• Guest, The rules for the P & N subforum have been updated to prohibit "ad hominem" or personal attacks against other posters. See the full details in the post "Politics and News Rules & Guidelines."

Asus Matrix r9 280X CF = nothing but the issues

b3ka

Junior Member
Oct 19, 2013
22
0
0
Hey to all, new to the forum and of course with issues.
I switched from Nvidia to ATI yesterday and I wasn't even able to play any game comfortable since then :)

My cards are www.asus.com/Graphics_Cards/MATRIXR9280XP3GD5
My PSU is Seasonic GOLD 750W

and my issue is that after a couple of minutes of playing BF3/running Valley Benchmark/etc. one of cards is going from 90-99% USAGE to ~50% USAGE making everything so choppy it's not even possible to play.
Has anyone experienced it? I'm using the newest beta drivers from ATI.com. Tried to download ASUS one (that are supossed to be launched yesterday), but after installing them gaming is simply not possible, everything is choppy from beginning.

So is it drivers based issue? Or maybe my PSU is not enough and one of cards is getting a voltage cut? I metered the usage and entire box is using around 600-650W while gaming (with 90-95% of PSU efficiency).
 

VulgarDisplay

Diamond Member
Apr 3, 2009
6,194
2
76
List all your hardware, resolution, and if you are an nvidia employee lol.

Warning issued for thread crapping.
-- stahlhart
 
Last edited by a moderator:

blackened23

Diamond Member
Jul 26, 2011
8,556
1
0
You bought TRIPLE SLOT cards to crossfire with? That was your first mistake. I don't care what kind of case you have, your temps will get too high and you won't be able to maintain temperatures no matter what kind of GPU you have (nv or AMD).

You can't crossfire or SLI triple slot cards. I've tried it. The temps of the upper card will get so high that it will BSOD because they are sandwiched. I'm telling you right now, this won't work, you should RMA one or both because 3 slot cards in mGPU never work for more than a couple of minutes due to temperatures.
 
Last edited:

b3ka

Junior Member
Oct 19, 2013
22
0
0
Ops, sorry about that
i5-3570K
asrock z77 ws
2x matrix asus r9 280x
16GB of ram
2x Samsing 830SSD in RAID0

playing on 1440p 27'

not a nvidia employee :)
 

b3ka

Junior Member
Oct 19, 2013
22
0
0
Probably. So let me get this straight, you bought TRIPLE SLOT cards to crossfire with? That was your first mistake. I don't care what kind of case you have, your temps will get too high and you won't be able to maintain temperatures no matter what kind of GPU you have (nv or AMD).

You can't crossfire or SLI triple slot cards. I've tried it. The temps of the upper card will get so high that it will BSOD because they are sandwiched.
Temps are actually pretty good, it's 70-80 degrees in full stress and 30-40 in idle. So it's not TEMPs I'm worried about.
btw asrock z77 ws is pretty good for CF/SLI as there is a nice GAP between cards, they are not overheating.
 

blackened23

Diamond Member
Jul 26, 2011
8,556
1
0
Temps are actually pretty good, it's 70-80 degrees in full stress and 30-40 in idle. So it's not TEMPs I'm worried about.
btw asrock z77 ws is pretty good for CF/SLI as there is a nice GAP between cards, they are not overheating.
It won't work. Trust me. After 10-15 minutes the top card will be 20C hotter than the lower at 100% GPU load. I've tried this many times and the temps go stupid after 20 minutes or so. You will never be able to game for an extended amount of time with 3 slot cards in mGPU.

IT WILL NOT WORK WITH 3 SLOT CARDS. You just need to RMA one or both.

Also, aside from this, your PSU is not powerful enough for these cards as they are factory overclocked. Tahiti uses a ton more power when overclocked and the Matrix is 1100 factory OC'ed IIRC? One card alone can do near 500W power draw with 1.3V on the Tahiti, let alone two.

But the main point is, what you're doing is a complete waste of time. It won't work. One because they're 3 slot. Two because your PSU isn't high enough for factory OC'ed 7970s....they use a ton of power when overvolted, and the matrix is overvolted.
 
Last edited:

b3ka

Junior Member
Oct 19, 2013
22
0
0
Thanks for your input, I've got 14 days to return, still have time to upgrade PSU and check if this is the issue. Anyone else with CF can say something about the issue?
 

Teizo

Golden Member
Oct 28, 2010
1,271
31
91
That Seasonic 750 should be able to handle those cards, though it may be reaching close to 85% under load. Get you a Killawatt P3 and see what it reads.

