AMD R9 Fury reviews!

Page 6 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Subyman

Moderator <br> VC&G Forum
Mar 18, 2005
7,876
32
86
Hardware.fr review.



This the problem facing the fury cards, custom cooled non-ref cards. Whats the best OC out of a fury?

Thats a good graph showing the actual landscape: Custom 980 vs custom fury. G1 pulls ahead out of the box at $519.99. I was hoping to see AMD upset the market, but it seems they simply slotted in where nvidia has been for nearly a year.
 
Feb 19, 2009
10,457
10
76
Basically with the superior XDMA, a pair of these in CF will match SLI 980Ti. SLI 980 has no hope of catching up.

With 1440p 144hz, you can't do it on a single GPU you need two at the least. Similar to 4K at 60hz.

So basically single GPU users are strictly 1080p 60hz or 1440p 60hz with toned down settings.

As soon as DSR/VSR is into the equation, we're talking multi-GPU territory.

For 1080p, go with a 970/390, best bang for buck for that resolution. 980 is caught in no man's land. It's too expensive for 1080p, too weak for 1440p. Even 1080p users can benefit with VSR from 1440p down, better AA quality overall.

Edit: Before anyone brings up multi-GPU is niche, it sure is, the same as anything beyond 1080p on el-cheapo monitors, you can find the numbers on steam. 720p & 1080p dominates. For these resolution, even a single R290 is enough so we can forget about being wasteful with $ for fancy hardware for these resolutions.
 
Last edited:
Feb 19, 2009
10,457
10
76
That graph shows the issue perfectly. Take the 390X. 100-150$ cheaper, close to the same performance and TWICE the VRAM. The competition against nVidia is bad enough. But putting 8GB on the 390/390X while the Fury series is "handicapped" with 4GB doesnt make it any easier.

That graph shows the problem with predominantly GameWorks titles. AMD scales poorly in Dying Light (with NV features HBAO & DOF, [H] found 30% perf hit), Project Cars, Witcher 3 with HairWorks (without Tessellation override).

When the amount of games widen, so does the lead Fury has on 980 as well as 390X.
 

Head1985

Golden Member
Jul 8, 2014
1,866
699
136
That graph shows the problem with predominantly GameWorks titles. AMD scales poorly in Dying Light (with NV features HBAO & DOF, [H] found 30% perf hit), Project Cars, Witcher 3 with HairWorks (without Tessellation override).

When the amount of games widen, so does the lead Fury has on 980 as well as 390X.
You have right.In normal games Fury is far faster than GTX980.I think average performance without Bugged/gameworks games is fury 15% faster than GTX980 in 2560x1440
Look at this i calculated performance difference fury vs GTX980 in each game
http://forums.anandtech.com/showpost.php?p=37549714&postcount=112

Btw can someone calculate performance summary from my post without gameworks/bugged games?
 
Last edited:

guskline

Diamond Member
Apr 17, 2006
5,338
476
126
Happy to see AMD have a solid release. Would make sense to just water cool the Fury Pro once the OCing is figured out.
 

shady28

Platinum Member
Apr 11, 2004
2,520
397
126
You have right.In normal games Fury is far faster than GTX980.I think average performance without Bugged/gameworks games is fury 15% faster than GTX980 in 2560x1440
Look at this i calculated performance difference fury vs GTX980 in each game
http://forums.anandtech.com/showpost.php?p=37549714&postcount=112

Btw can someone calculate performance summary from my post without gameworks/bugged games?


crysis3- fury 20.5% faster

Why are techpowerup's numbers you are quoting so out of whack with AT? PCPer doesn't show that kind of difference either?






 

Enigmoid

Platinum Member
Sep 27, 2012
2,907
31
91
Why are techpowerup's numbers you are quoting so out of whack with AT? PCPer doesn't show that kind of difference either?







Probably High vs. Ultra + MSAA

TPU has the unfortunate tendency to test games at unplayable settings, just like above where nothing is in an acceptably playable range.
 
Aug 11, 2008
10,451
642
126
Yeah I don't agree with that, 970 arguably has better price/performance than R9 390.

The VRAM issue sucks but if you're using it as a 1080p card as intended I don't see 3.5GB being much of a limitation.

R9 290s are a better buy than the 390 or 970, but they won't be around for much longer.

Yea, I dont even see the Fury as a compelling buy. It is a good buy, but seems pretty much a toss up compared to 980. Better performance, but costs more. And I cant believe even with the energy savings of HBM it doesnt beat an almost year old 980 in efficiency.
 

railven

Diamond Member
Mar 25, 2010
6,604
561
126
That graph shows the problem with predominantly GameWorks titles. AMD scales poorly in Dying Light (with NV features HBAO & DOF, [H] found 30% perf hit), Project Cars, Witcher 3 with HairWorks (without Tessellation override).

When the amount of games widen, so does the lead Fury has on 980 as well as 390X.

No offense, but Gameworks is a real threat. You can't just go "if you remove Gamework games, well AMD does better" because frankly put Nvidia is snagging some of THE must play games. We can't just keep removing them from benches because face it, gamers are gonna want to play those games WITH the bells and whistles. This isn't like PhysX where it is some randomly particles, these are more "visual" (at least from what I've seen in FFXIV DX11 Client).

