FineWine===>FineVinegar FuryX Aging Horrible in 2017

[ikeke]

Seems like not everyone agrees.



Can't do anything about it , fanboys will always cling to whatever they got.

I guess using latest drivers also helps, if one wants to level the playing field for new game release and it's proper [initial] support.

Guru3D 17.11.2

Radeon Software Crimson ReLive Edition 17.11.2 Highlights

Support For

• Star Wars™ Battlefront™ II

PCGHW 17.11.1

Fanboys truly cling to whatever they can, i'll drink to that.

 

[MuhammedAbdo]

PCGHW 17.11.1

Fanboys truly cling to whatever they can, i'll drink to that.

Did you read the article? the tested the new driver and found no difference whatsoever.
http://www.pcgameshardware.de/Star-...s/Star-Wars-Battlefront-2-Benchmarks-1243688/

 

[Xzibit]

Did you read the article? the tested the new driver and found no difference whatsoever.
http://www.pcgameshardware.de/Star-...s/Star-Wars-Battlefront-2-Benchmarks-1243688/

It says they tested them on a Frontier Edition. The gaming charts don't show a Frontier Edition.

Looks like they were doing a Frontier Edition review and attached/link the update. They should have just tested it on the previous cards for consistency.

The FE numbers line up exactly to the LC Vega but the clocks are different. They say they had to manual adjust speeds on the FE test for consistency but Its amazing/strange the numbers are exactly the same.

We wanted to know exactly and therefore we performed a series of measurements with constant clock rates. It uses a Radeon Vega Frontier Edition, which, except for the storage capacity, exactly one RX Vega 64 corresponds. With manual optimization, our test card always keeps a good 1,500 MHz core and 1,100 MHz memory clock; the room temperature is constant 21 ° C

On top of the FE & LC numbers being exactly the same, they were done using 2 different systems.

Core i7-5820K @ 3,0 GHz, Radeon Vega Frontier Edition/16G @ ~1.500/1.100 MHz, Windows 10 x64
Results not necessarily comparable to previous reviews / results not comparable with earlier reviews

vs

Core i7-6800K @ 4.4 GHz (Ring / L3: 3.6 GHz), 4 × 8 GiByte Corsair DDR4-3000, MSI X99A Tomahawk
Results not necessarily comparable to previous reviews / results not comparable with earlier reviews

Found that part funny since the numbers are exactly the same

 

[MuhammedAbdo]

It says they tested them on a Frontier Edition. The gaming charts don't show a Frontier Edition.

Don't give a damn about Vega, focus on the FuryX vs 980Ti numbers please.

 

[Xzibit]

Don't give a damn about Vega, focus on the FuryX vs 980Ti numbers please.

You gave enough of a "Dam" to reply and link.

Let me remind you.

Did you read the article? the tested the new driver and found no difference whatsoever.
http://www.pcgameshardware.de/Star-...s/Star-Wars-Battlefront-2-Benchmarks-1243688/

 

[ikeke]

Guru3D tested using SWBF2 launch drivers for nvidia and amd, on the same platform, and got different result (which is from performance standpoint more in line with the cards performance).

I understand it paints a bad picture for the "finevinegar" crowd hence a more mix-and-match review is used to make their case.

Anyway, im out.

 

[MuhammedAbdo]

Guru3D tested using SWBF2 launch drivers for nvidia and amd, on the same platform, and got different result (which is from performance standpoint more in line with the cards performance).

They tested using DX12 which reduces fps on all cards, from both NVIDIA and AMD. NVIDIA cards are hit more than AMD though, which skews the results in the favor of FuryX. If you test the robust DX11 which everyone plays the game with, the 980Ti is considerably ahead.

 

[TheoneandonlyMrK]

Morale of the story: don't buy high end AMD GPU. And FineWine is now a lie.

I bought every generation bar that one , all the rest improved with age ,fine wine style, my vega has since launch , your being way to dramatic and one card is not Amds entire lineup so hyperboling an entire company with your skewed opinion is flawed imho.

I won't comment on your fury x issues since I have No experience but i will say I haven't ever bought a gpu bar the 5870 that lasted more then 2.5 years in a useable form for high res ultra spec gaming , its never panned out as possible due to tech ageing.
That includes a gtx 680 and gtx 480 a titan and will include the 1080ti ,in two and a half years some dev always eclipses its performance envelope with features in game or in the game engine.

