AMD Radeon R9 Fury X Video Card Review @ [H]

Ugh, I'm a bit disappointed in the performance but I do appreciate the nice tidy and quiet package. Hopefully drivers can help the performance a bit.

Overall I wonder if they're making much money per unit considering this is a flagship card but it has a lot of low volume tech in it.
 
bah what a disappointment. this helps no one, from a competitive market standpoint.
 
Well, I guess that settles my decision. Now I just need to decide which 980ti to get. Like Brent said, if this card had been priced at $550, I may have actually been interested.
 
Ronin said:
Brent / Kyle...

http://www.tomshardware.com/reviews/amd-radeon-r9-fury-x,4196-5.html

http://www.tomshardware.com/reviews/amd-radeon-r9-fury-x,4196-4.html

http://www.tomshardware.com/reviews/amd-radeon-r9-fury-x,4196-6.html

Here the "picture" is slightly different... :confused:

Any comments ?
Click to expand...

I have no idea how they test and cannot verify their methods or similarities or differences.

I can however tell you how we test. I play each game, hands on, for hours at a time, find playable settings, and record fraps data. For ap2ap I run the games at the same settings on each card, each run-through is live, dynamic, and as real-world as you can get because it is just me, a gamer, playing the game as the game is meant to be played.
 
Ronin said:
Brent / Kyle...

http://www.tomshardware.com/reviews/amd-radeon-r9-fury-x,4196-5.html

http://www.tomshardware.com/reviews/amd-radeon-r9-fury-x,4196-4.html

http://www.tomshardware.com/reviews/amd-radeon-r9-fury-x,4196-6.html

Here the "picture" is slightly different... :confused:

Any comments ?
Click to expand...

Probably stuck to "suggestions" in the reviewer's guide? Ie: go to this part of the game, at these settings at this resolution and performance is x compared to it being y here. Which would account for the "internal benchmarks" amd released showing the fury to be consistently faster than the ti.

Surprised at this point they don't email out save game and config files with specific areas and settings to be tested.
 
Brent_Justice said:
http://www.hardocp.com/article/2015/06/24/amd_radeon_r9_fury_x_video_card_review/6
Click to expand...

It's really interesting the gap is much less at 4k % wise. Was something turned down where there was less of a VRAM bottleneck or is the 980ti bottlenecked as well?

980ti 40% faster at 1440p
980ti only 13% faster at 4k

I'm assume you turned down shadow maps or distance at 4k. In the 980ti review the 980ti averaged 40 FPS where here it's averaging 45 (could be drivers too.) In the 980ti review the Titan X was only 5% faster. Doesn't look like the 980ti is bottlenecking...
 
Ronin said:
Brent / Kyle...

http://www.tomshardware.com/reviews/amd-radeon-r9-fury-x,4196-5.html

http://www.tomshardware.com/reviews/amd-radeon-r9-fury-x,4196-4.html

http://www.tomshardware.com/reviews/amd-radeon-r9-fury-x,4196-6.html

Here the "picture" is slightly different... :confused:

Any comments ?
Click to expand...

You are correct. THG has been confusing me for years too! Seriously, I have no comment about Tom's Hardware testing.

All I can say to you is that HardOCP uses REAL WORLD GAMEPLAY when we test GPUs, not canned benchmarks. We also do NOT reuse old results from previous reviews with outdated drivers. If you play canned benchmarks with outdated drivers rather than real games with up-to-date drivers, you might look elsewhere for your performance metrics.
 
Seems frame times are not good on the Fury, in some of TechReport's tests it's doing worse than a 980 despite higher raw FPS. AMD has some serious work to do on the drivers


http://techreport.com/review/28513/amd-radeon-r9-fury-x-graphics-card-reviewed/6
http://techreport.com/review/28513/amd-radeon-r9-fury-x-graphics-card-reviewed/12
http://techreport.com/review/28513/amd-radeon-r9-fury-x-graphics-card-reviewed/5

Conclusion:

Speaking of which, if you dig deeper using our frame-time-focused performance metrics—or just flip over to the 99th-percentile scatter plot above—you'll find that the Fury X struggles to live up to its considerable potential. Unfortunate slowdowns in games like The Witcher 3 and Far Cry 4 drag the Fury X's overall score below that of the less expensive GeForce GTX 980. What's important to note in this context is that these scores aren't just numbers. They mean that you'll generally experience smoother gameplay in 4K with a $499 GeForce GTX 980 than with a $649 Fury X. Our seat-of-the-pants impressions while play-testing confirm it. The good news is that we've seen AMD fix problems like these in the past with driver updates, and I don't doubt that's a possibility in this case. There's much work to be done, though.
Click to expand...
 