If you're going to tell people "well you can turn that feature off" I get the feeling they'll ask you "why would I want to?"

And that is where it seems AMD continues to just fall on it's face. Which totally sucks, for them, and primarily for their users. When TressFX hits the scene and everyone is all "oooohh open standards" it gets picked up in LESS games with Hairworks (that name is terrible) and AMD users sit on the side lines going "well you can turn it off."

[Super drunk, spell check for the win, venting as a long time AMD user. Off to dinner <3]
 

RaistlinZ

Diamond Member
Oct 15, 2001
7,629
10
91
When you factor in overclocking, I think a 980 is still more compelling at $499.99 than a Fury at $549.99.

Performance will be more even between the two, the 980 will play well in Gameworks games, it'll use less power, and it'll save you $50.00.
 

VulgarDisplay

Diamond Member
Apr 3, 2009
6,193
2
76
When you factor in overclocking, I think a 980 is still more compelling at $499.99 than a Fury at $549.99.

Performance will be more even between the two, the 980 will play well in Gameworks games, it'll use less power, and it'll save you $50.00.
The 980 power consumption will explode when overclocked. Also people are touting 980 ocing like its a piece of cake to hit 1500mhz on every single gpu.

1300mhz is probably a safer bet with golden chips hitting higher, and 1500mhz requiring stupid amounts of voltage. I am talking stable overclock s here not what forum attention seekers call stable where they can run a game long enough to complete a test.
 

n0x1ous

Platinum Member
Sep 9, 2010
2,572
248
106
With 1440p 144hz, you can't do it on a single GPU you need two at the least. Similar to 4K at 60hz.

So basically single GPU users are strictly 1080p 60hz or 1440p 60hz with toned down settings.
in BF4 all Ultra settings I'm maintaining locked 96hz @ 1440p on my Korean with a single 980ti OC to 1400. the difference from 96 to 60 is massive and more noticeable than going from 96 to 120/144. 1440p and high refresh with max settings is doable on 980ti.
 

shady28

Platinum Member
Apr 11, 2004
2,520
397
126
The 980 power consumption will explode when overclocked. Also people are touting 980 ocing like its a piece of cake to hit 1500mhz on every single gpu.

1300mhz is probably a safer bet with golden chips hitting higher, and 1500mhz requiring stupid amounts of voltage. I am talking stable overclock s here not what forum attention seekers call stable where they can run a game long enough to complete a test.

1500mhz is like 35% over stock.

No one is talking about a 35% OC, they're talking mostly about AIB cards that come out of the box with a 5-20% OC and have the potential for more.

It basically makes the top range of OC AIB GTX 980s air-cooled cards like the Classifed equal to the Fury at 1080p and 1440p.

But they cost more than the lesser 980s (running $530-$560 mostly).

I think the Fury is a little bit better even vs the OC 980s, as in it still wins on 4K. But then it also takes more power, makes more heat, etc.

So Fury kinda winds up being a big 'meh'.

This is how various 980s stack up. Notice how much faster they are than stock GTX 980. Consider this when you see comparisons against stock 980s - cause not much of anyone who is buying a 980 buys a stock clock 980.

 
Feb 19, 2009
10,457
10
76
in BF4 all Ultra settings I'm maintaining locked 96hz @ 1440p on my Korean with a single 980ti OC to 1400. the difference from 96 to 60 is massive and more noticeable than going from 96 to 120/144. 1440p and high refresh with max settings is doable on 980ti.

Sure, some games especially older ones will easily hit >60 fps all the time at 1440p on a single top GPU. Newer titles, nope.

@Railven
GameWorks is a real threat, I've always said that. As for actual HairWorks, it ruins performance for little visual gains so even NV users run with it off, I noted that in the steam forums.



In Dying Light, disabling NV DoF, AMD GPUs run 33% faster. It's a no brainer for AMD gamers to turn that effect off. Project Cars just suck on AMD regardless. Just look at the [H] review where they stack GW titles and GW features to gimp AMD. AMD would actually need to be 25-50% faster and more efficient overall just to compete in such a line-up.

Basically people who use AMD have to play most GW titles a few months after release (once its officially patched up & AMD improve their drivers) to get decent performance (look at FC4, ACU etc now). If you're not OK with that, go with NV GPUs. That's the bottom line.

I'm OK with that personally because A) I rarely play titles on launch, I buy it on a steam sale and B) I'm NOT OK with anti-competitive closed source & code obfuscated GameWorks. But I understand people have different views. The way I see it, AMD is more than competitive because GW features are OFF in my gameplay.

Funny enough, Batman: AK which has 4-5 GW features, seems to run just fine on AMD, not sure what happened there!!

 

syngyne

Junior Member
Jun 22, 2015
8
0
6
I have to say, that power consumption of Asus Strix Fury is pretty much impressive.

I think it is possible to squeeze that GPU to 125W and put into Mac Pro.