 

[MuhammedAbdo]

I won't comment on your fury x issues since I have No experience but i will say I haven't ever bought a gpu bar the 5870 that lasted more then 2.5 years in a useable form for high res ultra spec gaming , its never panned out as possible due to tech ageing.

Only big Kepler aged badly for NVIDIA. Tesla, Fermi, and Maxwell aged fine. The GTX 480 still beats 5870 in the majority of games to this day.

 

[Vya Domus]

The GTX 480 still beats 5870 in the majority of games to this day.

And it was more expensive and it used 50% more power. And it ran so hot some people had to undervolt it so it wouldn't blow up.

But yes it was 10% faster and they only needed a die that was 70% bigger. Let us applaud Nvidia.

 

[TheoneandonlyMrK]

Only big Kepler aged badly for NVIDIA. Tesla, Fermi, and Maxwell aged fine. The GTX 480 still beats 5870 in the majority of games to this day.

I didn't comment on which was best or what beats what or what aged best please re read my post , you will find my point to be.

Gpus dont do max res ultra spec gaming on all new titles after two years and to expect so is unrealistic given that devs move the goal posts.

There you go in one sentence simplified, please do not try to troll me with bias influenced nonsense , i wont bite.

 

[MuhammedAbdo]

Gpus dont do max res ultra spec gaming on all new titles after two years and to expect so is unrealistic given that devs move the goal posts.

That's natural. but hat's not the point at all. The point of aging badly is that GPUs fall below their natural position compared to the competition, or to cards from different tiers. The FuryX should be always faster than RX580, but not in a lot of new titles, it's not.

 

[TheoneandonlyMrK]

That's natural. but hat's not the point at all. The point of aging badly is that GPUs fall below their natural position compared to the competition, or to cards from different tiers. The FuryX should be always faster than RX580, but not in a lot of new titles, it's not.

If they used exactly the same tech then maybe but no it doesn't and stuff surprisingly gets improved between generations so no your just putting your expectations in, as terms of good quality, your expectations mean nothing to me so im diacounting them as irrelevant as is your argument since the 580 is not using the same tech as a furyx.

And new tech is optimised and designed for what devs requested since the last tech was out , the mistakes of the fury x were absorbed and curtailed on later products, what a surprise.

 

[xkm1948]

Let me guess, i was a stupid ass for buying AMD's flagship GPU then and hoping it would do well down the road. Yeah i agree i have been naive thinking like that. Mistakes made and lessons learned. Not buying AMD flagship GPU again. Never believe in stuff like "FineWine". Just go straight for the performance one need. We can move on now.

 

[cucker tarlson]

That's natural. but hat's not the point at all. The point of aging badly is that GPUs fall below their natural position compared to the competition, or to cards from different tiers. The FuryX should be always faster than RX580, but not in a lot of new titles, it's not.

It was when 290X started beating 780Ti, now it isn't
AMD started gimping Fury X even before Vega came out, if ppl were following quality tech sites like pcgh,computerbase,purepc,sweclockers they'd know that already. Now it's gotten embarrasing, I think they've outdone themselves to let it fall behind Polaris too. I've seen people buying Fury (3584sp air coolerd versions) when rx480 hit the market as their prices fell and cut Fiji still had its nose ahead of Polaris. They must be delighted now.

 

[xkm1948]

Below a 970 some time?? Oh my gawd.

That is just sad.


So yeah, sucks to be me for buying this stupid FuryX in the first place.

 

[cucker tarlson]

don't look, don't look, don't look..... this is what spoilers are for, silly me.

I remember ppl saying Vega 10 would have to be 1.5x the performance of Fury X to beat GTX 1080.

AMD found a way to make it even more than 1.5x

 

[xkm1948]

One of most experienced Fury/X modder @gupsterg posted these on r/AMD. Same things can be found over overclock.net and overclockers.com

Overclockers.com link
https://forums.overclockers.co.uk/t...-on-fiji-after-v16-12-2-whql-driver.18790677/

Reddit link
https://www.reddit.com/r/Amd/comments/6lkxf3/r9_fury_fury_x_owners_how_did_performance_change/


@fullinfusion

If this is not actively crippling/gimping by AMD's driver team then I don't know what it is then. New FineWine features?


