Yeah, there isn't much difference between them, less than what there is between the GTX 970 and 980 and they don't performance all that far apart.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
** AMD R9 NANO NOW IN STOCK & AVAILABLE!! **
- Thread starter Gibbo
- Start date
Yeah, there isn't much difference between them, less than what there is between the GTX 970 and 980 and they don't performance all that far apart.
Permabanned
- Joined
- 12 Sep 2013
- Posts
- 9,221
- Location
- Knowhere
That's why having sold some at 599 surprised me when you still had them at 529, I suppose some people simply wanted the right sticker brand and they could afford to not let the difference concern them.
Thinking about it if I won the lottery I'd probably do the same seeing as I have a ROG board and wouldn't be looking at the price.
That is pretty darn smart!
I've not owned/used an NVIDIA card since the 8800GT, many years ago. I frequent this forum and several others, and noticed quite a few NVIDIA users complaining of driver issues using a single GPU in recent months.
From the top of my head, there was an issue with hardware acceleration in flash that affected a few browsers, some BSOD issues and driving stopped responding issues, that I saw quite a few times.
I understand that you've owned both NVIDIA and AMD recently, so are in a better position to judge on driver issues for single GPU's etc, though as I understand it you sold your AMD cards, as you preferred NVIDIA?
I guess we're both a little biased, I favour AMD and you favour NVIDIA, though I'm being 100% truthful when I say I've yet to encounter a driver issue over the last 10+? years using a single AMD GPU.
99.9% of all AMD driver complaints seem directed towards crossfire profiles, something I have no interest in due to the scaling issues and the fact that you don't get any use of the 2nd/3rd/4th card until a specific driver/profile is written for the specific game you play.
Last edited:
Actually on some games and benches the single card mode support for Fury Xs is dreadful and the 290X often embarrasses them lol.
Here is a good example in DX12 !!!! where an overclocked 290X is trading blows with a Fury X lol.
2160p
DX12
- Average FR 20.8, GPU 980 Ti @1452/1842, Normal FR 25.5, Medium FR 19.9, Heavy FR 18.2, CPU 5930k @4.4, rickyjb Link
- Average FR 20.3, GPU TitanX @1127/1752, Normal FR 24.5, Medium FR 19.4, Heavy FR 18.0, CPU 5960X @4.0, Kaapstad Link
- Average FR 19.0, GPU Fury X @1050/500, Normal FR 23.4, Medium FR 18.1, Heavy FR 16.7, CPU 5820k @4.4, ALXAndy Link
- Average FR 18.6, GPU 290X @1200/1500, Normal FR 22.0, Medium FR 18.3, Heavy FR 16.4, CPU 4790k @4.7, bloodkil Link
DX11
- Average FR 23.8, GPU TitanX @1491/2002, Normal FR 31.2, Medium FR 22.6, Heavy FR 20.1, CPU 5960X @4.0, Kaapstad Link
- Average FR 22.3, GPU 980 Ti @1452/1842, Normal FR 28.4, Medium FR 21.2, Heavy FR 19.2, CPU 5930k @4.4, rickyjb Link
[*]Average FR , GPU @/, Normal FR , Medium FR , Heavy FR , CPU @,
[*]Average FR , GPU @/, Normal FR , Medium FR , Heavy FR , CPU @,
I find it likely that when this game is released, AMD and NVIDIA will have developed better drivers that will be able to extract every ounce of power out of the GPU used.
One thing's for sure though, the 290/390 series are not to be trifled with
Decent backup cards at least
During your use of AMD GPU's (single GPU's) did you ever have a driver issue of note?
Soldato
- Joined
- 29 Aug 2010
- Posts
- 8,244
- Location
- Cornwall
Despite humbug cleverly trying to derail Kaap's point, I think the benchmark does prove it a bit.
While I'm sure nobody is more upset than Oxide games that Nvidia are doing well in the benchmark (probably missing the days of Mantle when Nvidia could be excluded more easily), Kaap's point wasn't really about AMD vs. Nvidia performance (the way I read it).
