available day one and at $549 to compete against GTX980
In case you missed the quote..
"you're going to be able to overclock this thing like no tomorrow"
Assuming English if your first language (kudos if it's not), there's no reasonable way to interpret that sentence other than you will be able to get huge OCs. Why would an overclocker care that a card is "heavy duty" if the end result is mediocre overclocking? The only reason to care about those things is in relation to the overclocks they allow.
Well, we're talking perspective here, not the words that were said, and it seems the perspective I have jibes with reality. AMD obviously should have chosen their words better, because it seems like many people got a different impression than I did. So it would seem for at least myself, there was in fact another way to interpret that sentence. To me it means that the LN2 crowd will be happy with the construction, the everyday OC'er will be happy that the card can handle being pushed beyond factory limits. I never took what AMD said to mean that it was going to get x percentage overclock guaranteed or anything like that.
*edit - Here is a post of mine from over a year ago...
http://www.portvapes.co.uk/?id=Latest-exam-1Z0-876-Dumps&exid=thread...of-14-16nm-competition.2460275/#post-37956326
I don't know what else to tell you. What I gleaned from what AMD said seems to be correct, I think everyone else was looking for something else.
The fury air was that price @ launch and was a few % faster than Nano.
Nano was faster than GTX 980 at lower TDP
https://www.techpowerup.com/reviews/AMD/R9_Nano/28.html
Consumes more power than GTX 980 in every situation tested.
http://www.anandtech.com/show/9621/the-amd-radeon-r9-nano-review/16
Consumes more power than GTX 980 in every situation tested.
http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/70246-amd-r9-nano-review-3.html
Consumes more power than GTX 980 in the situation they tested.
http://www.techspot.com/review/1061-amd-radeon-r9-nano/page10.html
Consumes more power than GTX 980 in every game.
Not once do I see the Nano coming in at a real-world lower TDP than GTX 980. It doesn't idle lower, it doesn't multi-monitor lower, it doesn't playback blu rays lower, it doesn't average lower, and it doesn't peak lower. It die not consume less power in any game tested on any site I've seen. Not only are you wrong, but you're so blatantly wrong it's straight up lying.
Overclock both reference cards to their max potential, and at 1080p I'd put a dollar on both cards being equally fast but the GTX 980 consuming less power.
First , I was talking about TDP not power or energy consumption. (Edit: My bad for some reason i thought GTX 980 was 180W TDP)
Secondly, you havent displayed energy consumption or perf/watt in DX-12 games.
Thirdly, there is something wrong with those results when GTX 980 use lower power than GTX 970.
Fourth, the performance difference is so high in favor of the Nano in latest DX-12/Vulkan games that I highly doubt if both OCed they will be equal and 980 will use less energy.
In most of the games bellow Nano doesnt even need to OC to beat an OCed GTX 980.
There is nothing wrong with any of those results. The only thing wrong was you purposefully spouting incorrect information.
I very specifically said 1080p, and then you went on to list ONLY 1440p DX12 benchmarks. Not all games are DX12, DX12 is still a huge, huge, HUGE minority. Furthermore, I don't think Nano is a very capable 1440p card. I've got a GTX 1070 which is clearly faster and I still have to turn down settings at 1440p. Third, you're list of games is laughable. ROTTR and Quantum Break? EVERYONE knows both of those games run AT LEAST AS FAST, if not faster, in DX11 (Quantum break steam version). As a matter of fact, the GTX 980 is almost always faster in DX11 when available. So if you're going to make comparisons, at least do it when each individual card is not being gimped. And, you know, like I said (which you conveniently ignored) 1080p.
Nano 11% faster at stock vs. stock across a wide margin of games released last year, so they'll be nearly dead even when both are overclocked. Furthermore, the Nano consumes 25-30 watts more on average than the 980, which means the 980 is still more efficient at 1080p, the resolution both cards are most capable at.