Last edited:
Ronin said:
Brent / Kyle...

http://www.tomshardware.com/reviews/amd-radeon-r9-fury-x,4196-5.html

http://www.tomshardware.com/reviews/amd-radeon-r9-fury-x,4196-4.html

http://www.tomshardware.com/reviews/amd-radeon-r9-fury-x,4196-6.html

Here the "picture" is slightly different... :confused:

Any comments ?
Click to expand...

All I can say is Tom's is the only one having different results. Hexus, PC Gamer and others seem to be coming to the same closure as [H]. It might be better to ask Tom's how they reached such different results.
 
I completely agree this should have been done for a mid range card. If even two cut-down versions were selling at $250 and $400 it would be one hell of a deal. NVIDIA would have crapped their pants.

Here's what gets me: did no one at AMD benchmark the 980Ti? It isn't an unknown quantity. Someone has to "uh...guys...we're getting spanked. We need to change something."

I was rooting for AMD this time around but it looks like a $650 farce. Brent, kudos on the great review. Much appreciated!
 
Dayaks said:
It's really interesting the gap is much less at 4k % wise. Was something turned down where there was less of a VRAM bottleneck or is the 980ti bottlenecked as well?

980ti 40% faster at 1440p
980ti only 13% faster at 4k

I'm assume you turned down shadow maps or distance at 4k. In the 980ti review the 980ti averaged 40 FPS where here it's averaging 45 (could be drivers too.) In the 980ti review the Titan X was only 5% faster. Doesn't look like the 980ti is bottlenecking...
Click to expand...

The 4K results in Dying Light are at the "Best Quality" setting, so yes the shadow map size is turned down to "High" and the view distance is at the half-way slider.

Whereas the ap2ap for 1440p is with maximum game settings, shadow map at "Very High" with view distance slider maxed out.

Mainly because it was very very difficult to get a run-through on the 290X at the highest possible settings in Dying LIght at 4K, it was physically hard to do the run-through cause the framerates were so low, so I chose a more sane setting.
 
ElBurroDeMuerte said:
All I can say is Tom's is the only one having different results. Hexus, PC Gamer and others seem to be coming to the same closure as [H]. It might be better to ask Tom's how they reached such different results.
Click to expand...


From their testing page. Older reused benchmarks with out of date drivers.

Software And Drivers

As with our GeForce GTX 980 Ti, many benchmark results carry over from previous stories. Nvidia’s GeForce GTX Titan X, 980, Titan, and 780 Ti are predominantly tested using driver build 347.84. The GeForce GTX 980 Ti is benchmarked with 352.90. Similarly, the results for all Nvidia cards in Grand Theft Auto V and The Witcher 3: Wild Hunt employ 352.90.
Click to expand...

Here is their tests outlined.

Middle-earth: Shadow of Mordor
Built-in benchmark, 40-sec Fraps, Ultra preset
Battlefield 4
Custom THG Benchmark, 100-sec Fraps, Ultra preset
Metro Last Light
Built-in benchmark, 145-sec Fraps, Very High preset, 16x AF, Normal motion blur
Thief Version 1.7, Built-in benchmark, 70-sec Fraps, Very High preset
Tomb Raider
Version 1.01.748.0, Custom THG Benchmark, 40-sec Fraps, Ultimate preset
Far Cry 4
Version 1.9.0, Custom THG benchmark, 60-sec Fraps, Ultra preset
Grand Theft Auto V
Build 350, Online 1.26, In-game benchmark sequence #5, 110-sec Fraps, FXAA: On, MSAA: 2x, Texture Quality: Very High, Shader Quality: Very High, Shadow Quality: High, Reflection Quality: Very High, Water Quality: High, Particles Quality: Very High, Grass Quality: High, Soft Shadows: Softer, Post FX: Very High, Anisotropic Filtering: 16x
The Witcher 3: Wild Hunt Version 1.03, Custom THG Benchmark, 110-sec Fraps, Post-processing Preset: High, Graphics Preset: Ultra, HairWorks: Off
Click to expand...

You will see a bunch of canned benchmarks, and even their "custom" FRAPS benchmarks may only be 1 or 2 minutes of actual gameplay, which of course is not enough to give you a true look at "gaming."


That all said, let's move on from discussing THG here. If you want to start a new thread please do. Let's keep this thread on topic.
 
Ronin said:
Brent / Kyle...

http://www.tomshardware.com/reviews/amd-radeon-r9-fury-x,4196-5.html

http://www.tomshardware.com/reviews/amd-radeon-r9-fury-x,4196-4.html

http://www.tomshardware.com/reviews/amd-radeon-r9-fury-x,4196-6.html

Here the "picture" is slightly different... :confused:

Any comments ?
Click to expand...
Kyle laid it all out above. I'll just add this.