If you look at the difference between the 970 scores and the 980TI scores you'll see the 980TI is on it's way to doubling the 970 scores. If you compare the 290X (close to a 390X?) to the Fury X you can't say the same. So either AMD's new flagship card isn't much better than last generations or the FuryX isn't performing at it's peak yet. Which would surely be down to drivers?
Aren't there also benchmarks done by AMDMatt showing 2 x FuryXs to scale more than double that of 1 FuryX? It seems more likely to me that a single card is underperforming due to drivers rather than 2 cards overperforming. Unless there's a glitch in the drivers or benchmark meaning the results are unreliable.
While I'm sure nobody is more upset than Oxide games that Nvidia are doing well in the benchmark (probably missing the days of Mantle when Nvidia could be excluded more easily), Kaap's point wasn't really about AMD vs. Nvidia performance (the way I read it).
If you look at the difference between the 970 scores and the 980TI scores you'll see the 980TI is on it's way to doubling the 970 scores. If you compare the 290X (close to a 390X?) to the Fury X you can't say the same. So either AMD's new flagship card isn't much better than last generations or the FuryX isn't performing at it's peak yet. Which would surely be down to drivers?
Aren't there also benchmarks done by AMDMatt showing 2 x FuryXs to scale more than double that of 1 FuryX? It seems more likely to me that a single card is underperforming due to drivers rather than 2 cards overperforming. Unless there's a glitch in the drivers or benchmark meaning the results are unreliable.
Any chance of having Greenland or Pascal in 6 months?
I hope "2016 release" doesn't mean "December 31st 2016"
Soldato
- Joined
- 30 Nov 2011
- Posts
- 11,427
The FuryX has more stream processors and texture units, but the same ROP's, so it could be that under current software it is being bottlenecked by having the only the same number of ROPs, it would appear that it needs very very specialised software to maximise stream use whilst not increasing the load on the ROPs to actually blend it all together
I guess it will depend on how many devs AMD can get on board to code not only for AMD in general but more specifically for the FuryX
Last edited:
depends where the bottleneck is.
you code for dx12 features not a specific card.
if the card do software emulation like the 980ti does then it wont work as well as a fury x that can do async shaders in hardware.
its why you see a 390 equal a 980ti there.
AMD built for the future Nidia didnt.
This is actually what i said, i didn't go into details as to why but clearly something is holding the thing back.
Decent backup cards at least
During your use of AMD GPU's (single GPU's) did you ever have a driver issue of note?
Not really, a couple of minor things like monitor waking but nothing that was a nuisance.
Soldato
- Joined
- 30 Nov 2011
- Posts
- 11,427
Rops. Read up on what Rops do and then think about why a 290X and a FuryX are getting nearly the same scores even in a DX12 benchmark.
And no a 390 is not equal to a 980ti in Ashes, I suggest you look at all of the results across all of the resolutions in the bench thread before jumping to conclusions
Soldato
- Joined
- 28 May 2007
- Posts
- 10,204
Keep in mind that the Fury was at stock and the 290x was overclocked. Do you still think the 290x will be keeping up when Fury X is overclocked with voltage. I expect the Fury X will be right up to where the nvidia cards are.
+1 replied thinking i was in the shes thread after reading some.
Soldato
- Joined
- 30 Nov 2011
- Posts
- 11,427
What are we expecting then to hit with voltage?
Soldato
- Joined
- 28 May 2007
- Posts
- 10,204
It's to soon but from the little i have seen a minimum of 1200 core but like the crazy people with 290's on the high end over 1300. I think middle ground will be around 1250 as all fury x's are aio cooled so thermals should not restrict them to much. Memory seems to do pretty well percentage wise due to the low base clock. Time will tell i might be way off.
Permabanned
- Joined
- 12 Sep 2013
- Posts
- 9,221
- Location
- Knowhere
I don't think it'll ever do a lot, Possibly 1200/1220 at best, 1160 or so for the Fury. It's odd that after hearing about the fella who had had some luck with unlocking the voltage it all went silent. That doesn't bode well.