Fury X Crossfire Almost 100% scaling @ 4k !

[Nena Trinity]

So erm does anyone here wanna trade me a Fury X for my 970?

[p5ych00n5]

They do it to make the Fury X look stronger than the Nvidia cards. HBM makes the Fury X have insane bandwidth when compared to the Maxwell GPU's, so turning on copious amounts of AA will slow the Nvidia cards down in terms of FPS, while the Fury X and crunch through the extra AA with ease. It has little to no effect visually, because of the resolution being so crisp as is. It only skews the results even more. If you looked at the benchmarks AMD provided about the Fury X, it showed plenty of games with AA on max at 4k. The only time the Fury X was ahead of the 980 Ti, was with those games having high AA on. When they showed a game with AA turned off, the 980 Ti tied the Fury X (which is still different than the results posted by several reviewers).

 

If you see a test with 4k AND AA, just assume the person benchmarking the cards is clueless and move on. It really does nothing to show people how these cards will perform on average.

 

I'm seeing a lot of fanboying and toxicity on this site especially towards certain companies.

 

Not trying to flame but this is the only thread I've seen (and correct me if I'm wrong) which puts the FuryX on level territory or even beating the topend NV cards despite the multitude of ACTUAL real life instances, on these forums, it's only getting drowned out by the HuurDuur's..

 

This is not a rant directed at anyone personally but this I find distasteful and it's driving ACTUAL Tech literate peeps away from what was once a fair and balanced site..

 

Regards 

p5ych00n5

[ablangc]

I am posting these results at the request of a few who asked me to do this. @Whorax @LordSlimJimm @Grrizz I did this for me as well; however, I am not an expert when it comes to benchmarking. I am doing this just to see where the 980 Ti SLI  stands in regards to my rig setup.  These results are also based on a OVERCLOCKED CPU AND GPU.  These are NOT stock setting runs.  Ok, I ran these tests utilizing DSR on a 1080p Screen.  The OP website does not even state what settings they used on Bioshock Infinite and GTA V.   Just read the comments for each test.

 

Test System from Website: http://www.eteknix.c...ossfire-review/

• Motherboard – Gigabyte X99-Gaming G1 WiFi LGA 2011-3 Motherboard
• Processor – Intel Core i7 5820K at Stock 3.3GHz
• RAM – 16GB (4 X 4GB) Crucial Ballistix Sport DDR4 2400MHz
• CPU Cooler – Thermaltake Water 3.0 with Gelid GC-Extreme
• Power Supply – Be Quiet Dark Power Pro 11 1200W
• Main Storage Drive – Crucial M550 512GB
• Chassis – Lian Li T80 Test Bench
• Displays – AOC U2868PQU 4K
• Operating System – Windows 8.1 Pro 64 Bit

My Test System

• Motherboard – Z97 Deluxe
• Processor – Intel Core i7 4790k @ 4.7 GHz (Delid)
• RAM – 16GB (4 X 4GB) G.Skill Sport DDR3 2400MHz CAS 9
• CPU Cooler – Liqtech 240
• Power Supply – Corsair AX1200i
• Main Storage Drive – Samsung Evo 250GB
• Chassis – CM Storm Scout 2 Advanced
• Displays – Sharp 70 Inch TV 1080p (Using 4k DSR for testing purposes)
• Operating System – Windows 8.1 Pro 64 Bit
• EVGA GeForce GTX 980 Ti Superclocked+ ACX 2.0+ SLI with BIOS MOD
• NVIDIA Drivers 353.38 (Hotfix)

Firestrike Ultra:  Graphics drivers are not approved.  I have the 353.38 hotfix and not using 353.30

http://www.3dmark.com/fs/5271279

Score: 8905

Graphics Score: 9531

 

GTA V:

Everything is maxed out except I turned off AA.  The website never posted any type of settings so it is hard to gauge what settings they used.  So I am going to assume everything at max except AA.

Frames Per Second (Higher is better) Min, Max, Avg

Pass 0, 37.285763, 112.175400, 86.098907

Pass 1, 37.270332, 100.165657, 67.501480

Pass 2, 41.945621, 144.800674, 87.711945

Pass 3, 30.580616, 160.600525, 95.634476

Pass 4, 26.104675, 162.058365, 79.871712

Minimum: 26 FPS

Average: 83 FPS

 

Bioshock Infinite:

Everything on Ultra AA= ON, Post processing: Alternate (DX11)

Average FPS, Min FPS, Max FPS

125.08, 49.49, 324.62, Overall

 

Unigine Valley:  I will post shortly.  I am having problems with DSR on that one.

 

I would do stock runs but I really do not feel like flashing my bios back to its original state since I have spent a lot of time testing my rig for stable overclock settings. 

[Whorax]

I'm seeing a lot of fanboying and toxicity on this site especially towards certain companies.

