PDA

View Full Version : 7790 Display driver amdkmdap stopped responding and has successfully recovered



mvedda
04-21-2013, 01:47 PM
Hi,

I am having some issues with my new SAPPHIRE 100356OCL Radeon HD 7790 OC 1GB DDR5 PCI-Express Video Card.

This card is an upgrade from my 5750 that had been running smoothly.

From time to time I will recieve the message "Display driver amdkmdap stopped responding and has successfully recovered." I receive this message while doing non gaming activities.

While playing world of warcraft(in windowed mode) and bioshock infinite(fullscreen) I have never once received this message.

I have done a bit of browsing on the web for a fix but all attempts have not solved my issue.

I was first using the drivers available on the AMD website but as this issue started to occur I switched to the drivers that came on the CD with the card. I have performed a full driver uninstall with drivesweeper in safe mode. My install of windows 8 is fairly fresh but was in place before the card was installed.

I added a TDRDelay of 8 to hklm\system\currentcontrolset\control\graphicsdriv ers

This seems to have made the issue occur less frequently as it will wait a little longer before timeout but it has not resolved the issue fully.

I have also underlocked the factory OC to standard settings but the issue still occurs.

Any help would be appreciated.



My system specs are : (this is a pre built PC with minor changes from gateway, Model: FX6840)

CPU: Intel core I7 860 @ 2.8ghz

RAM: 8gb DDR3

Graphics card: SAPPHIRE 100356OCL Radeon HD 7790 OC 1GB DDR5 PCI-Express Video Card.

PSU: 450w Max 500w peak. 18a on 2x 12v rails.

BIOS: American Megatrends inc. Version P01-A3

Motherboard: Gateway FX 6840

SSD: Intel 335 Series Jay Crest SSDSC2CT240A4K5 2.5" 240GB SATA III MLC Internal Solid State Drive (SSD)

CCC Version 2013.0215.1631.29608

EWP
04-22-2013, 06:16 AM
I'm just curious... Are you using FireFox when these crashes happen by any chance?? If you are you need to disable the hardware acceleration in both Flash player and within FireFox itself (Options, Advance Tab, Under "Browsing", and uncheck "Use hardware acceleration when available".)

mvedda
04-22-2013, 01:23 PM
These crashes are mainly occuring when using chrome actually. I do not use firefox as in my opinion it is straight horrible in windows 8. I would assume that the same hardware acceleration settings could be turned off tho. I will have to look into it once I get home from work.

Thank you very much for the suggestion I had not thought of that, but I think at least chrome has always been open in some fashion when these crashes occur.

Fallen_Crow
04-22-2013, 01:32 PM
check this:

CMD windows-> Net stop uxsms

mvedda
04-23-2013, 12:23 AM
Thanks for the suggestions guys.

@ Crow That service does not seem to run in windows 8. or at least it does not run on my PC. But i think i already have aero disabled.

I Rma'd the card thru newegg. Lets hope that in the week or so it will take to get another card that it wont have these issues.

pOLLYpOCKET
05-13-2013, 10:08 PM
I also had this problem. It never crashed when gaming, just regular browsing online.

I requested an RMA from Sapphire as Newegg wouldn't RMA it for me because I didn't have the UPC which was required for the mail in rebate... Go figure.

I will report back to this thread because I've tried every fix I can think of including registry settings and nothing worked. Hopefully a new card does.

Babylonia
05-14-2013, 07:52 AM
PSU: 450w Max 500w peak. 18a on 2x 12v rails.

Reading this thread. Just a thought.

Have you checked if e.g. the graphics card and 12v power of e.g. the motherboard or other devices are not hooked to the same 12v rail?
18 Ampere is a rather low value (maximum 216 Watt).
Maybe even for just the graphics card alone at full power drain.
Just hook up other devices (like hard-disks) to the other 12v rail.

In spite you have no problems playing games (that exhaust power drain), maybe the hardware acceleration used within a browser (e.g. for running 1080p Youtube movies at full screen) is just hitting the borders now and then for the PSU ???

Sapphire requirements for the HD-7790 is a power supply of at least 500 Watt

I have no HD-7790 model (just the HD-7770) but have no problems using hardware acceleration within a web-browser (in my case Firefox). My PSU: 620 Watt, one rail 12v 50 Ampere.