You may also want to raise your power/thermal limit. The new R9's have a boost similiar to NV's GPU Boost 2.0. So it may be throttling down to stay at the 80 degree temp limit.
 

wand3r3r

Diamond Member
May 16, 2008
3,187
0
0
Do you have the +20% power limit on? Which driver version are you running? They fixed crossfire in 13.8 beta with the framepacing thing or so and I think 13.10 added some crossfire fixes. I think you might have to enable frame pacing in CCC?
 

b3ka

Junior Member
Oct 19, 2013
22
0
0
That Seasonic 750 should be able to handle those cards, though it may be reaching close to 85% under load. Get you a Killawatt P3 and see what it reads.

You may also want to raise your power/thermal limit. The new R9's have a boost similiar to NV's GPU Boost 2.0. So it may be throttling down to stay at the 80 degree temp limit.
Running 13.11 Beta Driver. Didn't play with power/thermal limit yet.
Checked for these limits, seems there is power limit set to 100% but can't find anyting related to temperatures. I think you might be right about 80 degree being the spot for some cut.
Still investigating.

Do you have the +20% power limit on? Which driver version are you running? They fixed crossfire in 13.8 beta with the framepacing thing or so and I think 13.10 added some crossfire fixes. I think you might have to enable frame pacing in CCC?
I think I don't havea any limits increased past 100%. Will look for frame pacing.
 
Last edited:

wand3r3r

Diamond Member
May 16, 2008
3,187
0
0
Put the power limit to +20%, there was mention a long time ago with clock speeds and the power limit so it should be a quick test. Make sure you have framepacing enabled too.
 

b3ka

Junior Member
Oct 19, 2013
22
0
0
Framepacing enabled by default.
What is interesting that default voltage for card is 1,2510.
I think it's the 80 degrees spot where the card is starting to lower itself, however im monitoring vCORE and memory and there are no changes.

Will try power limit.
 

Imouto

Golden Member
Jul 6, 2011
1,244
2
81
Why don't you use some WHQL drivers?
Did you remove the previous drivers properly?
 

b3ka

Junior Member
Oct 19, 2013
22
0
0
I removed the old ones.
WHQL drivers for this card, are there any? AMD is pointing me to the beta ones.
 

24601

Golden Member
Jun 10, 2007
1,683
38
86
@OP: Don't even try on this forum.

The AMD paid viral marketers will not give any real help.

Note how every single poster so far have foisted blame on you instead of helping.

Warning issued for thread crapping.
-- stahlhart
 
Last edited by a moderator:

wand3r3r

Diamond Member
May 16, 2008
3,187
0
0
@OP: Don't even try on this forum.

The AMD paid viral marketers will not give any real help.

Note how every single poster so far have foisted blame on you instead of helping.
What are you helping him with, much less what are you doing in the thread other than falsely accusing people of being paid shills? :thumbsdown:

I don't blame him for anything, just trying to figure out a couple things to verify.
 

Rvenger

Elite Member <br> Super Moderator <br> Video Cards
Super Moderator
Apr 6, 2004
6,284
5
76
no WHQL drivers for the R9 280x yet but I would certainly use the 13.10 betas over the 13.11!

OP also try just running 1 card at a time to see if it could just be a bad card. Try the 13.10 betas then try separate cards.




@OP: Don't even try on this forum.

The AMD paid viral marketers will not give any real help.

Note how every single poster so far have foisted blame on you instead of helping.


This is called thread crapping and don't listen to this garbage.
 

b3ka

Junior Member
Oct 19, 2013
22
0
0
Beta 13.11 played a couple of minutes in BF and it started to happen around 76 degrees with Voltage of 1.2000 and Power Limit +20%.

Gonna install 13.10 betas right now. Already test single card setup and they are not defected it seems.
EDIT: 13.10 drivers (it seems these are same as on ASUS website) make the entire game too choppy & just got a system hang. There is no framepacing option in 13.10 too.
 
Last edited:

lehtv

Elite Member
Dec 8, 2010
11,900
74
91
Also, aside from this, your PSU is not powerful enough for these cards as they are factory overclocked. Tahiti uses a ton more power when overclocked and the Matrix is 1100 factory OC'ed IIRC? One card alone can do near 500W power draw with 1.3V on the Tahiti, let alone two.
This is factually false. The 280X Matrix Platinum uses about 240W.



Running two cards will get you nowhere near 750W, and the Seasonic unit can be run at full power continuously. It's closer to 850-900W in peak capacity.
 

blackened23

Diamond Member
Jul 26, 2011
8,556
1
0
This is factually false. The 280X Matrix Platinum uses about 240W.