 

Not trying to flame but this is the only thread I've seen (and correct me if I'm wrong) which puts the FuryX on level territory or even beating the topend NV cards despite the multitude of ACTUAL real life instances, on these forums, it's only getting drowned out by the HuurDuur's..

 

This is not a rant directed at anyone personally but this I find distasteful and it's driving ACTUAL Tech literate peeps away from what was once a fair and balanced site..

 

Regards 

p5ych00n5

Fanboys will be fanboys, and they'll be them with pride. It really amazes me that people are loyal to a brand no matter the value of other brands' products.

Oh well, welcome to the internet and people in general I guess.

[p5ych00n5]

Fanboys will be fanboys, and they'll be them with pride. It really amazes me that people are loyal to a brand no matter the value of other brands' products.

Oh well, welcome to the internet and people in general I guess.

 

*Deep Breath*

 

I probably am being a wee bit precious, but misinformation really rustles my jimmies

[ablangc]

I'm seeing a lot of fanboying and toxicity on this site especially towards certain companies.

 

Not trying to flame but this is the only thread I've seen (and correct me if I'm wrong) which puts the FuryX on level territory or even beating the topend NV cards despite the multitude of ACTUAL real life instances, on these forums, it's only getting drowned out by the HuurDuur's..

 

-Snip-

 

Regards 

p5ych00n5

 

 

You do realize that the OP posted charts showing  crossfire Fury X compared to single cards right?  To say that the Fury X is on level territory or even beating the topend NV cards based on those charts is massively skewed.  They did not even post the 980 TI SLI benchmarks with whatever settings they used.  Albeit, the OP was stating that it was getting near 100% scaling which is quite impressive.  I wish the same was for NVIDIA but as history has shown, AMD scales better when it comes to Crossfire vs SLI.  I can't wait to see what AMD has in store for the next generation of GPU's because I will definitely be keeping an eye on that.  Possibly replace what I have now.

[p5ych00n5]

You do realize that the OP posted charts showing  crossfire Fury X compared to single cards right?  To say that the Fury X is on level territory or even beating the topend NV cards based on those charts is massively skewed.  They did not even post the 980 TI SLI benchmarks with whatever settings they used.  Albeit, the OP was stating that it was getting near 100% scaling is quite impressive.  I wish the same was for NVIDIA but as history has shown, AMD scales better when it comes to Crossfire vs SLI.  I can't wait to see what AMD has in store for the next generation of GPU's because I will definitely be keeping an eye on that.  Possibly replace what I have now.

Good point, I did try showing vs to vs, but got flamed (elsewhere) for showing charts from different sites even thought the FPS were + - either way by 1 - 2 FPS to either card. The thing is there is no direct head to head yet, I just showed a RELATIVE correlation of results.

 

Until there is a definitive head to head the jury is out

[ablangc]

Good point, I did try showing vs to vs, but got flamed (elsewhere) for showing charts from different sites even thought the FPS were + - either way by 1 - 2 FPS to either card. The thing is there is no direct head to head yet, I just showed a RELATIVE correlation of results.

Until there is a definitive head to head the jury is out

I am actually waiting for benchmarks to come out to show SLI versus crossfire in regards to 980 TI versus fury X. Then show results from stock and overclock settings. I remember reading on a website that they exposed a flaw on the fury X and was able to overclock the memory. With scaling damn near perfect on the Fury X, and hopefully unlocking overclock settings on the card, i'm pretty sure it will be neck and neck. I almost want to say it might be faster.

[Alurit]

snip

well based on these avg framerates they do seem to be using max AA in op's chart

[ablangc]

well based on these avg framerates they do seem to be using max AA in op's chart

 

Yeah I am about to do a GTA V run with MSAA 2 and MSAA 4 and post the results later. 

 

As to what MSAA they used is beyond me.  Since HBM is new to a lot of people, I know that if they were to use MSAA 4 at 4k, it would exceed the 4GB of VRAM but it's HBM. HBM 4 may equal to 6GB of DDR5? 

[maizenblue]

Crossfire always seems to beat SLI these days. Something Nvidia fans continually ignore. In fact you still see them claiming the exact opposite.

 

These results don't surprise me. The Fury X didn't obviously beat the 980 ti at launch and clearly had some teething issues, so the Nvidia fanboys pounced, but in the medium term I think this cards reputation and performance will continually improve, while the 980 ti with its excessive noise and heat will probably look less impressive.

[Agent181]

Das nice

[IrshaadH]

those minimum fps values really suffer though...

[-rascal-]

Because it's water cooled, if it where air it would be horrible

 

You need to understand, TDP is a measurement of heat.

Heat as in ,Q, measured in Watts.

Temperature =/= Heat

 

If you were to go by TDP given by NVidia / AMD

980 TI: 250W

R9-FuryX: 275W

Not a HUGE difference.

[amintha1992]

does arkham knight even support sli/cf?

no it does not. that is why there is no performance improvement. 

