Interested to hear your findings.
You seem about as impartial and level headed as they come on these forums.
-
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.
The Fury(X) Fiji Owners Thread
- Thread starter Kaapstad
- Start date
Interested to hear your findings.
You seem about as impartial and level headed as they come on these forums.
Soldato
- Joined
- 28 May 2007
- Posts
- 10,147
That new score doesn't sound promising considering a nicely clocked 980 Ti/Titan X can get in the neighbourhood of 21k
Actually that's a 10% gain from a bit of driver tweaking. Add 10% onto the guru 3dmark scores at stock and it would beat 980ti/Titan X in all Firestrike benchmarks. If it overclocks well with voltage then it could well top the Bench threads. It's to soon to say but a 10% gain to Fury in gaming performance would put Fury X around where most wanted it.
One things for sure this gain needs to be shown in the Fury pro review if indeed Amd can produce it.
http://www.guru3d.com/articles_pages/amd_radeon_r9_fury_x_review,25.html
Last edited:
It's not really about the raw score in 3dMark. It's the possible increase in performance it could represent across the board.
If the driver improvements equate to a 10% performance increase across a broad spectrum of games, that's pretty cool.
If the driver improvements equate to a 10% performance increase across a broad spectrum of games, that's pretty cool.
Soldato
Some don't see it that way LOL. I'm waiting for the second one before I do anything - or will be playing with it in an i5 box next week. No point stripping three TITAN X blocks out of a loop for one Fury X.
It's pretty obvious the first thing that needs to be done is put it in scenarios where it is likely to be constrained by the memory - or not. Numbers can lie but frametime analyse doesn't
Soldato
- Joined
- 28 May 2007
- Posts
- 10,147
Yep and all this guy did was use a newer driver and force it onto a Hawaii and Tonga code path. With proper code for Figi maybe more is possible.
Indeed... will be very interesting to see what the standard Fury brings to the table. If it's a pale imitation of the Fury X then it's going to struggle to steal many sales from the 980, nevermind the Ti, unless they price it VERY aggressively. You never know though, we may see some magic in the coming months, voltage unlocks across the board and significant performance increase... or not lol.
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 30,841
Indeed... will be very interesting to see what the standard Fury brings to the table. If it's a pale imitation of the Fury X then it's going to struggle to steal many sales from the 980, nevermind the Ti, unless they price it VERY aggressively. You never know though, we may see some magic in the coming months, voltage unlocks across the board and significant performance increase... or not lol.
AMD driver guy/s will be going hell for leather until the 14th no doubt
That new score doesn't sound promising considering a nicely clocked 980 Ti/Titan X can get in the neighbourhood of 21k
And on another piece of news, Apparently the whole of north america had a total stock count of around 700 Fury X's according to a mate of mine who works in NewEgg which is a lot lower than I thought it would be, AMD must really be having production issues.
This guy who did benchmarks has quite weak cpu though. Also drivers are missing codepath for Fury, so he's running codepaths for Tonga and Hawaii.
Also ignore that score for Hawaii, at it shows it has tesselation factors changed.
That new score doesn't sound promising considering a nicely clocked 980 Ti/Titan X can get in the neighbourhood of 21k
And on another piece of news, Apparently the whole of north america had a total stock count of around 700 Fury X's according to a mate of mine who works in NewEgg which is a lot lower than I thought it would be, AMD must really be having production issues.
Either he lies, you lie, or he's misinformed.
It was the #3 best selling card on Amazon.com alone ... and the only high selling expensive card (the others were mostly 970s). 700 for the whole of north america wouldn't even dent the sales chart.
Indeed... will be very interesting to see what the standard Fury brings to the table. If it's a pale imitation of the Fury X then it's going to struggle to steal many sales from the 980, nevermind the Ti, unless they price it VERY aggressively. You never know though, we may see some magic in the coming months, voltage unlocks across the board and significant performance increase... or not lol.
Dunno how you think that. The 390X is a much faster (and cheaper) card than the 980 already at 2560x1440 and 4K. It's nearly 50% faster in quite a few games at 4K. It's very close at 1920x1080. With the latest leaked drivers you might see it overtake the 980 at 1920x1080 too.
The Fury's obviously going to be faster than the 390X / 980 and it inhabits a price point that NVIDIA doesn't (at least with reference cards).
It probably would in such a small timescale ( ie if all 700 sold on launch day)
Where does this keep coming from? From the reviews i have read the 390X is not even faster than the 980 at 1440p, let alone much faster!
Last edited:
It wouldn't. On those figures, Amazon would have been lucky to get more than 50 cards.
Yes it is. The only reviews that showed it not being didn't use the 15.15 launch drivers with all the improvements. They used 15.5.
Caporegime
That may even be 15.5 ^^^^
Its only faster than a 980 at 4K, by a cats whisker, none the less a quick card overall at any res snapping at the 980's ankles.
By that measure the 290X would be just as quick clock for clock.
Its possible AMD have had some huge Driver based performance improvements putting Hawaii right up there with the full fat GM204.
Mainstream reviewers just haven't retested Hawaii for some time.
Hawaii is actually quite an impressive bit of GPU these days.
Its only faster than a 980 at 4K, by a cats whisker, none the less a quick card overall at any res snapping at the 980's ankles.
By that measure the 290X would be just as quick clock for clock.
Its possible AMD have had some huge Driver based performance improvements putting Hawaii right up there with the full fat GM204.
Mainstream reviewers just haven't retested Hawaii for some time.
Hawaii is actually quite an impressive bit of GPU these days.
Last edited:
Can you point me to some reviews of this then? All the reviews i have seen have it only matching or beating the 980 at 4k. From all the reviews i have seen, at 1080p the 980 is decent bit ahead and at 1440p the 980 is still faster in the majority of games.
Last edited:
As referenced above, I see no evidence that the 390X is MUCH faster than the 980... especially when pegged against the high performance varieties. Besides, I didn't say the Fury wouldn't be a better card than the 980, I just wonder how well it's going to sell if it ends up being a cut down Fury X (as rumoured) and the performance is hindered by a lack of ability to OC (which I can't believe will happen, but you never know)... and then it's all going to depend on how competitively it's priced vs the £500 Fury X. So many unknowns of course. Interesting times ahead.
Indeed. the Guru 3d review showed a reference 980 being generally faster across the board at 1080p and 1440p when compared with the gaming edition of the 390x.