Overclockers' dream, FineWine. Yeah right.


I will quote gupsterg post before he moved on to 1080

What we have here IMO is AMD have placed a block within driver.

My card will not do 700MHz HBM (ie 1400MHz effective speed) on any driver prior to v17.x.x, but with any v17.x.x driver it will and I still get performance as card is at 500MHz. Reference zip in 1st post of thread and a linked screenie at end of post 1.

There is no doubt in my mind that AMD have gimped Fiji.

There is no AMD FineWine for Fiji.

I have noted over the course of ownership ~5% gains at best in performance with drivers. Was I after more? not really. I just wanted bug fixes and things not taken away that have been there.

SWBF has been an issue of Fiji for months as well. Not only experienced by me but by others as well, link. I have religiously used AMD driver feedback page since every v17.x.x driver to highlight issue with SWBF. Lowering Mesh Quality is a workaround and not a fix, this workaround works on all v17.x.x drivers but on v16.12.2 WHQL I do not need this workaround.

Also thank you @fullinfusion for trying to communicate with AMD's driver team. From what @gupsterg said AMD is pretty much set on pulling life support for the Fiji line. Yeah it will limp along the other older GCN cards as long as AMD put Fiji on supported driver list. I honestly have no hope that AMD's driver team would look at any Fiji related driver issues.

 

[Vya Domus]

So let me get this straight AMD is purposely gimping Fiji , used by rather old cards that are no longer in production and that were never that popular in the first place.

We are reaching levels of fanboyism that shouldn't even be possible. Nvidia gimping conspiracy theorists ain't got nothing on this.

 

[Xzibit]

One of most experienced Fury/X modder @gupsterg posted these on r/AMD. Same things can be found over overclock.net and overclockers.com

Overclockers.com link
https://forums.overclockers.co.uk/t...-on-fiji-after-v16-12-2-whql-driver.18790677/

Reddit link
https://www.reddit.com/r/Amd/comments/6lkxf3/r9_fury_fury_x_owners_how_did_performance_change/

@fullinfusion

If this is not actively crippling/gimping by AMD's driver team then I don't know what it is then. New FineWine features?

Overclockers' dream, FineWine. Yeah right.

I will quote gupsterg post before he moved on to 1080

Also thank you @fullinfusion for trying to communicate with AMD's driver team. From what @gupsterg said AMD is pretty much set on pulling life support for the Fiji line. Yeah it will limp along the other older GCN cards as long as AMD put Fiji on supported driver list. I honestly have no hope that AMD's driver team would look at any Fiji related driver issues.

Unfortunately I took the time to read the thread start to finish.

That person gupsterg is complaining about newer drivers not allowing vbios changes to overclock the HBM. So hes upset about an unsupported feature that is not there anymore, which AMDMatt addresses in that forum thread and several people pointed it out to him.

I did chuckle at this part

I have noted over the course of ownership ~5% gains at best in performance with drivers.

Does not fit your narrative

 

[TheoneandonlyMrK]

Let me guess, i was a stupid ass for buying AMD's flagship GPU then and hoping it would do well down the road. Yeah i agree i have been naive thinking like that. Mistakes made and lessons learned. Not buying AMD flagship GPU again. Never believe in stuff like "FineWine". Just go straight for the performance one need. We can move on now.

You are downplaying the improvement brought by Polaris in architecture ,i wouldn't imply any of that but honestly I think things moved on way to quick for fiji to contend, Nvidia have changed little of their shader array and lean heavy on their optimization via driver ,plus Nvidia and Amds performance moved up a notch , it's time to move on regardless imho.

 

[eidairaman1]

Xkm1948, what brand is that furyx specifically?

 

[xkm1948]

Sapphire. It is in my system profile

 

[MuhammedAbdo]

Hardware Unboxed arrived at the same conclusion: The 980Ti is aging considerably better than the FuryX!

 

[xkm1948]

Am i the only sucker that bought into AMD's Fury/X cool aid around TPU?? Where are tge other Fury/X owners?