@pOLLYpOCKET
Also check you power supply if it meets the recommended specs.

pOLLYpOCKET
05-14-2013, 06:12 PM
This is the PSU I'm running: http://www.newegg.com/Product/Product.aspx?Item=N82E16817152041&nm_mc=KNC-GoogleAdwords&cm_mmc=KNC-GoogleAdwords-_-pla-_-Power**Supplies-_-N82E16817152041&gclid=CK3u8oOZlrcCFYU5Qgod4zcACw

Babylonia
05-14-2013, 09:42 PM
It has two 12 volt rails of 22 Ampere each (= 264 Watt each).
The same advice as for the topic starter. Be sure the rail that is used to power up the graphics card, don't hook up other devices or motherboard to the same outlet / rail.

pOLLYpOCKET
05-18-2013, 04:01 PM
I don't exactly understand what you're telling me to do. Are you saying not to connect the same cable that's connected to the motherboard to the graphics card?

Babylonia
05-19-2013, 01:28 AM
You have to look to the technical description of the power supply. As by the specs. the 12 volt connections are divided into two different groups (rails). Be sure the graphics card is connected to one group, other devices like hard-disks, DVD player, motherboard, to the other group.

It is just like the fuse box of a normal house power cabinet. Running a washing-machine and and electrical cooking plate at the same fuse group, you do get into trouble. You have to separate the connections into two different power groups.

For the power supply of a PC as in your case having two different 12 volt groups, you have to separate the devices that are hooked on them in a comparable way.

(Hope my English is not to bad, and you do understand what I wrote).

Rubenoa
05-19-2013, 10:08 AM
@pOLLYpOCKET, the question in here is, what else do you have connected to the **12v rail? Just the card?
And, guys, remember that those 22as on each rail are Max. Peaks, so combining both rails will not add 44a in a single rail, I think that PSU will deliver only 30A and sustained, so in the limit....

jrweiss
05-19-2013, 11:23 AM
432W or 36A according to the specs at http://www.raidmax.com/psu/rx_600af.html

That SHOULD be enough total power, if the rail loads are balanced...

Rubenoa
05-19-2013, 03:22 PM
Thatīs max peaks, no sustained values. The 30A I mentioned are sustained.... ;)
So, on the limits of what the card needs. If you have another piece of hardware attached and it requires **12v load, then the card is underpowered.

Regards"

jrweiss
05-19-2013, 09:48 PM
I couldn't find the peak vs sustained values in my brief visit to the web site, or anything other than the 432W I cited.

IMO, a quality PSU today should be capable of sustained operation at its rated output, with peak output ABOVE its nominal rating. Also, the 12V rating should be VERY close (90%**) to its total output.

Rubenoa
05-20-2013, 10:16 AM
You could always ask the manufacturer about it, or meassure it yourself with a multimeter and a little bit of ingenious.
Then, if you know about the formula for Watts is Volts times Amps, you should know also between the differences on Max. Peak and sustained/continuous values when talking about intensity.

And for your last statement, you said well, a QUALITY PSU. Enermax for example rates its PSU values in sustained values, and it is almost the only company doing so, the rest present their rates ALWAYS in MAx. Peak values.... which are way inflated when comparing those with the real sustained values, you should also check into that when doing some tests ;)

Regards"

jrweiss
05-20-2013, 12:20 PM
I generally don't bother to ask a mfgr what his ratings are based on, and neither does the average builder. I would like (though I know it's only wishful thinking) that they would use a standard and realistic rating... Also, I don't have the equipment to test every component I buy, nor do I expect a vendor would take one back if I were to fry one in some off-the-wall test.

OTOH, it appears Enermax is not the only one to rate their PSUs on continuous or realistic power output... Several reviews I've read (techpowerup, tweaktown, techspot) indicate the Kingwin Stryker 500 I have is capable of 600W...

BTW, where did you find the peak vs continuous output ratings for the OP's PSU?

Rubenoa
05-20-2013, 01:45 PM
Yeah, I know, it is a real pain in the ass for us the buyers to find out those details, but it is the only way to know about for real and also it will tell us about a particular brand or model, so I guess we will have to do our homework prior buying any products...
About the kingwin Stryker, again, that doesnīt tell us anything about what we need, right? What about its sustained values? A PSU that can reach 600W is like saying that this PSU can rise its Max. Peak ratings above average 500Wsī PSUs, but nothing about its sustained value...

Normally we know about them making our "homework" at home (testing them by ourselves) and comparing the results between 4 or 5 of us in the forum, so we could avoid bad readings or a bad test. Thatīs why we talk about it. Also, sometimes, as there are many models of PSUs out there, asking the manufacturer straight with an email give us a aprox. number of it (we prefer to rely on our tests, but...).

Again, I know this is a pain in the ass, but this is the way it is at the moment :(

Drtzekos
06-06-2013, 03:58 PM
I am facing the same problem.

A week ago I upgraded from 6790 to 7790 (2GB OC Sapphire). At start, i did not remove any driver, I just installed the new one over the old.. instantly I experienced very low fps in games that did not require a lot of performance (Like Dota-Warcraft3 or LoL).

Of course then I removed ALL drivers with their uninstallers and re-installed the new drivers.. Then although the performance improved in to normal levels, in random times during gaming (every 5-10') I would get a flicker for less than half a second which was more than annoying..

I removed again all drivers, this time using a driver-sweeper in safe mode and re-installed the new drivers . Since the "cleaned/sweeped" install I am getting the "Display driver stopped responding and recovered" message every half an hour two times one ofter the other.. (I did this 3 times already using 3 different sweepers) The odd thing is that it never happened during gaming but during browsing (with G Chrome). I didn't try to see if I am also still having this flickering during games..

Yesterday night after un/installing I also updated G Chrome (although it does not seem to be the problem) and went to sleep..

Later today I am planning to try

1) this --> http://support.microsoft.com/kb/2665946
2) Then if it does not work, i will try to increase a bit (1-2%) the voltage if possible (it was suggested in an other forum)
3) Increase a bit the minimum clocks (as the problems appears to occur only on "idle")
4) Format my pc and re-install everything from start
5) Rma my new GPU (i still have the old, I just changed because this one has **1 DVI port)

My configuration is:
OS: Windows 7 64Bit Professional
PS: 600W OCZ 80Plus
CPU: AMD FX 6100
MB: Asrock 970 Extreme 3
RAM: 2x 4 GB DDR3 Kingston @1600MHz
HDD: OCZ Vertex 3 120GB
Monitor: 3x Dell 2312HM (DVI/DVI/DP) ** one Toshiba TV (HDMI)

Alhon
06-07-2013, 11:12 PM
Having the same issue here. Oddly enough, It works fine on Bioshock Infinite at max settings, but has client-side lag and crashes on Star Trek Online: Legacy of Romulus.


My PS is a Single-Rail, (or so it is advertised), so I'm not sure if there's anything I can do on that part.

My build:

PS: Thermaltake TR2 700W
OS: Windows 7 64Bit Professional
CPU: Intel i5 3570K
MB: Gigabyte Z77-DS3H

jrweiss
06-08-2013, 02:47 AM
Neither of you mention which driver versions you have tried. While 13-1 and 13-4 each work for some people, 12-6 and 12-8 have also worked for others.

Once in a while, some people find that an older version works better than the "latest & greatest". I'm running 13-1 on 1 machine (stable) and 13-4 (NOT stable; neither was 13-1) on another. I'm still working to find the right mix on the 2nd machine...

In any case, a thorough CCleaner or DriverSweeper run after each uninstall is a good idea before installing new drivers.

Drtzekos
06-18-2013, 08:36 PM
Neither of you mention which driver versions you have tried. While 13-1 and 13-4 each work for some people, 12-6 and 12-8 have also worked for others.

Once in a while, some people find that an older version works better than the "latest & greatest". I'm running 13-1 on 1 machine (stable) and 13-4 (NOT stable; neither was 13-1) on another. I'm still working to find the right mix on the 2nd machine...

In any case, a thorough CCleaner or DriverSweeper run after each uninstall is a good idea before installing new drivers.

We are talking about a graphic card approximately 3 months old. In this time period there was only one version (if i am not mistaken) which supports this graphic card.. the latest..

jrweiss
06-18-2013, 09:47 PM
Sapphire lists 13.4 and 13.5 beta 2.

AMD cites a special release for the 7790 on http://support.amd.com/us/kbarticles/Pages/RadeonHD7790CatalystPerformanceDriver.aspx

From the info on the web pages, it appears all 3 are different. So, which version are you using, and from where did you download it?

If you haven't downloaded the special version directly from AMD, I recommend you do so.

Drtzekos
06-19-2013, 08:27 AM
Sapphire lists 13.4 and 13.5 beta 2.

AMD cites a special release for the 7790 on http://support.amd.com/us/kbarticles/Pages/RadeonHD7790CatalystPerformanceDriver.aspx

From the info on the web pages, it appears all 3 are different. So, which version are you using, and from where did you download it?

If you haven't downloaded the special version directly from AMD, I recommend you do so.

Well if you don't count beta (cause it is beta..), and you just go to AMD.com there is the option "7xxx series" and "7790 PCIE". Theoretically "7xxx" should support 7790 but when there is a "specific for 7790" you go for the specific one (logically there should be a reason to have a specific version, and I guess it is not only the profiles, which you can separately download)

So, I am using the amd.com "7790 special release".

(On: W7 Pro 64Bit, upgrading from 6790. I used driver sweeper and I meet the power requirements etc as I already described).


