Soldato
4k isn't even testing them. On to 5k 
-
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
Soldato
- Joined
- 30 Nov 2011
- Posts
- 11,375
yeah because we all love a 5 or 20fps minimum 
Caporegime
yeah because we all love a 5 or 20fps minimum
Seems to be a glitch as it's only like that in a benchmark sequence for the first second. Happens at 1080P as well.
Soldato
New driver ?
that's I think early beta driver. we are now on 15.20 or something
Permabanned
Look between the lines. In a few of the games they have AA disabled or TressFX off or are not running maximum settings. In addition, why have they got a £800 cpu running at a stock setting of 3.5Ghz? That's not how you bench a 3/4 GPU setup.
Here's what Thief looks like on a QuadFire Fury X setup.
Thief
4K Max Settings
Mantle
4x FuryX @1125/500Mhz
15.15
I could beat their four card score with three cards.
Here's what Shadow Of Mordor looks like. Note, I'm using higher settings than they are.
5960X @4.9Ghz
4x Fury X
15.7.1
Windows 10
2160P
@1050/500Mhz
@1115/500Mhz
I could go on, the Battlefield 4 and Grid Autosport scores are miles off too, but i think you get the point.
This is meant to be about the scaling between 2, 3 and 4 cards.
As for there cpu it doesn't mention speed so presumably they are running it at stock speeds which seems a bit dumb considering it's an open bench and they're using an AIO cooler.
Someone with Facebook should post on Tweaktown and ask them if they are overclocking the cpu and why not, When they reviewed the cpu they do so including a 4.4 overclock which they said was very easy to get so you'd think they'd have the sense to use it if there not when doing these card tests. Especially considering how low the stock clock is and how most games don't care for the extra cores it might have.
Anyway scaling Matt. What's that like?.
Last edited:
I have seen it happen way too much with tech sites where they bench 3 or 4 cards and don't use max settings or the CPU is way too slow or even at stock.
Even worse depending on which way the site leans they will blame the drivers and say either SLI or C/F does not scale as well as the other (cross out SLI or C/F here).
I think Tweaktown also said in the review that they will be also benching 4 TXs to compare. That should be good for a laugh as I have 4 FXs and 4 TXs along with a couple of 5960X based systems so I may post some of my own scores when they do the review.
The sort of games I play @4k I can get away with using a single card but when I use the extra cards they do scale and pull their weight. Averaging 170fps+ in BF4 maxed @2160p in a 60+ player game is possible, shame my monitor will only do 60htz.
Permabanned
We always seem to be getting results where the 3 and 4 scaling is not worth the cost and often it adds very little over 2 way. So if properly maxed hardware changes that it would be good to see some results put up as a comparison against the usual testing when you get a chance too.
Soldato
The cost? Kaaps/AmdMatt's setup is pretty much my dream setup (with the exception that dream included dual CPU's). I would have that in a heartbeat if I could.
8K screen, or 3x 4K's Let's do this!!! 
8K screen, or 3x 4K's
Let's do this!!! shame my monitor will only do 60htz.
hardware without proper display needs an upgrade today
I have a 2560p 144hz screen and since I wanted one since ages ago (Tolkien age) its good to find they are really good monitors today.
4k and 8k 144hz etc...will make me behind the curve there.
but 120+hz is just to good to pass up.
Soldato
Starting to get really annoyed with mine now, GTA V performance is terrible, just constant stutters and fps drops, regardless of if i Alt enter in and out, Bf4 is crashing after about 15 mins even with the cards at stock 
tried rolling back to 15.7.1 drivers but the same crap happens.
tried rolling back to 15.7.1 drivers but the same crap happens.
Caporegime
Starting to get really annoyed with mine now, GTA V performance is terrible, just constant stutters and fps drops, regardless of if i Alt enter in and out, Bf4 is crashing after about 15 mins even with the cards at stock
tried rolling back to 15.7.1 drivers but the same crap happens.
What message are you getting on BF4 mate? I'm playing both of those games using 2x Fury X at the moment and 4K and they both work well. GTA has some off stutters, but BF4 is as smooth as a babies bottom on Mantle and DX11.
Is your system stable? CPU overclock?
Permabanned
While 120+ hz would be nice to have (I've never owned one) I think I'll prefer something like this which at 75 hz is plenty considering how much grunt you'll need to push this resolution anyway.
https://www.overclockers.co.uk/showproduct.php?prodid=MO-090-AC&groupid=17&catid=948
Soldato
My system system is overclocked matt, but i even put it all back to stock and got the same thing, game just locks up and CTD, with error message Something like "AMDDeviceRemoved, incorrect command"
I did take a pic on my phone but i've left it in my GF car.
It's breaking me bud, GTAV and BF4 and pretty much my two most played games too.
Witcher 3, MGS5 etc have all been rock solid.
It's definitely an issue with either Crossfire the AMD Drivers on my system though.
Caporegime
Have you tried both API's in BF4?
Are the Fury X cards at stock?
Which drivers are you using? Am using 15.7.1 here from amd.com
Soldato
Cards are at stock bud, i've tried 15.7.1 and 15.8, both have the same issue, not tried Mantle no, you did say to use DX11 so i've just continued to use it.
Caporegime
- Joined
- 18 Oct 2002
- Posts
- 30,833
But what about your system?
Use OCCT mate, if it can pass 64-bit Large Data Set for 5 mins then it's solid
Caporegime
I've been testing Mantle on the Fury X and it works perfectly, even without Dice optimising for it. Give it a go.
Soldato
But what about your system?
Use OCCT mate, if it can pass 64-bit Large Data Set for 5 mins then it's solid
lol System is Solid bud, the crash happens even at stock, it's a graphics driver/ issue.
Matt are you using Mantle? I'd rather use DX11 bud tbh.