I want to take that Asus Strix card and cram the Sapphire cooler onto it.
 

boozzer

Golden Member
Jan 12, 2012
1,549
18
81
well, that is that. this gen of cards are not worth it. I guess my 290 will survive another year or 2. my 50% increased performance bar was not reached.
 
Feb 19, 2009
10,457
10
76
well, that is that. this gen of cards are not worth it. I guess my 290 will survive another year or 2. my 50% increased performance bar was not reached.

Yeah and in AUS, our prices are ridiculous due to the falling exchange rate. Used to be 1:1 and now its 0.73 USD: 1 AUD... 980Ti are $1,059-1,159 here, Fury X is $999.

The 980 is ~$750-800 and 390 is at $499. I got my R290s for less than that at release!!

Tempting to play with new hardware but makes no sense.
 
Feb 19, 2009
10,457
10
76
This is how various 980s stack up. Notice how much faster they are than stock GTX 980. Consider this when you see comparisons against stock 980s - cause not much of anyone who is buying a 980 buys a stock clock 980.


52 -> 57 (avg of those factory OC models) is 10%.
 

shady28

Platinum Member
Apr 11, 2004
2,520
397
126
52 -> 57 (avg of those factory OC models) is 10%.

Which is enough to match them up fairly equally against fury in sub 4k resolutions.



So you can get one of the good AIB OC'd 980s and match up against Fury at sub 4K resolutions, but be slower at 4k.

Or, you can get Fury and take up more power / make more heat but be faster at 4k.

Or, you can get one of the slower 980s at ~ $480 and bank $70, and probably not be able to tell the difference if you don't run 4k.

Fury does seem marginally better from a performance standpoint with the 4k advantage, but it didn't exactly break new ground. It slots into the market like a high-end GTX 980 but has better 4k performance.
 

monstercameron

Diamond Member
Feb 12, 2013
3,818
1
0
Sure, some games especially older ones will easily hit >60 fps all the time at 1440p on a single top GPU. Newer titles, nope.

@Railven
GameWorks is a real threat, I've always said that. As for actual HairWorks, it ruins performance for little visual gains so even NV users run with it off, I noted that in the steam forums.



In Dying Light, disabling NV DoF, AMD GPUs run 33% faster. It's a no brainer for AMD gamers to turn that effect off. Project Cars just suck on AMD regardless. Just look at the [H] review where they stack GW titles and GW features to gimp AMD. AMD would actually need to be 25-50% faster and more efficient overall just to compete in such a line-up.

Basically people who use AMD have to play most GW titles a few months after release (once its officially patched up & AMD improve their drivers) to get decent performance (look at FC4, ACU etc now). If you're not OK with that, go with NV GPUs. That's the bottom line.

I'm OK with that personally because A) I rarely play titles on launch, I buy it on a steam sale and B) I'm NOT OK with anti-competitive closed source & code obfuscated GameWorks. But I understand people have different views. The way I see it, AMD is more than competitive because GW features are OFF in my gameplay.

Funny enough, Batman: AK which has 4-5 GW features, seems to run just fine on AMD, not sure what happened there!!

Maybe amds dx10 driver support is better than nvidias?
 

monstercameron

Diamond Member
Feb 12, 2013
3,818
1
0
Which is enough to match them up fairly equally against fury in sub 4k resolutions.



So you can get one of the good AIB OC'd 980s and match up against Fury at sub 4K resolutions, but be slower at 4k.

Or, you can get Fury and take up more power / make more heat but be faster at 4k.

Or, you can get one of the slower 980s at ~ $480 and bank $70, and probably not be able to tell the difference if you don't run 4k.

Fury does seem marginally better from a performance standpoint with the 4k advantage, but it didn't exactly break new ground. It slots into the market like a high-end GTX 980 but has better 4k performance.
On look at you trying to keep the more power\heat meme alive. Good try!

So will the factory oc 980s use the same power as the throttling reference 980s?
 

iiiankiii

Senior member
Apr 4, 2008
759
47
91
Which is enough to match them up fairly equally against fury in sub 4k resolutions.



So you can get one of the good AIB OC'd 980s and match up against Fury at sub 4K resolutions, but be slower at 4k.

Or, you can get Fury and take up more power / make more heat but be faster at 4k.

Or, you can get one of the slower 980s at ~ $480 and bank $70, and probably not be able to tell the difference if you don't run 4k.

Fury does seem marginally better from a performance standpoint with the 4k advantage, but it didn't exactly break new ground. It slots into the market like a high-end GTX 980 but has better 4k performance.

An AIB OC GTX 980 will consume more power than the reference one.

Example: Look below. The difference is 18watts on average.... no biggie



 

3DVagabond

Lifer
Aug 10, 2009
11,951
204
106
I see it the same way you do, I don't understand why everyone is jumping up and down saying how great this card is. Wooooooopy! it can beat a year old card and costs more ,hurray! I guess at this point anythings better than those 290 class cards.:thumbsup:

Come on now, happy medium.

Hawaii is now killing big Kepler, so no not anything is better than the 290 class cards. Just like Tahiti killed the GK104. Pretty bad if you feel like you have to exaggerate the age of the 980 to make AMD sound bad. Fury is going to end up faster than big Maxwell too. Mark my words.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |