ASUS Radeon R9 Fury STRIX 4 GB

[Casecutter]

Nice review!

Hey, @Casecutter , that DC3 cooler looks pretty damn similar to the that found on the 980 Ti doesn't it?

Figured we'd get back to this...
Well they made some changes to the side rail(s), looks like they were able to revise it to back-fit the 980 Ti. Although, they have a custom board and they lengthen from the chip to the I/O bracket, that spacing from the FuryX PCB was my point. While Asus almost appears to have brought in the spacing for their 4 mount holes to the bracket/frame in ever so slightly, verses FuryX ... but maybe not. But in that lies their folly, to use that cooler they made a new PCB, and now they have an expensive product that doesn't make substantial use of any of it. Appears a bad concession.

 

[Basard]

What is up with that thermal crud? Looks like some crap you'd find in an e-machine!

Can you please explain, w1zzard?

 

[Joss]

I'd still buy the 980 simply because... I stopped trusting AMD. Too many mistakes, flaws, misses, pure stupidity.
Trust is easy to loose and difficult to regain, and I shouldn't be the only one in this boat.

 

[majevica]

I'd still buy the 980 simply because... I stopped trusting AMD. Too many mistakes, flaws, misses, pure stupidity.
Trust is easy to loose and difficult to regain, and I shouldn't be the only one in this boat.

You mean like Nvidia with their ultimate fuck you to customers with their faulty laptop chips that rendered my 1500e laptop useless after a year and in my country they didn't offered extended warranty or anything.

 

[nem]

the nvfangilrls looks jealous

meanwhile AMD fanS be like

 

[HumanSmoke]

Figured we'd get back to this...
Well they made some changes to the side rail(s), looks like they were able to revise it to back-fit the 980 Ti.

GPU packages (die + substrate + support bracing) and mounting points differ, thus the "side rails" ( hold down/retention bracket) differ for different GPUs using the same cooler. Even small differences in die size necessitate either a change in mounting hardware, or a multi-fit approach. Good luck trying to fit an AMD mount (53.2mm x 53.2mm) to an Nvidia GF100/GF110/GK104/GK110/GM204/GM200 (58.4mm x 58.4mm) without a multi-option mounting plate.

 

[happita]

Really? If people are hung up on the fact that W1zz doesn't include another aftermarket card to compare with the current card being reviewed then you guys are just being lazy. It isn't that hard to just look up the different ASUS STRIX cards (be it a 970, 980, etc.), take the temps, OCability, power consumption and compare it to another STRIX card that you're interested in. No need to put more work on the table where it isn't needed.

 

[RejZoR]

Then again, why not do that in the review itself and be done with it? Why should user/consumer chase data around?

 

[happita]

Then again, why not do that in the review itself and be done with it? Why should user/consumer chase data around?

Reviewing a card has always been like this on TPU. The card being reviewed is the star of the show, whatever manufacturer/brand/aftermarket card that might be, against ALL other STOCK REFERENCE cards out there. Now, whether that dynamic needs to be challenged, is up to the people of TPU and it is obviously W1zz's decision.

 

[Aquinus]

If only it cost less.

Side note: @W1zzard , can we stop calling no VGA signal on DVI a con? Do you really think anyone is buying a Fury to use a VGA display on it? Honestly, I thought I was late to the game switching from VGA to DVI in 2007.

 

[rubenclavs]

Strix 980 vs Strix Fury

I gathered this from the reviews here in TPU. The 980 used 1600p and Fury used 1440p so I skipped it and used the common benchmarks for 1080p instead.

All the common games on both cards below.

Batman Arkham
Radeon Fury 255.8
GeForce 980 176.9

Battlefield 3
Radeon Fury 141.6
GeForce 980 137

Battlefield 4
Radeon Fury 87.3
GeForce 980 90.8

Bioshock Infinite
Radeon Fury 181.3
GeForce 980 156.7

Crysis 3
Radeon Fury 54.3
GeForce 980 49.6

Metro Last Light
Radeon Fury 97.7
GeForce 980 92.9

Tomb Raider
Radeon Fury 78.3
GeForce 980 70.5

Watch Dogs
Radeon Fury 85.4
GeForce 980 94.6

Wolfenstein
Radeon Fury 84.2
GeForce 980 104.6

Total Average:

Fury - 118.4333333
980 - 108.1777778

Cheers!

 

[Steevo]

Great review, and perhaps with voltage control software we will get a higher overclock, but this is pretty nice performance and the power consumption numbers are in line with what they should be.

 

[Fluffmeister]

Strix 980 vs Strix Fury

I gathered this from the reviews here in TPU. The 980 used 1600p and Fury used 1440p so I skipped it and used the common benchmarks for 1080p instead.

All the common games on both cards below.

