AMD Polaris architecture – GCN 4.0

[arc@css]

Does anyone run a 980ti at stock?

I do.

 

[Kaapstad]

Does anyone run a 980ti at stock?

My 980 Ti @stock has no trouble beating an overclocked and overvolted Fury X.

I always game with the GPUs at stock.

Spoiler

I will get my coat.

 

[JediFragger]

I do.

Weirdo

 

[r7slayer]

Please tell me you have those cards under water Kaap?
Also didn't realise they had full coper heatsinks. What temps do they run at if you don't have them under water?

 

[arc@css]

Weirdo

That goes without saying.

 

[maonayze]

I do.

LOL Straight outta Monty Python, that one.

The lone little fella in amongst the thousands.

Good point though. What a strategy by Nvidia....Bring out underclocked cards and then no-one will care what efficiency they do or do not have after overclocking. That's a clever bit of slight of hand I must admit.

Sneaky edit

 

[Overlag]

I do.

so you've disabled GPU boost?

 

[Kaapstad]

Please tell me you have those cards under water Kaap?
Also didn't realise they had full coper heatsinks. What temps do they run at if you don't have them under water?

I use them on air.

I got them for the copper coolers.

Flat out they run around 55c.

Stock volts they can do over 1570 on the cores.

 

[arc@css]

so you've disabled GPU boost?

I've no idea.

Will be interesting to see how she runs when pushed, though. I'll have to have a go some time.

 

[bru]

So comparing an overclocked 980ti power usage to a stock 290x is fair, either overclock both or neither.

 

[Screenbeat]

Still have my hd7950 and loving it!!

 

[Overlag]

So comparing an overclocked 980ti power usage to a stock 290x is fair, either overclock both or neither.

Well you have 50mhz head room on the AMD setup vs 300-400 on the Nvidia, which is the point he was trying to make. The AMD setup is at the limit, hence the TDW is higher vs the Nvidia which is downclocked to make it look more efficient. Use GPU boost (or overclock) and it puts it straight up there with AMD GPU power usage.

 

[David Bisset]

Aye, people suggest comparing Nano power use isn't fair for the same reasons.

 

[Ian Evey]

It wouldn't surprise me if pascal is way faster at 1080 / 1440 res than Polaris is on release but Polaris is way faster @ 4k.

 

[Boomstick777]

AMD Unveils “Polaris 11” And “Polaris 10” GPUs – To Deliver The “most revolutionary jump in performance so far”

Read more: http://wccftech.com/amd-unveils-polaris-11-10-gpu/#ixzz3xQyWdqsX

AMD’s Raja Koduri unveiled two next generation 14nm GPUs, dubbed Polaris 11 and Polaris 10, to power high-end desktops and bring console-class gaming performance to notebooks. Both of these graphics processing units are based on AMD’s recently announced next generation”Polaris” graphics architecture. Which features the high performance Samsung/Globalfoundries 14LPP FinFET manufacturing process technology

 

[Simon Belmont]

Opps wrong thread

 

[FoxEye]

Hmm, a "console class GPU" is something a bit better than a 7870, iirc. Not as good as a 7950 by a long way.

That is one of the announced GPUs, intended mainly for notebooks, and very low-end desktop (equal to the 270/370, roughly...)

That means one of two things, then. They have one new GPU to power everything from the new Fury class GPU, then the new 490 class, then the new 480 class.

Or they are going to have to re-use some of the current range. Like the Fury and the 390.

Otherwise it would be like nV having to use the same chip (GM100) for Titan, the 980ti, the 980 AND the 970. Could they actually do this? Seems unlikely.

So... we should expect some re-brands in the next-gen lineup?

 

[Simon Belmont]

Hope Nvidia have something to counteract it?

 

[fs123]

AMD seems to be showing the next gpu's while nothing seems to be coming out from Nvidia when Pascal is supposed to be here first. Am I missing something?

 

[drunkenmaster]

Hmm, a "console class GPU" is something a bit better than a 7870, iirc. Not as good as a 7950 by a long way.

That is one of the announced GPUs, intended mainly for notebooks, and very low-end desktop (equal to the 270/370, roughly...)

That means one of two things, then. They have one new GPU to power everything from the new Fury class GPU, then the new 490 class, then the new 480 class.

Or they are going to have to re-use some of the current range. Like the Fury and the 390.

Otherwise it would be like nV having to use the same chip (GM100) for Titan, the 980ti, the 980 AND the 970. Could they actually do this? Seems unlikely.

So... we should expect some re-brands in the next-gen lineup?

No, it's a new process that is more expensive, lower yield and much more complex than the last. Neither AMD nor Nvidia made their big gpu first. It's not at all feasible to do so.

Two gpus are coming, a smallish one, probably somewhere around the 125-175mm^2 size, this is the console level gpu. Then an enthusiast card, probably 300-400mm^2 somewhere. It will be aiming for probably 15-20% beyond Fury performance, depending on architecture it may be more or less. It depends heavily on how dense they can make the gpu on the new process compared to yields.

The Fury or Titan type lets say 450mm^2 or larger cores, won't come in 2017 full stop. I think AMD may well have HBM2 on the enthusiast card and gddr5 on the lower end one. If we see a huge core on 14/16nm at all from either company I don't know. 10nm is genuinely pretty close behind so it simply may not be worth making a big expensive 14nm part when 6 months later they may be able to do another 300-350mm^2 10nm part instead.

Effectively this year we're getting the 7870 and 7970, with the 290 /Fury to come in 2017 if at all. Nvidia we will see the 680gtx type core, NOT the Titan.