R9 Fury Versus GTX 980 Revisited

ultima_trev

Member
Nov 4, 2015
148
66
66
It's been nearly 10 months since Fiji's NDA lift. At the time performance between these two was close however due to the $50 premium that the R9 Fury commanded, GTX 980 was generally seen as the better buy. Flashforward, we have seen price cuts for both and new drivers that should boost Fury quite a bit, albeit I haven't seen any reviews in recent days with these current drivers or prices in mind. However, in newer game previews, the Fury tends to come out ahead. Credit to Guru of 3D :

http://www.guru3d.com/articles_pages/hitman_2016_pc_graphics_performance_benchmark_review,6.html

http://www.guru3d.com/articles_pages/the_division_pc_graphics_performance_benchmark_review,6.html

http://www.guru3d.com/articles_pages/far_cry_primal_pc_graphics_performance_benchmark_review,7.html

Not sure what boost R9 Fury sees in older games, if any, but here it seems to edge out a win from the GTX 980 at all resolutions. Anyone know of recent reviews with a more comprehensive battery of tests?
 

tential

Diamond Member
May 13, 2008
7,348
642
121
index.php

index.php

index.php
 

BigDaveX

Senior member
Jun 12, 2014
440
216
116
While Hitman seems to really favour the AMD cards, in the other two games the Fury's only slightly ahead of the 980, and at the cost of being more expensive and having much higher power consumption.

If anything, the real problem for the Fury is actually another AMD card, the 390X, which is much cheaper, has more memory and lower power consumption, and is generally around the same speed as the Fury.
 

96Firebird

Diamond Member
Nov 8, 2010
5,738
334
126
What timing; this 980 is being sold for $400. Not a great cooler, but maybe the better 980s will see price cuts in the coming months.
 

tential

Diamond Member
May 13, 2008
7,348
642
121
While Hitman seems to really favour the AMD cards, in the other two games the Fury's only slightly ahead of the 980, and at the cost of being more expensive and having much higher power consumption.

If anything, the real problem for the Fury is actually another AMD card, the 390X, which is much cheaper, has more memory andlower power consumption, and is generally around the same speed as the Fury.

1436520543zZMsl7GpwE_8_1.gif

14359838676XjxIP4bGb_6_1.gif

The Fury uses slightly more power than the GTX 980 and is slightly faster....

It helps to use graphs/quote things when you make claims so people can understand where you're coming from....
 

BigDaveX

Senior member
Jun 12, 2014
440
216
116
The Fury uses slightly more power than the GTX 980 and is slightly faster....

It helps to use graphs/quote things when you make claims so people can understand where you're coming from....

I was going by AT's own review of the Fury cards, which shows a more pronounced difference between the 980 and most of the Fury series. The STRIX isn't actually too bad from a power consumption standpoint - though it does seem to be one of the more expensive Fury cards, at least in terms of UK pricing - but the rest of the Fury line-up is a different story.
 

Rvenger

Elite Member <br> Super Moderator <br> Video Cards
Apr 6, 2004
6,283
5
81
The Nano is where its at right now. Vanilla fury should be no more than $400.
 

SimianR

Senior member
Mar 10, 2011
609
16
81
Fiji is easily one of the biggest GPU failure of all time.

I think that might be an exaggeration. It's still a decent GPU, I think its launch price was way off, it should have been priced similar to the 290/290X IMHO, $400 for the Fury and $549 for the Fury X. It does feel like Fiji was a test run for HBM though rather than an high-end competitor to the 980 Ti / Titan X.
 

Raising

Member
Mar 12, 2016
120
0
16
Now overclock the 980ti and watch the Fury X eating dust..

Even in Hitman 63fps+ at least 20%oc = 75fps~
 

linkgoron

Platinum Member
Mar 9, 2005
2,598
1,238
136
Fiji is easily one of the biggest GPU failure of all time.

Easily? probably not. Could you compare it to the FX series? The Fermi launch was also very late and wasn't exactly a perf/watt monster.