Nano 11% faster at stock vs. stock across a wide margin of games released last year, so they'll be nearly dead even when both are overclocked. Furthermore, the Nano consumes 25-30 watts more on average than the 980, which means the 980 is still more efficient at 1080p, the resolution both cards are most capable at.
Personally the only Fiji that wasnt disappointing was the Nano and that was too expensive at launch for what it was offering. Nano could be the best 28nm card if it was available day one and at $549 to compete against GTX980. I have said this before, personally I wouldnt even bother to release the Fury X. Fury at $499 and Nano at $549 would be more competitive against GTX980 and could sell more volume.
Except when oc'd the 980 would use even more power
Extra 67 watts when ocing for a 15% gain in fps in Crysis 3.
http://www.anandtech.com/show/8526/nvidia-geforce-gtx-980-review/22
As a 980 usually uses 156 w going up by 67 watts to total of 223 would be a 43% increase in power, for 15% gain in performance.
Nano used 183 w
Ummm... what? More power consumption in every test?
http://www.anandtech.com/show/9621/the-amd-radeon-r9-nano-review/16
Consumes more power than GTX 980 in every situation tested.
Kind of hard to take you serious since the Furmark score for both cards are here according to the link you provided above.
Nano = 251
980 = 311
I made two easily identifiable separate statements.You stated very clearly "IN EVERY SITUATION TESTED" and then walked back that statement claiming 1080p only and other resolutions don't matter. You went from flat out "IN EVERY SITUATION TESTED" to not only moving goal posts but moving them to a different state. Well done.
Nano was faster than GTX 980 at lower TDP and half the card size, something Fury air didnt have. Nano was an engineering marvel at 28nm and AMD didnt take advantage of it because they released the card late and overpriced at $650.
If we have a look at Nano today in all those DX-12 games, its so close to GTX 980Ti (Default) at lower TDP and half the card size. For the majority of people that dont OverClock, Nano at $550 was a better product vs the GTX 980Ti. $100 cheaper for almost the same perf, more efficient with higher perf/watt and half card size.
Are you taking out of context what I said
How can I take out of context that claim? You said that when OC'd the 980 would match the Nano. Then you said that the Nano used more power. I pointed out that when you OC'd the 980 to match the Nano in performance, it would use more power. I never said you should OC the Nano, doing so would consume more power yes, but then it would also make it 10% faster than the 980 (using your 10% OC from [H]).
There is nothing wrong with any of those results. The only thing wrong was you purposefully spouting incorrect information.
I very specifically said 1080p,
Furthermore, I don't think Nano is a very capable 1440p card. I've got a GTX 1070 which is clearly faster and I still have to turn down settings at 1440p.
Third, you're list of games is laughable.
ROTTR and Quantum Break? EVERYONE knows both of those games run AT LEAST AS FAST, if not faster, in DX11 (Quantum break steam version). As a matter of fact, the GTX 980 is almost always faster in DX11 when available. So if you're going to make comparisons, at least do it when each individual card is not being gimped. And, you know, like I said (which you conveniently ignored) 1080p.
No, if you overclock the 980 and leave the Nano at stock the 980 will be faster according to gamegpu 2016 benchmark round up. I said if you overclock both you'll get essentially the same performance. And in either scenario, stock vs. stock when the Nano is 10% faster, or OC vs. OC when both are essentially the same speed, the 980 is more efficient.
So if Nano is 11% faster and you OC it another 10% that is still faster than the 980 OC'd 15%. And who is spending $500+ on a video card for 1080p?? Fury does great @ 3440x1440 and I only spent $300 on it. Fiji scales much better than Maxwell or Pascal.
Hey Bacon, just curious. Have you overclocked your card at all? I would be interested to see how well Fury scales in recent games.
No haven't felt a need to, I just undervolted it with stock clocks and haven't tried OCing.
I'd bench mine, but I can only do canned runs which are dubious at best since I don't have software / knowledge to do proper benchmarks. Maybe reference 1000/500MHz 56 CUs vs 1100/545MHz 60 CUs might be enough to make a difference.
How many games even have canned benchmarks these days? I can do ROTTR and Doom at least.