All of the tests from Tom's was done using some sort of benchmark or gameplay demo loop that is around a minute long. [H] sometimes shows framerates over a real gameplay session up to 30 minutes long.
 
Last edited:
PiSquared said:
Here's what gets me: did no one at AMD benchmark the 980Ti? It isn't an unknown quantity. Someone has to "uh...guys...we're getting spanked. We need to change something."
Click to expand...

By the time companies get to this point, they are way too far down the rabbit hole to get out. They will attempt to spin things as much as possible to save face and look for a way to make money off their investment.
 
pandora's box said:
Would be nice to see a follow up article that takes a 980 Ti and a Fury X, overclock them to the max and bench. Even better would be closed loop cooled 980 TI (EVGA) vs Fury X. That way we can see what these cards can do when both are cooled in the same manner.
Click to expand...

I don't see us covering Fury X overclocking till we see more controls unlocked that allow us more variables to play with. Overall we are only seeing a +10% clock bump, and quite frankly that is not going to show anything compared to 980 ti's overclock levels.

But yes, we are keeping our eye on it.
 
PiSquared said:
I completely agree this should have been done for a mid range card. If even two cut-down versions were selling at $250 and $400 it would be one hell of a deal. NVIDIA would have crapped their pants.

Here's what gets me: did no one at AMD benchmark the 980Ti? It isn't an unknown quantity. Someone has to "uh...guys...we're getting spanked. We need to change something."

I was rooting for AMD this time around but it looks like a $650 farce. Brent, kudos on the great review. Much appreciated!
Click to expand...

They did and included comparison numbers in the reviewers guide that showed Fury X as on avg 9% faster at 4k :p

The 980ti though was released just a few weeks ago, there is only minor adjustments that can be made in that short of a time frame.

If you mean the price some people already suspect 980ti capped Fiji at $650 and that was originally intended to released even higher. Also the other consideration is yes they could price lower but how much would that kill margins? How many extra sales would be needed to cover that deficit?

Also AMD certainly has tried before to be very aggressive with prices (don't really understand the viewpoint of AMDs pricing being more consumer friendly, I've brought this up before), see the FX-9590 as the most notable example.
 
PiSquared said:
I completely agree this should have been done for a mid range card. If even two cut-down versions were selling at $250 and $400 it would be one hell of a deal. NVIDIA would have crapped their pants.

Here's what gets me: did no one at AMD benchmark the 980Ti? It isn't an unknown quantity. Someone has to "uh...guys...we're getting spanked. We need to change something."

I was rooting for AMD this time around but it looks like a $650 farce. Brent, kudos on the great review. Much appreciated!
Click to expand...
You should read up a bit on GPU manufacturing. Even if AMD knew that the 980 Ti is faster, GPU architectures are planned and built literally years in advance. Even if there some was something they could go in and quickly change (unlikely), doing a respin of the GPU takes at least 3 months, and then they would need to redo QA, update drivers, etc. You'd be looking at them pushing the card out 6+ months at least. Once the GPU design is finalized, the only thing they can do is play with clock speeds and optimize the drivers. Failing that, you can cut prices, but AMD needs the money, and HBM is expensive.
 
Looking at other reviews it seems the tendency is the same.

I'm hoping it gets better with future drivers, as it bounces from 390x to 980Ti. At least is faster then the GTX980 for the most part.
 
Was hoping it was going to be at least faster then the titian x. For 650 water cooled it's not a bad deal if you want the cooler. I others have said if the can do a air cooled card at 550 they would be set. As low as the temps are on water and with the memory and gpu being so close together I don't see air cooling as an issue. Should be easier to build an air cooled card. Hell they could make the cooler huge and still only be as long as a common high end video card. I'm not really gaming that much any more so I'll wait till 16nm till I pick up something new.
 
Rizen said:
You should read up a bit on GPU manufacturing. Even if AMD knew that the 980 Ti is faster, GPU architectures are planned and built literally years in advance. Even if there some was something they could go in and quickly change (unlikely), doing a respin of the GPU takes at least 3 months, and then they would need to redo QA, update drivers, etc. You'd be looking at them pushing the card out 6+ months at least. Once the GPU design is finalized, the only thing they can do is play with clock speeds and optimize the drivers. Failing that, you can cut prices, but AMD needs the money, and HBM is expensive.
Click to expand...

If AMD needs the money, dropping the price is exactly what they should have done. With this launch, their customer base isn't going to change for the better, which obviously isn't good enough for their financial outlook. Only buyers will be the die hard AMD fanboys where there isn't enough of and even that number is falling. Lower profit margin with higher volume would have been better for them, at least they would have secured people on the fence and if they showed good support for the card moving forward, those are customers they could have retained for the next gen. As it stands now, all they've done is lost more of their already dwindling customer base.