Running two cards will get you nowhere near 750W, and the Seasonic unit can be run at full power continuously. It's closer to 850-900W in peak capacity.
I don't really agree with that. I can dig up 280X reviews showing past 500W on a single card, you can be selective with the reviews if you want but the bottom line is that the 7970 and 280X are pigs with regard to power consumption once you over volt. From my rough memory, the Matrix is factory over-volted? Although I could be wrong. This also matches my experience with 7970s which I bought at launch (although i've long since sold them ), they used a TON of juice when over volted and overclocked. It all depends on which BIOS is used and to what level the card is being over-volted. In general, Tahiti is an absolute pig when over volted - and it can and will use a ton of juice, period. I do think the Matrix is factory over-volted, but correct me if i'm wrong.

That said, the main issue I see here is mGPU with 3 slot cards. I've tried this on several occassions with both nv and AMD cards and it without exception was never sustainable, ever. The top card will have temperature "creep" when you do high 3d application loads, it will become 20C hotter than the lower card slowly over time. And that eventually will cause TDRs and BSODs. I can't really think of any elegant solution for 3 slot cards to work in mGPU configurations - which is why reference or 2 slot aftermarket is always preferred for sli/CF configs. If you manage to get mGPU working with 3 slot cards sandwiched, they *might* work for 20-25 minutes or so , and then the temperature creep on the upper card will cause you to BSOD or TDR.

Maybe i'm jumping to conclusions here, but i've tried on many occassions to get 3 slot mGPU to work. And it just wouldn't. It would work for maybe 20-25 minutes max, as mentioned - and then fail over time due to the temperature creep issue I mentioned. The heat output from the bottom card, inside the case, will cause the upper card to become EXCESSIVELY hot after a period of time of heavy 3d application load. There is no way around this unless you use an open air test bench or something along those lines. Of course, it sounds like his direct issue may lie elsewhere with respect to CF drivers or something - but even when that issue is sorted out, it still won't work due to the temp creep. IMHO.

This is why my personal opinion is, to RMA one or both cards. Getting that (3 slot mGPU) to work will be an exercise in futility, especially when they're factory over-volted.
 
Last edited:

b3ka

Junior Member
Oct 19, 2013
22
0
0
13.101 betta CF performance is average, cards utilization between 60-80%

Maybe i'm jumping to conclusions here, but i've tried on many occassions to get 3 slot mGPU to work. And it just wouldn't. It would work for maybe 20-25 minutes max, as mentioned - and then fail over time due to the temperature creep issue I mentioned. The heat output from the bottom card, inside the case, will cause the upper card to become EXCESSIVELY hot after a period of time of heavy 3d application load. There is no way around this unless you use an open air test bench or something along those lines. Of course, it sounds like his direct issue may lie elsewhere with respect to CF drivers or something - but even when that issue is sorted out, it still won't work due to the temp creep. IMHO.
Actually played ~20-30minutes and difference between cards is MAX. 10 degrees.

Summing up as for now

13.11 - the best performance, but one card GPU usage drop around 10-15minutes of playing
13.101 - average CF performance, no framepacing in settings, no usage DROP as it's staying around 60-80 all the time
13.10 - average CF performance as with 13.101 + system hangs, no usage DROP as it's staying around 60-80 all the time
13.09 - good perforamnce (like 13.11), GPU usage drop and chopping after a couple of minutes of playing.
 
Last edited:

badb0y

Diamond Member
Feb 22, 2010
4,013
27
91
My advice is completely wipe the drivers of both vendors and start fresh with the latest drivers. I don't know what's the defacto driver removal utility nowadays but I used to use Driver Sweeper when I switched between vendors.
 

railven

Diamond Member
Mar 25, 2010
6,604
560
126
Easiest way to determine if it's a temp issue, load up 13.11 beta (drivers that gave you best load and experience) remove your side panel (so heat from cards is dumped quickly outside of case) and go see how long it holds.

I have a similar setup to yours (the Me in Sig), and on a 750W Corsair, I can run my cards at 1125/1575 @ 1.5v for 99% of my games.

Game runs fine at 1.175v too. I've never tried to see if they run at decent clocks with 1.225v. I get the feeling my PSU won't hold out for long.
 

b3ka

Junior Member
Oct 19, 2013
22
0
0
Easiest way to determine if it's a temp issue, load up 13.11 beta (drivers that gave you best load and experience) remove your side panel (so heat from cards is dumped quickly outside of case) and go see how long it holds.

I have a similar setup to yours (the Me in Sig), and on a 750W Corsair, I can run my cards at 1125/1575 @ 1.5v for 99% of my games.

Game runs fine at 1.175v too. I've never tried to see if they run at decent clocks with 1.225v. I get the feeling my PSU won't hold out for long.
Will try that too, right now I'm reinstall the whole system.
Could you tell me what are your idle/full usage temps with this setup?
 

ASK THE COMMUNITY