Batman Arkham
Radeon Fury 255.8
GeForce 980 176.9

Battlefield 3
Radeon Fury 141.6
GeForce 980 137

Battlefield 4
Radeon Fury 87.3
GeForce 980 90.8

Bioshock Infinite
Radeon Fury 181.3
GeForce 980 156.7

Crysis 3
Radeon Fury 54.3
GeForce 980 49.6

Metro Last Light
Radeon Fury 97.7
GeForce 980 92.9

Tomb Raider
Radeon Fury 78.3
GeForce 980 70.5

Watch Dogs
Radeon Fury 85.4
GeForce 980 94.6

Wolfenstein
Radeon Fury 84.2
GeForce 980 104.6

Total Average:

Fury - 118.4333333
980 - 108.1777778

Cheers!

Good work, damn that tiny GM204 is a monster, gotta love those margins.

 

[Initialised]

OK so where's the softmod to Fury-X for those who want it watercooled but don't want the noisy CM pump?

 

[HumanSmoke]

Strix 980 vs Strix Fury
I gathered this from the reviews here in TPU. The 980 used 1600p and Fury used 1440p so I skipped it and used the common benchmarks for 1080p instead.

Unfortunately you aren't taking into account game patches and driver revisions. Comparing one review from October 2014 to one from June 2015 is a flawed exercise at best.
Case in point: The Strix 980 review had the stock 980 at 169.3 f.p.s. in Batman:AO at 1080p. The current review has the same card producing 231.8 f.p.s.

 

[xvi]

OK so where's the softmod to Fury-X for those who want it watercooled but don't want the noisy CM pump?

Patience, young grasshopper.

 

[fullinfusion]

What's that horrible thing in page 5 photo #2, could that be thermal paste? it looks like scratched metal.

View attachment 66393

@W1zzard , yeah whats that, a pad?

And I'm sure you bench the card before pulling it apart correct?

 

[btarunr]

And I'm sure you bench the card before pulling it apart correct?

That is correct.

 

[dwade]

Overclock vs overclock, my money's on the 980.

 

[newtekie1]

Figured we'd get back to this...
Well they made some changes to the side rail(s), looks like they were able to revise it to back-fit the 980 Ti. Although, they have a custom board and they lengthen from the chip to the I/O bracket, that spacing from the FuryX PCB was my point. While Asus almost appears to have brought in the spacing for their 4 mount holes to the bracket/frame in ever so slightly, verses FuryX ... but maybe not. But in that lies their folly, to use that cooler they made a new PCB, and now they have an expensive product that doesn't make substantial use of any of it. Appears a bad concession.

All Fury cards will be custom PCBs, there is no reference PCB for Fury. And the Fury X PCBs are all manufactured by AMD(Sapphire actually) and given to the other companies for distribution, so they can't even use that PCB for their Fury cards. Well, Sapphire can.

 

[buggalugs]

Reviewing a card has always been like this on TPU. The card being reviewed is the star of the show, whatever manufacturer/brand/aftermarket card that might be, against ALL other STOCK REFERENCE cards out there. Now, whether that dynamic needs to be challenged, is up to the people of TPU and it is obviously W1zz's decision.

Exactly, its funny how guys are complaining about this now, its always been the case and its usually AMD that suffered worse from this. Now that Nvidia looks worse some people are upset about it. A perfect example is the 290X, which was criticized for heat and noise based on the early reference blower cooler. Those opinions and early benchmarks blighted the card but when the non-reference versions came out, it was a different card. An Asus DirectCUII 290X is silent, not audible during gaming and gaming temps are very low in the 60's-70's. Most of the non-reference cards also had a decent overclock, so it was a much better performer all around in benchmarks than the early reference designs indicated.

Same can be said for most of the 290X non reference versions from Sapphire, MSI, Gigabyte etc, but there are still guys(and benchmarks) that judge it based on the reference.

Its not the reviewers fault, there are so many options a reviewer cant test them all but people should be aware of it.

The other funny thing is how people judge cards based on performance differences of 3% or 5% or 7%. I guarantee most people would not notice a performance difference that small in the real world.

 

[NC37]

Price is too high. If I were that close to a reference 980 Ti ($70ish), I'd go on and finish my way there and get something far more capable for 1440p or greater. Otherwise, I'd drop down to $500 or less (980, R9 390X) for 1080p.

AMD has a real problem with pricing.

You said it. Probably one of the biggest disappointments with Fury to me has been the price. AMD really could have had a good price/performance leader but they decided to pull a nVidia. Heck even $50 less would have done AMD some good.

 

[Caring1]

This. Strix is an aftermarket enthusiast series. Comparing it to stock reference models is silly. Strix R9 Fury should only be compared to Strix GTX 980 (despite the fact that R9 Fury is using cooler found on GTX 980Ti). Only this way it's apples vs apples.

The only difference from reference is the cooler, the clocks are the same.

 

[fullinfusion]

That is correct.

Yeah cool, reason I asked is the thermal pad or whatever they used looked, well, like a pad, I dont think the temperatures would be as good otherwise..

And for your info do you know what pad that is?
I really think that's the reason for such cooler temp's

 

[newtekie1]

And for your info do you know what pad that is?
I really think that's the reason for such cooler temp's

I looks to me like it is a Liquid Metal pad, it's better than most thermal pastes.