AMD fans one claim that justified them being AMD fans was the "value proposition" and AMD has now taken that away. There's nothing left.
 
Brent_Justice said:
Our gaming suite has been recently all replaced, now with new games. These are new games, popular games, games people are playing now. My only regret was not having enough time to also include Batman or Project Cars.
Click to expand...

I'm really hoping more sites pick up PCars. Doubly so since it's the game I'm most interested in right now.

Are you planning on including it and just couldn't for this review, or will you guys be passing over it as well?
 
shad0w4life said:
It's crazy reading other reviews eg TPU has Civ V, Dead Rising 3 on it and shows Fury X trouncing 980Ti in those as two examples be it stock benchies but it's curious. And other ones had games like Bioshock and Metro doing well yada yada.

Someone give H Moar monies so we can get more games included. I play none of the ones used, where's the Skyrim & Civ V benchies :D

I'm not sure if it's as DOA as made out to be, the 980Ti compared to it is pumping out some room heating air, where's all the guys claiming to live in Florida now and needing the coldest possible card.
Click to expand...

Your post makes no sense. Why would [H] be testing a brand new GPU arch on old titles that existing mid-range hardware can run at hundreds of frames per second?
 
Yet they seem to keep selling out. Doom and gloom does not make reality. These will sell based on card size, HBM (new tech sells regardless of actual performance or price) , and simply the looks of the card.
 
Looking at reviews, why wouldn't one just buy a 295x2 for better performance than a 980 Ti or a Fury at the same price? I know it's older, but it appears to win every bench.
 
durquavian said:
Yet they seem to keep selling out. Doom and gloom does not make reality. These will sell based on card size, HBM (new tech sells regardless of actual performance or price) , and simply the looks of the card.
Click to expand...

How many were shipped? Because Unless you know the quantity shipped, "sold out" is a completely useless metric.


At least when the 980Ti was selling out, you had dozens of forum goers bragging they got their order in and many dozens more upset they didn't get their order in on time. You don't see that here.


Reality makes reality and doom and gloom is AMDs reality.
 
frunction said:
Looking at reviews, why wouldn't one just buy a 295x2 for better performance than a 980 Ti or a Fury at the same price? I know it's older, but it appears to win ever bench.
Click to expand...
Because dealing with CrossFire or SLI is inherently worse than a single card solution when they're the same performance (give or take a few %).
 
durquavian said:
Yet they seem to keep selling out. Doom and gloom does not make reality. These will sell based on card size, HBM (new tech sells regardless of actual performance or price) , and simply the looks of the card.
Click to expand...

Sure, performance, efficiency and value don't matter... :rolleyes:

*Newsflash* This video card isn't made by Apple.
 
RamonGTP said:
If AMD needs the money, dropping the price is exactly what they should have done. With this launch, their customer base isn't going to change for the better, which obviously isn't good enough for their financial outlook. Only buyers will be the die hard AMD fanboys where there isn't enough of and even that number is falling. Lower profit margin with higher volume would have been better for them, at least they would have secured people on the fence and if they showed good support for the card moving forward, those are customers they could have retained for the next gen. As it stands now, all they've done is lost more of their already dwindling customer base.


AMD fans one claim that justified them being AMD fans was the "value proposition" and AMD has now taken that away. There's nothing left.
Click to expand...
It's hard to say without knowing what the margins are on these chips or what the yields on Fiji are like.
 
I can't imagine yields are that horrible considering it's on the most mature manufacturing process in GPU history (that I'm aware of)


I'm sure HBM isn't cheap, but again, when all you're doing is losing customers because you've priced yourself out of competition, that can't be good for making money.
 
frunction said:
Looking at reviews, why wouldn't one just buy a 295x2 for better performance than a 980 Ti or a Fury at the same price? I know it's older, but it appears to win ever bench.
Click to expand...

295x2 isn't that far ahead of a 980 Ti and you give up a lot going with it. Higher power consumption, dual GPU issues, no HDMI 2.0 are the main things that come to mind. Going past 2 GPUs doesn't net good scaling either. Adding a second 980 Ti you're looking at around 70% more performance, but since the 295x2 is already dual GPU if you add a second you'll only get around 30% boost at best. You also need a fairly beefy PSU for a pair of 295x2... 1 kW wouldn't cut it for overclocking, whereas that would be fine for OC'd 980 Ti's.

I'd also venture to say, OC to OC the 980 Ti can probably match a 295x2. Hawaii doesn't overclock as well as Maxwell.
 
how disappointing. I was really hoping AMD would give us a competitive product. Instead we get a big fat flop.

The price needs to come down $100 for this to be an alluring product.
 
You must log in or register to reply here.
Back
Top