I am still facing the problem but a bit rarely, maybe once per day.. but unfortunately it happened once while gaming also..

On the first chance I will make a clean install (format and again from zero) to my PC, to see if the problem insists.. I guess I will bother with this problem at the end of next month due to some obligations.. Hopefully there is going to be a newer driver version that will save us some time..

Sai
06-19-2013, 09:38 AM
I'm having the same problem!

I've done everything I could think of and research to fix this problem with zero luck.

I've tried different drivers (driver sweeper and AMD cleanup before each)
-13.4
-13.5 Beta 1 and 2
-13.6 Beta 1
-and Currently, 13.6 Beta 2

I tried changing Windows TDR. I've tried disabling Aero (granted the crashes happen a bit less, but STILL happen. But, I didn't pay over 100 dollars to turn off Aero.)

I've tried overclocking & underclocking my CPU and GPU

I've ran all sorts of tests eg: memtest, OCCTPT, Prime95

And today, I did a fresh format and clean install of Windows out of desperation.
I was only able to Install windows updates and a few programs before my first crash.
http://img402.imageshack.us/img402/9589/pn98.jpg (http://img203.imageshack.us/img203/5341/7zof.jpg)
My webcam software being the last thing I installed, moments before my first crash.

To make things more awkward I crashed as I started to type this post.
http://img441.imageshack.us/img441/1922/9nj7.jpg (http://img29.imageshack.us/img29/8651/ysre.jpg)


My crashes tend to happen when I'm just doing desktop stuff. Chatting using IM or clicking around folders.
When it happens during gaming, it makes the game unplayable because framerates drop from 60**fps to 10-20fps until I reboot. It also causes some of my games to go gray-screen at unacceptable moments.

When I crash once, I tend to crash every few mins thereafter until I reboot.

Specs
Windows 7 Ultimate x64
Intel i5 3570k
Kingwin ABT-650mm 650W PSU
Sapphire Radeon HD 7790 Dual-X OC
Kingston HyperX 2x4 GB RAM

I don't know what to do anymore...

Drtzekos
06-19-2013, 10:14 AM
I'm having the same problem!

I've done everything I could think of and research to fix this problem with zero luck.

.......

My crashes tend to happen when I'm just doing desktop stuff. Chatting using IM or clicking around folders.
When it happens during gaming, it makes the game unplayable because framerates drop from 60**fps to 10-20fps until I reboot. It also causes some of my games to go gray-screen at unacceptable moments.

When I crash once, I tend to crash every few mins thereafter until I reboot.

......



The only thing in common (except of the problem of course :D ) is that we have both logitech camera drivers :p

I have AMD CPU and Asrock Mobo (other chipset other drivers).

Btw, after crash I do not experience bad performance, even in game it fully recovers (Chrome maybe needs restarting or refreshing of tabs)

Lets hope that AMD will solve the problem, or we have to change card again :/

One question: what for screen set-up do you have? (No of screens and resolutions)

I am running with 4@1920x1080 each 3 are extended desktop and the 4th is a mirroring of the 3rd (my TV.. for movies @ HDMI), most of the times I have the TV powered off and one of the other 3 turned to my laptop "video input" (so I am using 2, without deactivating the other 2 in settings)

Sai
06-19-2013, 10:42 AM
The only thing in common (except of the problem of course :D ) is that we have both logitech camera drivers :p

I have AMD CPU and Asrock Mobo (other chipset other drivers).

Btw, after crash I do not experience bad performance, even in game it fully recovers (Chrome maybe needs restarting or refreshing of tabs)

Lets hope that AMD will solve the problem, or we have to change card again :/

One question: what for screen set-up do you have? (No of screens and resolutions)

I am running with 4@1920x1080 each 3 are extended desktop and the 4th is a mirroring of the 3rd (my TV.. for movies @ HDMI), most of the times I have the TV powered off and one of the other 3 turned to my laptop "video input" (so I am using 2, without deactivating the other 2 in settings)

I'm using 1 Monitor @ 1280x1024 Using VGA/DVI adapter that comes with the package.

Fallen_Crow
06-19-2013, 04:54 PM
If you have the lastest GPU-Z look out, cuz there is a bug that makes your clock work @full speed.

Sai
06-20-2013, 08:59 AM
I should also add, nearly all my driver crashes happen during user input, such as mice clicks.
Be it playing a game or clicking around file folders.

Fallen_Crow
06-20-2013, 02:30 PM
Sai, try to clean EVERYDRIVER out there, looks like windows ships with some old 12 drivers which happened on my case.

Also if you go to hardware devices, in the "view hiddeen devices" do you find something called AODDRIVER?

