which gpu has better gaming performance?

GTFOOH

Reputable
Aug 18, 2015
28
0
4,530
Sapphire Tri-X Radeon R9 FURY 4GB HBM or GALAX GeForce GTX 980 HOF 4GB GDDR5 ??

the same price around me!
 
Solution
Without giving some basic data, the question is kinsa like like asking "what is the best tool" ?

a) it kinda depends on what ya wanna do with it ... bang in a nail, tighten a bolt or put in a screw each will get a different answer.
b) it much depends on how big the project is ... it it's putting in a screw, I'd have a different answer than if it was put in 1,000 screws

Tho the release of the 980 Ti kinda kills the 980 from a bag for the buck standpoint., the 980 is still an easy choice over the Fury. So I will assume that the question is 1440p and gaming only

The relative ranking of the cards at 1440p is 98 to 91 for the reference 980 so if you are an "out of the box" GFC card user then the Fury is the obvious choice. But since you are looking at overclocking cards.....

http://tpucdn.com/reviews/Sapphire/R9_Fury_Tri-X_OC/images/perfrel_2560.gif

However the Fury has almost no overclcoking headroom with the Sapphire getting you just 7.7 % (101.5 / 94.2) over reference.

That would raise the 98% score of the Fur X to 105.5 (98 x 1.077)

However if we take the 91% score of the 980 and apply the 22.7% (160.5 / 130.8) overclocking potential of the 980, we get 91 x 1.227 or 111.7.

http://www.techpowerup.com/reviews/MSI/GTX_980_Gaming/28.html

Based upon TPUs data anyway, that's about 6% faster than the Fury



 


I disagree, the Fury beat the 980 85% here based on Guru3D reviews http://s1068.photobucket.com/user/loki1944/media/AMD%20VS%20NVIDA_zpszmsk5jfx.png.html.

80% of the time @1440p in the Andantech review: http://www.anandtech.com/show/9421/the-amd-radeon-r9-fury-review-feat-sapphire-asus/14

100% of the time @1440p in the eurogamer review: http://www.eurogamer.net/articles/digitalfoundry-2015-amd-radeon-r9-fury-review

Only time the 980 gets an edge is at lower resolutions like 1080p.
 
See rest of post above ... as i said, if you are buying reference cards and using them outta the box, the Fury is the way to go/ However, that's not generally the mindset of those paying a price premium Sapphire's and HOFs. These folks are, way more often than not, gonna overclock the begeezes outta them.

The Fury, like the FuryX and all the 3xx cards has single digit overclocking headroom. This is because all AMD has done is apply a significant overclock before putting the card in the box.

With both cards overclocked "balwz to the wall", the out of the box performance is more than eclipsed by the 23% overclocking headroom compared to the Fury's 7%

The only time the Fury makes any sense, if you can call it that, is at 2160p and, at this time, no single GPU makes sense at 2160p.... it's hard to make a case for any 2 cards at 2160p.
 
disagree, the Fury beat the 980 85% here based on Guru3D reviews http://s1068.photobucket.com/user/loki1944/media/AMD%20....

80% of the time @1440p in the Andantech review: http://www.anandtech.com/show/9421/the-amd-radeon-r9-fu...

100% of the time @1440p in the eurogamer review: http://www.eurogamer.net/articles/digitalfoundry-2015-a...

Only time the 980 gets an edge is at lower resolutions like 1080p.

It's true until you start overclocking. The gtx 980 can reach north of 1400mhz in pretty much every version of gpu while the fury can barely reach 1150mhz. Once overclocked, the gtx 980 would win in all resolutions.

https://youtu.be/-BTpXQkFJMY
 

Th3-Hunter333

Reputable
Dec 15, 2014
508
0
5,060


Not even all that mate, ive seen quite a large number of 970's/980's break 1500 mhz on the core with ease.

I.E. Im running my 970 fully stable on 1500 core at stock voltage
 

Reaper_7799

Distinguished
Everyone loves to compare fury and fury x stock to reference cards stock because the perfomance is pretty even to a 980 or 980 ti but then they always conveniently leave out overclocking as if it doesn't matter. In the case you don't overclock...it's kind of a waste if you don't but then the cards are pretty even but who doesn't love more performance, especially at this level.
 


Strix 980 vs Strix Fury; Fury wins @1440p in guru3d reviews. Overclocking results are not guarranteed. You could end up with poor ASIC quality. Also I did not talk about reference cards at all; I showed STRIX vs STRIX results. At 1440p even the Strix 390X trades blows with the Strix 980.