[othertomperson]

You need to understand, TDP is a measurement of heat.

Heat as in ,Q, measured in Watts.

Temperature =/= Heat

 

If you were to go by TDP given by NVidia / AMD

980 TI: 250W

R9-FuryX: 275W

Not a HUGE difference.

 

Which is all the more frustrating that we aren't going to see Sapphire or Gigabyte air-cool this thing. For me the CF performance is completely irrelevant because I don't want three AIOs in my case, I think it looks absurd. I get the 295x2 being water cooled only... I would love to see Gigabyte's claim that the Windforce can handle 600W tested but I get it; but not allowing any air cooled versions for the non-cut down Fury X is silly imo.

[-rascal-]

Which is all the more frustrating that we aren't going to see Sapphire or Gigabyte air-cool this thing. For me the CF performance is completely irrelevant because I don't want three AIOs in my case, I think it looks absurd. I get the 295x2 being water cooled only... I would love to see Gigabyte's claim that the Windforce can handle 600W tested but I get it; but not allowing any air cooled versions for the non-cut down Fury X is silly imo.

 

Absolutely agree with you there.

AMD isn't letting their board partners use their own PCB and cooler design.

 

Hopefully....they change their mind, or somebody like Sapphire can be sneaky   :ph34r:  

[p5ych00n5]

Which is all the more frustrating that we aren't going to see Sapphire or Gigabyte air-cool this thing. For me the CF performance is completely irrelevant because I don't want three AIOs in my case, I think it looks absurd. I get the 295x2 being water cooled only... I would love to see Gigabyte's claim that the Windforce can handle 600W tested but I get it; but not allowing any air cooled versions for the non-cut down Fury X is silly imo.

 

Meh, it's like Nvidia not allowing board partners to redesign a flagship design. (ala Titan/X/Z)

 

I'm seeing AMD knowing what the FuryX is capable of, riding out the initial backlash of "This cards shit", drivers mature, HBM yields improve, the card kicks absolute arse and HALO CARD MOFO. Lets AIB design boards for the non X Fury

 

Damn got Rascal Ninjad

[xjaffacake]

Notice the minimum frames are some of the lowest, which would suggest a not a smooth frame rate which could be equally as bad as a low frame rate

[Hunter7263]

That's actually really impressive good job AMD. It's a shame that the rest of the 300 series are just rebrands with slightly higher clocks  :mellow:

REFRESHES!!! Yes same gpu but NOT the same card. The upgraded other components on the board to give it better power delivery and stability which is what allows it to be more efficient and overclock the memory so far. In addition to that you CAN over clock these cards just as much as you could overclock the 200 series counterparts.

[p5ych00n5]

^^^^^^^^^^^^^^^

This, so much this

[alphaproject]

Ab's scores look much better than the Fury X. Sorry AMD. And what is with some of those low frames with CF? A tall spike is less important than minimum frame rates.

 

Plus that article is basically saying you NEED to Xfire the Fury X in order to get decent performance.

[DarkSmith2]

you will have to sli/cf every card in future to get decent performance.

[MageTank]

I'm seeing a lot of fanboying and toxicity on this site especially towards certain companies.

 

Not trying to flame but this is the only thread I've seen (and correct me if I'm wrong) which puts the FuryX on level territory or even beating the topend NV cards despite the multitude of ACTUAL real life instances, on these forums, it's only getting drowned out by the HuurDuur's..

 

This is not a rant directed at anyone personally but this I find distasteful and it's driving ACTUAL Tech literate peeps away from what was once a fair and balanced site..

 

Regards 

p5ych00n5

Yeah, i see it happening a lot too. I myself am very anti-fanboy. I see graphics cards as tools, not as sports teams one roots for just because they exist. People forget that we owe absolutely nothing to these companies, let alone our allegiance. They exist as a means to profit off of us based on what they can offer us. To defend them when they release mediocre products just causes more harm than good. Regardless of who the company is, they need to be held responsible for their mistakes. A consumer should not be making excuses on behalf of a company, nor should they be accepting of a companies excuse as to why their product did not live up to the advertised specs, or promised performance. 

 

It is a mindset i will not understand. Do people feel like they are being insulted when someone exclaims that X company has terrible products, because they happen to own a product from X company? If so, why do they feel that defending that company will make them feel better? Is it self rationalization? Are they trying to justify a purchase that may or may not have been the best decision? Are they trying to convince themselves that they are happy with something when they secretly are not? I would love to be filled in on this subject.

 

I just wish i knew when things became backwards. Consumers should not be fighting for the companies. The companies should be fighting for the consumers. Alas, my words will fall on deaf ears, and people will continue to have their "MY OBJECT IS BETTER THAN YOURS BECAUSE I LIKE IT" arguments, and every piece of factual information will be ignored. 

[anthonyc813]

thanks for sharing this and doing all the work!

<3