Drtzekos
06-20-2013, 08:28 PM
Sai, try to clean EVERYDRIVER out there, looks like windows ships with some old 12 drivers which happened on my case.

Also if you go to hardware devices, in the "view hiddeen devices" do you find something called AODDRIVER?

I can see this aoddriver 4.2.. Should I remove it? :confused: Does it make a difference?

Sai
06-20-2013, 09:48 PM
Sai, try to clean EVERYDRIVER out there, looks like windows ships with some old 12 drivers which happened on my case.

Also if you go to hardware devices, in the "view hiddeen devices" do you find something called AODDRIVER?
I don't seem to have this.

This is what's listed under hidden.
http://img17.imageshack.us/img17/7018/o44m.png

ACM
06-22-2013, 06:14 AM
Built my friend a PC with this card & it's having the same problem.

Kinda ridiculous.
I tried everything.

I don't really want to RMA it for him but it looks like it will have to go that way.

ScarSpidey
06-26-2013, 07:10 PM
Built my friend a PC with this card & it's having the same problem.

Kinda ridiculous.
I tried everything.

I don't really want to RMA it for him but it looks like it will have to go that way.

I just got my GPU (7790 dual x oc 1gb) back from an RMA with no fix. I have also clean installed 3 times both before and after the RMA.
Here is my build:
FX-3850
ECS- A990FXM-A (V1.1)
2x8gb Patriot PSD38G13332H Signature
Coolmax ZU 600W
Kingston V-100 64gb (OS)
OCZ Vector 256gb
Seagate 3tb
2x500gb Seagate (Raid0)
HD 7790 Dual X OC 1gb

I have tried with only 2 SSDs connected (OS and Game installs) I've tinkered with the RAM voltage. Tried different games (LOL, Batman, FarCry3, Bioshock Infinite) and I get the same error with frame drops after the event. The games start off so buttery smooth and beautifully until this dreadful error. I do hope I have hair left by time this is resolved, if at all. I really just want this to be stable.

ACM
06-27-2013, 02:07 AM
I just got my GPU (7790 dual x oc 1gb) back from an RMA with no fix. I have also clean installed 3 times both before and after the RMA.
Here is my build:
FX-3850
ECS- A990FXM-A (V1.1)
2x8gb Patriot PSD38G13332H Signature
Coolmax ZU 600W
Kingston V-100 64gb (OS)
OCZ Vector 256gb
Seagate 3tb
2x500gb Seagate (Raid0)
HD 7790 Dual X OC 1gb

I have tried with only 2 SSDs connected (OS and Game installs) I've tinkered with the RAM voltage. Tried different games (LOL, Batman, FarCry3, Bioshock Infinite) and I get the same error with frame drops after the event. The games start off so buttery smooth and beautifully until this dreadful error. I do hope I have hair left by time this is resolved, if at all. I really just want this to be stable.

Sounds like I should just return it to Newegg for a refund & get him a 7850.

ScarSpidey
06-27-2013, 02:18 AM
Sounds like I should just return it to Newegg for a refund & get him a 7850.

Actually, I moved it to a different computer
Intel i5 gen 3
Intel Mobo
Hyper X SSD 128gb
550W PSU

It works fine -_-
I am currently using a GTX550 Ti to see if I still get an error. It might be the mobo (ECS a990fxm-a (v1.1)) or PSU (Coolmax ZU 600b) I really want to get this thing working because I've seen what its capable of on its own. Im willing to buy another PSU or mobo to make it work like its supposed to.

ACM
06-27-2013, 02:21 AM
Actually, I moved it to a different computer
Intel i5 gen 3
Intel Mobo
Hyper X SSD 128gb
550W PSU

It works fine -_-
I am currently using a GTX550 Ti to see if I still get an error. It might be the mobo (ECS a990fxm-a (v1.1)) or PSU (Coolmax ZU 600b) I really want to get this thing working because I've seen what its capable of on its own. Im willing to buy another PSU or mobo to make it work like its supposed to.

Thing is my old 6870 works on it.

ScarSpidey
06-27-2013, 02:30 AM
Im convinced it has to be driver issues. Either the mobos and gpus arent matching up correctly or there are some game drivers that are screwing the way this new card works. Im going to try all the same games on both comps with swapped gpus. Maybe i can post more information that will help.

ACM
06-27-2013, 02:46 AM
Im convinced it has to be driver issues. Either the mobos and gpus arent matching up correctly or there are some game drivers that are screwing the way this new card works. Im going to try all the same games on both comps with swapped gpus. Maybe i can post more information that will help.