Shadow of Mordor 1440p: STRIX 980 79FPS / STRIX 390X 82FPS / STRIX R9 Fury 92FPS

Thief 1440p: STRIX 980 54FPS / STRIX 390X 55FPS / STRIX R9 Fury 67FPS

Tomb Raider 1440p: STRIX 980 102FPS / STRIX 390X 95FPS / STRIX R9 Fury 120FPS

Bioshock Infinite 1440p: STRIX 980 87FPS / STRIX 390X 81FPS / STRIX R9 Fury 99FPS

Hitman Absolution 1440p: STRIX 980 65FPS / STRIX 390X 72FPS /STRIX R9 Fury 79 FPS

2160p STRIX 980 vs STRIX R9 Fury:

SoM: STRIX 980 42FPS / STRIX R9 Fury 50FPS

Thief: STRIX 980 30FPS / STRIX R9 Fury 37FPS

Tomb Raider: STRIX 980 47FPS / STRIX R9 Fury 58FPS

Bioshock Infinite: STRIX 980 44FPS / STRIX R9 Fury 50FPS

Hitman Absolution: STRIX 980 33FPS / STRIX R9 Fury 41FPS
 
The number are the numbers..... it's not like they aren't consistent, there is very little variance among the reviews. Guaranteed, no.... but find just one where the 9xx series doesn't overclock 2-3 times better than a Fury / R9 series.

I have never .... ever ... gotten less than a 22% OC on 5xx, 6xx, 7xx or 9xx non reference card. 9 outta 10 break 25%. Ofc these were all non reference cards, well cooled and the one with the most robust PCB package.

Here's the Strix Fury and Strix 980 cards overclocked.... not exactly the strongest contenders available in the current series and not as if TPU goes "Bawlz to the wall". While in the 6xx series Asus was the king of the hill, they surrendered that title when they started going cheap on the PCBs. MSI and Gigabyte lead the pack here. MSI and Sapphire on the AMD cards. But back to *your numbers*

http://www.techpowerup.com/reviews/ASUS/GTX_980_STRIX_OC/28.html

Actual 3D performance gained from overclocking is 13.2%.

That is 19% over reference, not bad for the Asus design

http://www.techpowerup.com/reviews/ASUS/R9_Fury_Strix/34.html

Actual 3D performance gained from overclocking is 6.4%.

The Strix gives the same OC as the reference design....(94.2 fps)
http://www.techpowerup.com/reviews/Sapphire/R9_Fury_Tri-X_OC/33.html

So what we are left with is the Asus 980 has 3 times the overclocking headroom of the Asus Fury.

Arstechnica summed it up perfectly and they were just comparing out of the box performance

http://arstechnica.com/gadgets/2015/07/the-r9-fury-is-amds-best-card-in-years-but-just-who-is-it-for/

On the other hand, the R9 Fury is best thing AMD has put out in ages: It's a cooler, quieter graphics card that turns in a performance often matching its own more expensive R9 Fury X. And really, that's who the R9 Fury is for: the AMD user wanting the best the company has to offer without having to stretch to a Fury X. For everyone else, it's a much tougher sell.

No argument, AMD puts a very aggressive clock on the cards before they are boxed and if you don't OC, there's a strong argument to go that way. But for anyone who uses Afterburner or it's equivalent, it's a real hard sell.




 

Reaper_7799

Distinguished


Here they are both overclocked...they trade blows. Not 85% win for fury and certainly not by 15-20 fps every game up there. http://www.pcworld.com/article/2951297/components-graphics/fight-custom-radeon-fury-gtx-980-graphics-cards-brawl-for-pc-gaming-supremacy.html

dragon-age-overclocked-h2h-100597715-orig.png


mordor-overclocked-h2h-100597727-orig.png


metro-overclocked-h2h-100597724-orig.png


gta-v-overclocked-h2h-100597721-orig.png


 


Again, not guaranteed. 85% STRIX R9 Fury vs STRIX 980 is guaranteed. You get a 65% ASIC quality card and you're going to be in for disappointment thinking you can overclock a bunch. I've got a 65.9% ASIC quality STRIX 980Ti which will only go +105Mhz Core and +160Mhz Memory without over-volting.
 

Reaper_7799

Distinguished
Yeah okay whatever....I'd say 80-90% or more can get that oc and you know it. See this is how you can defend it right. It's not guaranteed...not everyone overclocks...yada yada....meanwhile for the ones that do and there are a lot of them...the results are much different.
 


I'm willing to bet hands down the games where the Fury beat the 980 that I showed would be the same story with both cards overclocked. Oh wait, I don't have to bet, Guru3D did it for me:

Hitman Absolution: @1440p: STRIX 980 O/C 74FPS / STRIX R9 Fury O/C 88FPS

Bioshock Infinite @1440p: STRIX 980 O/C 98FPS / STRIX R9 FURY O/C 105 FPS

Tomb Raider @1440p: STRIX 980 O/C 115FPS / STRIX R9 FURY O/C 129FPS

If we only consider these O/C results and results you showed then the STRIX R9 Fury still wins 75% of the time, even with both cards O/C.
 


That's a pretty minuscule representation of everyone who bought 4690K I'm sure, since that's less than 200 people unless I'm missing something, though overclocking to 4.5Ghz on a 4690K isn't impressive anyway since just turning on turbo through maximum performance gets you 4.4Ghz without overclocking anything. Even 4.8Ghz is a relatively small overclock of 400Mhz past stock turbo. I mean, even my i7 960 did 640Mhz past turbo (3.46Ghz turbo, 4.1Ghz O/C) on an H50 and my 5930K does 800Mhz (3.7Ghz turbo / 4.5Ghz O/C )past turbo on six cores. That still doesn't mean I'm guaranteed good overclocking ability; I've got a 4770K that won't go past 3.89Ghz.