Sapphire R7 260x black screen!!!

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.

Aravind92

Distinguished
Apr 1, 2014
699
9
19,065
Bought a sapphire R7 260x 2gb(says oc edition but core and memory clocks are lesser than referrence) 25 days back.
Whole system is new.Got a black screen just when sitting on the desktop two days after i built the pc.Restarted pc,everythjing fine for 2 weeks,black screen again,all components running fine inside the cabinet,still black screen tried unplugging and replugging DVi cable,no go,monitor still gets signal after black screen.This was with the the driver which came along with the gpu.
Then installed 13.12 from amd webiste,black screen next day,just when browsing the internet.
Uninstalled that too and installed 13.12c from sapphire which had catalyst version 14.1,worked fine for 8 days,black screen again when watching a movie.Movie sound form the speaker turned into a wird long loop,still monitor was recieving signal.So uninstalled that too and installed 14.3 beta.Restarted the pc and when installiong arkham origins,black screen again,restated the pc and uninstalled that too and installed 13.10.Worked fine for one day and black screen again just after starting the system when browsing thorugh my folders on the hard disk.

Everytime i do a clean uninstall and boot into safe mode and run driver sweeper and only then i install another version.

When gaming this has never happened,No heating issue,max temp is 76 even with crysis 3.To be noted:I live in India,chennai.It is summer now and temps are between 35 and 40c outside.

Even tried leaving the system on for 5 minutes after black screen,display does'nt comeback,all components would still be running.I have to hard shut down and restart everytime this happens.

Even before i bought this card i knew many are facing this issue,with all manufacturers like asus,xfx,sapphire and all.Stll took a risk by going with the card.I just wanna know if it is hardware issue or a driver related problem.If its latter i can wait for a table version.Just wanna make sure it is not a hardware issue,if so i will have to rma this card.

I could not relate each occurance with another,happens completely random.Everytime happened under diferrent scenario_Only thing i could get out of those were like the gpu did not have lot of work to do.Was almost at idle.

Tested drivers:
13.20(worked fine for 2 days black screen,again worked fine for 2 weeks,black screen) which came along with the disc(for some reason this was released before 13.10 or 13.12)
13.10(good for 1 day)
13.12(good for 2 days)
13.12c from sapphire website.(good for 8 days)
14.3 beta(not even for 1 hour).

My whole system config is:
I3-3220
Gigabyte B75m-D3h
Corsair value select 4gb 1600mhz ddr3
Seagate barracuda 1 tb
Sapphire R7 260x 2g Gddr5
Power supply:Corsair VS450(UK version)34A on 12v.
 
Solution
From mtdew332:

"So after two weeks of searching for the cause / solution to my Black Screen issue, I think i've finally nailed it down.
With some help from overclock.net forums.
I'll just quote it here.

(This black screen from power transitions occurs because the memory clock jumps up to full speed (1250, etc), while the GPU voltage (which also controls the IMC--this is the problem) is STILL running at idle volts of 0.9xxx). That idle voltage is too low to handle the IMC, which is controlling the memory running at that speed (this is NOT due to defective memory modules btw), and thus the card black screens.)

The fix, is to increase the GPU voltage, which ALSO increases the idle voltage to the IMC as well. (previous...
I would think that the 12V dropping to 11.78V (although not terrible) isn't as stable as it could be. The acceptability limits are +- 5%. That is (when measured accurately) 11.4 to 12.6 V. If it is that new, I think I would RMA it. I would recommend a 550 to 650 watt from a good brand , if you go that route.

You can bring your PSU to a computer repair shop and have them check it for you. They can also check the graphics card.

If you do not have a case speaker, I would recommend one. They are really cheap tool at around $5. The motherboard beep codes are particularly good at indicating memory issues.
 

Aravind92

Distinguished
Apr 1, 2014
699
9
19,065
Oh is that so,i thought i don't have a problem as long as it would not go below 11.6.And i already got ll the components tested with a computer shop,but i suggested the whole config and had them make sure everyhting is fine.For computer shops here in india they would always go with the psu's which come along with cabints,so the shop at which i tested my components felt like this psu is great.Because he had just used zebronics before.And the voltage diferrence is just .1 between idle and under load.now while i have firefox running it is wavering between 11.952 and 11.888.When under 100% load both on gpu and cpu it drops to 11.78.Is that a real concer because as i already said i don't have the money to go for a new psu now,probably i can change it later if it goes even lower.
 

Aravind92

Distinguished
Apr 1, 2014
699
9
19,065
Should i definetely change it then..In the review that i posted he has the same votage as what iam having.I can at max rma this psu.Because i cannot shell out a penny for anything now.Is that really required?
 

Aravind92

Distinguished
Apr 1, 2014
699
9
19,065
Psu reached 11.664 now with arkahm origins.Stayed there for the whole time when i played it.As soon as i exit the game,came to 11.880.Now is this game stressing the components more than crysis 3?Again tried crysis,that too reached a low of 11.664.Is this a real problem now?\What does the tolerance of +/-5 mean in psu voltages?
 

Karadjgne

Titan
Ambassador
Usually that supposed to be 12v, +/- 5%, because the OEM uses cheaper standard resistors, mosfets etc that have a tolerance of 5%. Means its rated 12v output can be anywhere between 11.40v and 12.60v, which in my estimation is a rather loose regulation. The fact that you are reading 11.664v means you are still in tolerance levels but only just. What that means is in power terms, your components are pulling more amperage on the 12v rail to make up the difference in voltage. That can have repercussions.
 

Karadjgne

Titan
Ambassador
I would. If you het the same reading as what you saw before, I'd take it back to the store and swap it out, if only for the same part. If you are seeing voltages that low now, is be surprised if the psu lasted to the end of the 3 year warranty.
 