That's what I'm thinking.
If you get anymore info please post it.
:)

ScarSpidey
06-27-2013, 07:01 PM
OK, I have tested different RAM 2x4gb Ballistix and a different power supply 550W Ultra and still got the error with the 7790 dual x oc 1gb. The Radeon is still working on the intel pc but not the amd pc. I am going to save up some cash and get another AM3** mobo, one that I wont cheap out on. This ECS a990fxm-a (v1.1) only has one BIOS available. Im not sure if thats a good thing. AS OF NOW its not the PSU, RAM, GPU, OS or SSD setup. It has to be the drivers, BIOS or mobo. BTW I have tried both Legacy and UEIF Booting. Both give me the same error but I am sticking with UEIF on Windows 7.

ACM
06-28-2013, 12:16 AM
OK, I have tested different RAM 2x4gb Ballistix and a different power supply 550W Ultra and still got the error with the 7790 dual x oc 1gb. The Radeon is still working on the intel pc but not the amd pc. I am going to save up some cash and get another AM3** mobo, one that I wont cheap out on. This ECS a990fxm-a (v1.1) only has one BIOS available. Im not sure if thats a good thing. AS OF NOW its not the PSU, RAM, GPU, OS or SSD setup. It has to be the drivers, BIOS or mobo. BTW I have tried both Legacy and UEIF Booting. Both give me the same error but I am sticking with UEIF on Windows 7.

That's weird because the build I did used an intel i5 & a nice Asrock mobo.
Must be motherboard bios problems or drivers like you said.

dubyaohohdee
06-30-2013, 02:00 PM
Has an RMA worked for anyone? Seems like this is a driver / hardware incompatibility issue.

Sai
07-01-2013, 03:39 PM
Has an RMA worked for anyone? Seems like this is a driver / hardware incompatibility issue.

I'd also like to know.
I am pretty sure it is much too late for me to try.
I'm still struggling to find a sweet spot to lower my crash chances. With the underclocking and turning things down, I feel like my 140 dollars was burnt in a fire and that I'd be better off using my onboard.

jrweiss
07-01-2013, 07:07 PM
Dunno 'bout the 7790 specifically, but I had to RMA my 7750 Ultimate twice. Original was DOA; second lasted 2 months; 3rd one is still running fine.

Has anyone here put in a trouble ticket and/or RMA request? If so, what was the response?

ScarSpidey
07-02-2013, 02:59 PM
Dunno 'bout the 7790 specifically, but I had to RMA my 7750 Ultimate twice. Original was DOA; second lasted 2 months; 3rd one is still running fine.

Has anyone here put in a trouble ticket and/or RMA request? If so, what was the response?

Just here to confirm that the RMA did not work for my GPU (7790 dual x oc). It still will not work with one of my builds however It does work, with no issues on another computer. I cannot confirm that the GPU (before the RMA) would have worked on the second build in the first place.

jrweiss
07-02-2013, 04:56 PM
If it works without problems in one computer, then the card is probably fine.

If the power supply in the other computer is good, then it likely comes down to a BIOS/driver compatibility issue. Unfortunately, it seems some of the new Z87 boards have issues with several high-performance components, including SSDs and GPUs. OCZ forums have a discussion about Asrock Z87 boards and Sandforce 2xxx controllers. I just built a rig on a MSI Z87-GD65 (BIOS 1.2) and i7-4770K, and it has no problems with RAID 0 SSDs or my HIS 7750 with Cat 13.4...

Jaraju
07-03-2013, 08:01 AM
I'm using Sapphire HD 7790 1GB GDDR5 DUAL-X OC and i'm experiencing driver crashing/recovering issue while on Windows but never while I'm gaming. Anyone found a solution yet for this problem?

jrweiss
07-03-2013, 02:49 PM
Have you opened a trouble ticket? The more info the techs get, the easier it will be to fix...

Lt.Catscratch
07-05-2013, 10:23 AM
Did any of you try bare minimum driver install yet ? Just Driver, or Driver**CCC ? I read somewhere that the not-so-brilliant amd codecs, transition, drag&drop etc bits were causing this. Especially on browsers with flash and videos where hardware acceleration is used.

PS: And i think there's really really very very few PSU s on the market that can't run 7790 anymore, i mean you really have to try to find a psu that won't work. It needs around 90watts for itself. 75w already comes from the pci-e slot and ~15w from the pci-e power, no hard work for any psu.

Maynardthecat
07-05-2013, 11:13 AM
I am having this problem on my 7790 too. I did an rma through newegg and I still have the same problem with my new 7790. I have also tried this GPU in a different computer but still had the same problem.

