to compete in regular games even without aggressive pricing. Nvidia's ~500 mm2 card, 2080, currently costs just shy of $800, even if prices reach the suggested base of $700 by the time AMD's cards come out, they could easily price their ~500 mm2 card at $650, which would be similar to the $1200-1000 2080 Ti in regular games. Not to mention a cut-down version like the non-X 290, or Fury, or Vega 56, that might have 85-90% of that performance for about $500 - a card faster than 2080 that costs as much as 2070. That ought to cause a stir.
As wonderful as ray tracing might be, in reality and/or as a concept, by pushing the technology in the here and now Nvidia finds itself in a situation where it not only has to compete with its previous products in regular games, but also with an AMD that would have the advantage of a new process and the lack of need to build GPU that are as big.
What do you think?
Yeah, maybe. Even if a ~500 mm2 die turns out unrealistic, a ~400 mm2 card might land above 2080, and easily sell for $450-500. The effect would be similar.On the the 7nm advantage:
Initially 7nm is going to be quite expensive, so initial parts will probably be on the small size from everyone, So I just don't see AMD shooting for a consumer die on a 7nm at ~500 mm2, at least not a while. By the time 7nm is refined for large dies, NVidia will likely be moving to 7nm as well.
I totally agree and hope AMD would have competitive RT products in the future. That's part of the reason I think it'd be beneficial for them to exploit this small window of opportunity for regular rendering in 2019 - not only would they make an extra profit, gain some market share, but they would also essentially slow down RT adoption, buying them more time, and making them look good - it's not hard to imagine how the people that bought those nice regular cards back in 2019 might be excited when some two years later the same company introduces a mature RT solution that not only works better, and is capable of powering their high-res monitors, but also exists in a landscape where RT games are much more widespread. By which point AMD becomes dominant and exploitative, and we start rooting for Nvidia. The circle of life.On AMDs die not having space devoted to RT:
Some may disagree, but IMO AMD will need it's own RT solution. RT is not PhysX or Hairworks. RT is a DX12 feature going forward, and it will be hard to ignore.
Yeah, maybe. Even if a ~500 mm2 die turns out unrealistic, a ~400 mm2 card might land above 2080, and easily sell for $450-500. The effect would be similar.
I totally agree and hope AMD would have competitive RT products in the future. That's part of the reason I think it'd be beneficial for them to exploit this small window of opportunity for regular rendering in 2019 - not only would they make an extra profit, gain some market share, but they would also essentially slow down RT adoption, buying them more time, and making them look good - it's not hard to imagine how the people that bought those nice regular cards back in 2019 might be excited when some two years later the same company introduces a mature RT solution that not only works better, and is capable of powering their high-res monitors, but also exists in a landscape where RT games are much more widespread. By which point AMD becomes dominant and exploitative, and we start rooting for Nvidia. The circle of life.
With RTX prices, the $300 market will swift to entry level. Im sure they can afford another 300-500mm2 die.
Polaris was introduced in 2016, are we all here suggesting that AMD doesnt have a better architecture to release in 2018-2019 ??
From what we have seen from the RavenRidge APUs, Vega mArch is doing ok, even without any tinkering and just porting it to 7nm will make it more than enough to compete in the $600+ market against RTX2070 and RTX 2080.
Sorry but at launch, Vega 56 was easily faster than GTX1070 at same price ($399) and Vega64 was very close to GTX1080 at lower price ($499).
But we should get some clues from the new 7nm Pro Vega coming later this year, and maybe they can make a high end consumer card out of it as well.
On AMDs die not having space devoted to RT:
Some may disagree, but IMO AMD will need it's own RT solution. RT is not PhysX or Hairworks. RT is a DX12 feature going forward, and it will be hard to ignore.
Vega has never ever been cheaper than or the same price as it's Nvidia counterpart. Vega 56/64 has been slower than its Nvidia counterpart since launch and in the past 12 months has gotten even slower. Vega has been getting slower each month as Nvidia has continued to update their drivers while AMD has done nothing for their failed vega architecture. Not only are the Vega cards slower, they consume way more power and their in-game quality is substandard. The amount of stutter I experienced with a vega56 which was purchased for mining was incredible. There are numerous complaints about this all over web. Sorry but, if you own a vega card and purchased it for anything other than mining you got totally ripped off.
https://www.youtube.com/watch?v=cJFm51OFcNA