Aravind92

Distinguished
Apr 1, 2014
699
9
19,065
I would like to know one thing.What is the basic difference between mine and yours,as you say +/-5 you are in the other side of the tolerance,does that make a difference.Because yours should not go+.2 where it would fail and mine should go -.2 where it would fail,what is the basic difference?
 

jayadratha

Distinguished
Sep 17, 2009
427
0
18,790

I also faced same issue. here is my thread.
http://www.tomshardware.com/answers/id-2084976/monitor-black-hangs-260x-2gb-gpu-catalyst-driver.html

I'm also using corsair VS550. And I think it is not the problem of PSU. Cz When I play Battlefield 4 or crysis 3 the R7 260x OC 2GB gets a 100% load in core clock, memory clock & also 1.8GB of graphics ram is used. Card temp goes to 75%, So I think this time card used 100%. So if PSU have problem then in this time I should face the problem also, but after playing like this for 4-5hrs I didn't faced it. I faced this issue only when I just use my pc just for work not for gaming. Dont know why.

& yes I faced this one day then next day also then next 2days I didnt faced, then at 3rd day I again faced it 2times. and then until today (2days) I didnt faced it.
 

fruitbowl33

Reputable
Apr 4, 2014
3
0
4,510



Hey man, I JUST ran into the same problem. So I just installed the drivers on the cd, and waited. For a very
long time. Like it was a 10 minute wait. Then windows suddenly loaded. So try giving it some time.
 

jayadratha

Distinguished
Sep 17, 2009
427
0
18,790


I didnt understand. Did you wait when the screen was black & pc freezes?? & after 10min or so again display came and pc worked well??

 

Karadjgne

Titan
Ambassador
You wanted to know the difference between my PSU and yours. Simply, mine is built by Seasonic. The are among to very best when it comes to building a psu. They use top shelf components, use more regulatory components than just about anyone. The tolerances on these components is so tight, there is almost no ripple on the outputs. Not only does my psu have just about every over or under current/voltage protection you can put in a psu, but Seasonic tests their psus to higher standards than most. My psu well exceeds atx standards at any load and when they say it puts out a certain wattage, its accurate.
That's the difference, and worth every extra rupee.
It's also dead silent at idle, and even at @70% load, its still very, very quiet, that's how good the thermals are.

 

Aravind92

Distinguished
Apr 1, 2014
699
9
19,065
I did not want to know the quality difference between yours and mine.Yours is made of better quality components i know.If you can read my post again and give a clear idea about the difference between +5 range and -5 range that would be great.You have completely skipped my question for some reason.
 

Karadjgne

Titan
Ambassador
Ah, sorry, misunderstood the question. Most everything electrical or electronic requires a certain voltage to run and part of this reason is the amperage draw. If you look at the components themselves, more than a few you can see 600V max ratings, but very few give max Amperage or current ratings, most of the time that's an unknown amount. So, if the gpu requires 11.7v to run, its because the internal components like diodes and transistors etc. have tolerances that have to be met or switching doesn't happen. Give it 12.4v and they are happy as a clam, the current at the higher voltage is lower. Give that same component 11.7v and the current goes up to maintain the wattage, and this affects the entire gpu. End result is a severly shortened lifespan on anything that is under-volted, simply due to over-current stress.

 

Aravind92

Distinguished
Apr 1, 2014
699
9
19,065
Perfect this is what i was expecting,now everything is clear.So will i be safe with this volt for some time until i change to a better psu?

will i be safe with seasonic eco class 500w if i change in few months?

And my card would use more watts than normal if i give it lower volt right?
 

Karadjgne

Titan
Ambassador
You should be ok for that time, but no guarantees, and the seasonic is very good.

If you give the card lower volts, the amps go up, watts stay the same. P = V x I or watts = volts x amps, eg. 12w = 6v x 2a, if you lower volts then, 12w = 4v x 3a, amps go up.

the 260x is a 150w gpu, 75 from pcie, 75 from 6-pin, that equates to 12.93A at 11.6v, and 12.19A at 12.3v. On my Seasonic, at 100% load, I would see the same temp as you pushing 94% load, so you would exceed my temp if you pushed 100% for example, which would burn your card out eventually, or make you lower graphics
 

Karadjgne

Titan
Ambassador
The temp gradient isn't that shallow. 1A difference could be as much as 10*C easily. Its the difference between you having a great time playing BF4 at 81*C maxed out, and throwing stuff across the room because you just black screened by thermal shut down during the best game of your life when the gpu peaked at 91*C. But you are thinking in the right direction, and being in India, I know ambient temps there get pretty nasty at times.
 

Aravind92

Distinguished
Apr 1, 2014
699
9
19,065
Yes absolutely right,now its 40*c here..

And i don't have temps that high because with long hours i just hit 76 at max with crysis 3 and it then wavers between 74 and 75 mostly.Would bf4 heat my gpu more than crysis 3.

But the thing is iam not a fps lover so i would'nt play bf4 probably,just thought of saying.
 

jayadratha

Distinguished
Sep 17, 2009
427
0
18,790


I played BF4 & crysis 3 running GPUz in background. After playing 5-6hrs I checked the max temp in GPUz, I saw it was 81degree Celsius. Is this the throttling temp for R7 260x?? Will R7 260x shut itself down in this temp???

At the time of playing game I didn't get any black screen issue. I got the issue only when I do my normal works which don't give load to the card so much. So at the 100% load and 81degree Celsius card didn't shut down itself then why the card getting shut down at 0-1% load & 40-44degree Celsius temp??? (Considering gpu shut down is the reason of black screen & PC freezing issue. as you said earlier.)

& yes last 3-4days I didn't get any black screen issue.