MCardona81
07-05-2013, 02:33 PM
I'm having the same problem oddly not when i'm running an game. I have noticed that when the driver fails the memory clock get stuck at 150 mhz and wont go up when the gpu activity is at 100% i have to reboot in order to play a game with good fps

Before diver fails:

http://s20.postimg.org/5rtyx40i5/before.jpg


When driver fails:

http://s20.postimg.org/r2rj1diml/error.jpg

When memory clock wont go up

http://s20.postimg.org/7yy7l15sd/after.jpg


I hope that this can be fixed fast :(

MCardona81
07-05-2013, 06:26 PM
I opened a trouble ticket and this thee response i get :


"Looks like its hardware related. Please provide a copy of the purchase receipt and send it to tech@althonmicro.com for RMA request."

Sai
07-05-2013, 08:06 PM
I opened a trouble ticket and this thee response i get :


"Looks like its hardware related. Please provide a copy of the purchase receipt and send it to tech@althonmicro.com for RMA request."
Guess I'll have to try this as well.

Shakaza
07-06-2013, 10:31 PM
I've been experiencing the exact same problem, except, when it recovers, the card won't fully utilize the pci-express slot it's in. It only uses pci-e x1 v2.0 instead of pci-e x16 v2.0. GPU-Z says it uses those less powerful modes to conserve power. After that, it no longer stops working, but I can't play games with a good FPS. Obviously.
My specs are as follows:
AMD FX-6100 @3.3 GHz
2x4 GB DDR3 1333 MHz RAM 1.5V (I dunno the brand)
Sapphire Radeon HD 7790 Dual-X OC 1075 MHz Core/6400 MHz Effective Mem. Clock
GIGABYTE GA-78LMT-S2 Motherboard
LEPA N Series N500-SA 500W ATX12V PSU 12V@34A
Toshiba DT01ACA100 1 TB HDD
Windows 8 64-bit
Catalyst 13.4
(I hope that's enough for you people. O_o)

A couple nights ago, I was able to avert the problem of not being able to fully utilize the pci-express x16 2.0 slot it was in by rolling back the driver, but I can't do it now. (I ruined it by reinstalling them because the CCC kept crashing. Yet it still does.) When I was playing Black Ops 2 during that period, I was monitoring the temperature and wattage using GPU-Z so I could assure myself that they were within reasonable levels. I did notice, though, that the wattage of the card would be almost constantly changing. It would fluctuate between 80 and 90 watts or so. Could this possibly be a sign of insufficient power? I'm totally hopeless when it comes to understanding power supplies. xD Also, when the computer first starts up, the card will perform optimally for a few minutes, before the performance returns to what I described before my specs.

jrweiss
07-07-2013, 12:17 AM
GPU-Z v0.7.2 also has a render test ("?" icon right of Bus Interface block) that loads the GPU in an attempt to shift it back to full rated speed and bus width. Use the Sensors tab to confirm whether the core & memory clocks and fan respond appropriately. Then shift back to the Graphics Card tab to confirm the PCI bus width. Close the Render Test window and look for the downclocking again. If yours doesn't shift up and down properly (PCI bus width will remain at the highest detected until GPU-Z is closed and restarted), it may be a hardware problem.

You may also want to check your BIOS Setup to ensure you don't have anything set that would adversely affect GPU performance.

In the past I have seen registry hacks that will disallow the power-saving downclocking, but I can't recall where they were posted or which cards they apply to.

Shakaza
07-07-2013, 12:42 AM
I should have mentioned that I tried the render test several times. When I do it now, it frequently switches between pci-e x1 v1.1 and pci-e x1 v2.0, and the difference is very noticeable in the test's FPS. Also, I know that the core and memory clock speeds do increase when doing the render test, and return to normal afterwards. I checked. And, I know that I haven't changed my BIOS settings in quite a while, much before I switched to this GPU.

What I'm going to do is contact the tech support for Sapphire. If I don't get an acceptable solution by Monday, I'm going to return the card to Newegg and try a different one, perhaps a Radeon HD 7850 or another brand of Radeon HD 7790 if the problems don't carry across brands.

Ale9HAck
07-08-2013, 01:03 AM
Hi, I have the same problem. It is not hardware problem
Testing in Windows 7/8 32/64 bit, it's the same
Testing in Debian and Ubuntu x64. with the proprietary driver. run fine. no problems ....
this leads me to think it is a serious problem in windows drivers.