The only problem with Fury was the price. The Fury X beats the stock 980ti at 4k (while overclocked 980ti are much better obviously). The fury basically beats the 980 with basically the same perf/w. The Nano also has good perf/watt.

I agree that it was disappointing and underwhelming though.
 
Last edited:

DrBombcrater

Member
Nov 16, 2007
38
0
61
The Nano is where its at right now. Vanilla fury should be no more than $400.
So much this. I'm surprised the Nano doesn't get more love for its great value. The price cut has made the Fury (and the 980 for that matter) look significantly overpriced.
 

poofyhairguy

Lifer
Nov 20, 2005
14,612
318
126
For being a member since 2005 you don't seem to have a very good memory.

I can't remember a GPU so snake bit. First it was the overclockers dream then it wasn't. Then the Fury X had problems with noise. The Fury Nano supposedly has that loud coil whine. The Fury non X was priced wrong. The Fury X2 model has been delayed and delayed. Just issue after issue.

I mean something like the Geforce FX 5800 Ultra was just terrible from the start, it was a bad design that was behind the times. Fiji could have been something much better and to me that is a bigger fail.

Can you imagine how different it would have been if the Fury X2 would have launched first? All that Fiji hype would have been focused on the excellent Crossfire performance of the card, which is much better than the 980 ti's SLI. The hype train would have kept rolling too, as the Fury X2 could neither confirm nor deny overclocking potential or exactly how a single Fury X would match up against a 980 ti. People would have been extrapolating off of the crossfire numbers, which would have left a lot of ambiguity for a while which could have helped AMD sell more cards.

Plus any issues with the AIO cooler would have been settled on a unit less people would have bought. By the time the regular Fury X is released it could have had a fixed cooler, plus people who buy multiple high-end GPUs would have loaded up on Furys instead of 980 tis because that X2 would have shown the potential. I mean eventually the lack of overclocking, or the fact a single Fury X loses to a 980 ti, would have been revealed. But AMD could have sold a lot of Fiji before the hype bubble burst.

Instead it was a lot of people who waited for the Fury only for the X to let them down, and Nvidia sold a lot of 980 tis instead.
 

railven

Diamond Member
Mar 25, 2010
6,604
561
126
Fiji was definitely a disappointment across the board for me.

Hawaii/Grenada below it and GM200 above it left Fiji struggling to find a place of it's own.

I'd take a 390X over a Fury, personally. (Actually I took a refurb 290X for $160, and slapped a water cooler on it for $20. Thing is silent and powerful! (RIP power bill, but its finally starting to get warm again :D).
 

Glo.

Diamond Member
Apr 25, 2015
5,930
4,991
136
Now overclock the 980ti and watch the Fury X eating dust..

Even in Hitman 63fps+ at least 20%oc = 75fps~

How do you know that if we would overclock Fury X it would not affect its performance in even higher way?

We did not see any tests on that in DX12, so far.
 

ElFenix

Elite Member
Super Moderator
Mar 20, 2000
102,393
8,552
126
I think that might be an exaggeration. It's still a decent GPU, I think its launch price was way off, it should have been priced similar to the 290/290X IMHO, $400 for the Fury and $549 for the Fury X. It does feel like Fiji was a test run for HBM though rather than an high-end competitor to the 980 Ti / Titan X.

that's been my impression of it since pretty much day 1. test part to make sure HBM works, with not a lot of volume actually planned.

although, if nvidia didn't seem to know exactly what amd was launching and to drop the real 980 3 weeks beforehand it would have been at/near the top on every list compared to the 970ti nvidia was selling, so it wouldn't matter that it was pricey. as it were, nvidia played it perfectly and so it ended up being bad optics for amd and yet another hit to its mindshare.

also, amd needs to tell its marketing people to shut up.



edit: and as i remember from launch day it seemed like power consumption figures were all over the place. toms with its high resolution monitoring seems like it has the best methodology. and since your power bill and the ability of the card to warm the immediate area are based on total kw-hr (iow, joules of energy) consumed, rather than peak draw, that's what we all should be most interested in.
 
Last edited: