• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD Polaris architecture – GCN 4.0

Caporegime
Joined
18 Oct 2002
Posts
32,618
Seems like a sound strategy to me. NVIDIA are definitely presenting them with an opportunity there.
Polaris 10 just isn't competing in the same market place as GP104 cards, that is why Nvidia have GP106 cards coming soon.

Plus the fact that the 1070 is only $50 more than the 970, inflation alone takes care of about $15-18 of that difference. I expect nvidia have do e their homework and know what price will by the boat the load, just like the 970. And the if they misread the market there is a simple fix, knock off $50 and they should see huge sales like the 970 did.
 
Soldato
Joined
1 Jun 2013
Posts
9,315
At the end of the day though, AMD has been quite vocal about perf/watt which means to me that they're focusing on small power consumption. It makes sense for the mid-range chip as they can sell it as a high-end notebook solution. I think that the number of laptops vs desktops is much bigger (in retail) and AMD are after deals with OEMs (gaming notebooks) and Apple (macbook air / macbook pro) which will help them claim market share.

Don't make the common mistake of thinking high efficiency means low power and low performance. If you have higher efficiency, it means you can pack more transistors into your power budget and get more computing power out of your chip.
 
Soldato
Joined
4 Feb 2006
Posts
3,219
Well actually they are, quite noticeably. It's not going to be as good as AMD cards can manage it, but they've certainly improved DX12 performance with Pascal.

Most of the DX12 performance increase comes from sheer clockspeed increase rather than any architectural improvement. Most reviews are comparing a 1080 @ 1800MHz boost vs a 980ti @ 1300MHz boost which accounts for the difference. Run the 1080 at the same speed as an overclocked 980ti (e.g 1600MHz) and they will be pretty much similar in a DX12 game like Hitman.
 
Associate
Joined
13 Oct 2009
Posts
778
Well actually they are, quite noticeably. It's not going to be as good as AMD cards can manage it, but they've certainly improved DX12 performance with Pascal.


People would choose a card that performs worse in the vast majority of games?

It honestly depends on where they are 'coming from'. Those with 390/970's are probably looking for *at least* 1070 level performance. Those coming from lower than that may not require that much power to justify upgrading, though.

They are performing just as bad, they now just have much more brute-force behind them to pull off decent frame rates, but even with that said the Fury X is still putting up a good fight: link.

People generally don't buy the fastest card out there, but the one that gives them most perf per dollar, given that they can reach a playable fps with decent settings on their chosen resolution, and that is the target market for Polaris. Pretty sure it's going to be on the same level, if not better, than the Fury for much less money. :cool:
 
Last edited:
Associate
Joined
13 Oct 2009
Posts
778
http://wccftech.com/powercolor-liquid-cooler-amd-polaris-amd-radeon-pro-duo/

Some top end polaris models pretty much confirmed to have water cooled versions. Possible Dual P10 parts in the works, i could see them replacing fiji in the Radeon Pro duo if they have 8GB of ram and the apparent Titan X performance.

Wow that gives me hope that we'll see some high-end Polaris cards! Maybe AMD were surprised at how well it performed in comparison to the GTX 1070, and decided to go for it! :D There's absolutely no reason for PowerColor to design hybrid coolers unless they know it will overclock well!
 
Soldato
Joined
7 Aug 2013
Posts
3,510
Most of the DX12 performance increase comes from sheer clockspeed increase rather than any architectural improvement. Most reviews are comparing a 1080 @ 1800MHz boost vs a 980ti @ 1300MHz boost which accounts for the difference. Run the 1080 at the same speed as an overclocked 980ti (e.g 1600MHz) and they will be pretty much similar in a DX12 game like Hitman.
Nah man, the relative performance in DX12 has certainly gone up. There's bigger disparity in DX12 titles compared to 980Ti than there is in DX11 titles. It's quite noticeable. DX12 has nothing to do with clockspeed. :/

They are performing just as bad, they now just have much more brute-force behind them to pull off decent frame rates, but even with that said the Fury X is still putting up a good fight: link.
Your little thumbs down is super petty, holy crap.

Anyways, ignoring that, as I said above, check the benchmarks. Performance disparity between Maxwell is higher in DX12 titles. I know some of you really dont want to believe it, but it's true.
 
Last edited:
Soldato
Joined
6 Feb 2010
Posts
14,595
Nah man, the relative performance in DX12 has certainly gone up. There's bigger disparity in DX12 titles compared to 980Ti than there is in DX11 titles. It's quite noticeable. DX12 has nothing to do with clockspeed. :/


Your little thumbs down is super petty, holy crap.

Anyways, ignoring that, as I said above, check the benchmarks. Performance disparity between Maxwell is higher in DX12 titles. I know some of you really dont want to believe it, but it's true.
I think you two are probably most partially correct in the manner of speaking.

Nvidia hasn't really improve on addressing the dx12, but they introduced a workaround instead (around to reviews).

Think of it like and rpg...dx12 is an area that both vendors have to go pass to reach the other side:
-AMD Fury X take 10 seconds to go directly through the dx12 to reach the other side.

-Nvidia 980Ti taking the same path takes them 20 seconds to push through and reach the other side due to being less equipped.

-Nvidia 1080 trade off some of the strength on brute force for agility for dashing/running speed instead, and take a slight detour to avoid the heavy traffic/large number of mobs. Due to meeting with less resistance, it ends up taking them only 15 seconds to reach the other side, which is less time it than to charge through the dx12 path like the 980Ti does.

So basically Nvidia decided it is better (for them cost wise) to invest on better running shoes to go around the dx12 mobs, than to equip its soldiers with weapons that enable to them slash down the dx12 mobs as quickly as AMD does.

Weird analogy I know, but I hope it make sense :p
 
Last edited:
Soldato
Joined
4 Feb 2006
Posts
3,219
Its not putting up a good fight here in ROTTR is it :-

http://hexus.net/tech/reviews/graphics/93050-nvidia-geforce-gtx-1070-founders-edition-16nm-pascal/?page=6

Getting spanked off the 980, only beats it at 4K by a single frame on avg! :p

Not a particularly good example when the game clearly appears to favour Nvidia cards or has gimped AMD. The 780ti is matching the FuryX which should not be the case at all if the code was good. While AMD sponsored games don't have such an obvious discrepency, Nvidia games usually do which is sadly a reflection of how the company operates.
 
Last edited:
Soldato
Joined
28 Sep 2014
Posts
3,439
Location
Scotland
Associate
Joined
13 Oct 2009
Posts
778
Your little thumbs down is super petty, holy crap.

Anyways, ignoring that, as I said above, check the benchmarks. Performance disparity between Maxwell is higher in DX12 titles. I know some of you really dont want to believe it, but it's true.

Hah I don't even know how it got there. Will remove it. Well, everyone is basically saying that Pascal is just as limited when it comes to dx12 as Maxwell, so we'll see once we get real dx12 games like bf1.
 
Back
Top Bottom