ScarSpidey
07-09-2013, 05:00 PM
-_- The GPU is back to error on the Intel PC. Originally, I had it running on an AMD 8-core when it started failing. After a lot of troubleshooting, I ran it on an Intel pc and it worked until I played a tasking game (FC3) now the errors are back randomly. So here is what I can gather. As soon as you use it to its potential, the drivers are corrupted until you move it to a different PC, install it and it works until you use it to its full potential again. Hardware/Driver error. The temps stay around 55(GPU: Dual X OC) and 20(CPU: Noctua NH-D14) under a full load. I've tried different Hardware combinations, both Intel and AMD with the same results. The Temps are fine and the PSUs are more than enough to support it. All BIOSs are up to date.
Ive tried:
AMD_Radeon_HD_7790_12.101.2.1-130416a
13-4_vista_win7_win8_64_dd_ccc_whql

On top of this, I even tried running a GTX 550 Ti along with the Radeon HD 7790 in different configurations ie: the main display on the GTX and visa versa. The GTX works fine either playing games or not while the Radeon gives me the error.

jrweiss
07-09-2013, 09:36 PM
Another possibility is that the VRAM or other component on the card is overheating. I've seen a couple reviews (Tom's Hardware?) where the reviewer received a card with the heat sink was only firmly attached to the GPU, and other parts overheated...

Sai
07-10-2013, 12:24 AM
Another possibility is that the VRAM or other component on the card is overheating. I've seen a couple reviews (Tom's Hardware?) where the reviewer received a card with the heat sink was only firmly attached to the GPU, and other parts overheated...
I'm still hoping it is just a driver issue that will be fixed. (Yeah, right.)
I'm still on the email part of my RMA request, currently looking into getting Microsoft Office so I can fill out the .doc they emailed me... -_-

I feel this is going to be a huge pain and I'll be without a GPU for a few months during said RMA. Took a week just for the email.

Ale9HAck
07-10-2013, 04:21 PM
DUAL X run in full load.... Iīm confused uxu

ScarSpidey
07-13-2013, 07:15 AM
DUAL X run in full load.... Iīm confused uxu

Im not sure if you are referring to my post but if you are, the numbers I posted are degrees Celsius at full load, meaning 100% or close to it.

CoolerKing
07-14-2013, 02:13 PM
Hello,

I'm having the same driver crashing issues with my Sapphire Radeon HD 7790 2GB GDDR5 OC and Windows 7 x64...

I've tried 3 different drivers from the AMD website:
13.4 (12.104)
13.6 Beta 2 (13.101)
AMD Radeon™ HD7790 Catalyst™ Performance Driver

ps. I've only installed the Display Driver in all these cases, so nothing else that comes in the driver package.

All these drivers have the same issue. In addition, the last driver gave me another issue: If I power off the monitor, leave the PC running for a while and then come back and turn on the monitor, I would find that the PC was locked up and the monitor remaining in standby, unable to receive image from the graphics card.

I've already had 3 crashes today, and that's just when running normal desktop apps (Chrome etc.)... I've owned several Radeon gpu's in the past, including Sapphire ones, and I never experienced such problems before.. It's either a driver issue or a hardware issue, but it seems from searching on google that people with different brands and different versions of the 7790 are all having these problems, so it's most likely a driver issue. I hope that Sapphire/AMD will pay attention to this issue ASAP.

Sai
07-18-2013, 07:37 AM
I'm still messing around with different configs and settings before I switch to that last resort. I REALLY don't want to deal with that.
As of now, I have went two days without a crash.

I switched out from using my PSU's 6 pin PCIE to the Molex-to-6-Pin connector that came with the card. (While I did crash with this still at my default settings, before changeing below.)
With that, I also set CCC's AMD OverDrive at default for everything except power control settings @ 20%.

Moments when I feel like I will crash, only my mouse locks up for a couple of seconds. (sucks in FPS but way better than a crash!)


Knowing my luck, I will crash soon after posting this...

Day 3: Crashed. :|

MCardona81
07-19-2013, 08:24 PM
Well, i went to the store when i get my card, i returned the card , they gave me a new card, with the new card I didn had any problems so far so i guess that the previous card was defective. I didn had any driver issue or any crash so far , on gaming the new card runs great

Sai
07-24-2013, 11:28 AM
I made it to the final stages of the RMA process, but it seems I won't be able to do that.
Going by what it says, I need to ship them the card out of my own expense.

I've never once used postal service, much less know what to do, nor want to pay for it.
Yay.

The biggest thing that would scare me is, if I was able to send it back, they would return it saying there isn't anything wrong with it.

ScarSpidey
07-24-2013, 03:24 PM
^This usually isn't the case as it will be my second RMA to be sent. HOWEVER, I just installed a second screen and what do you know, it hasn't given me the error. I hope I didn't just jinx myself but out of curiosity, are any of you guys running multiple screens on this GPU? (7790 dual x oc 1g)

Fallen_Crow
07-24-2013, 04:54 PM
Sai you could test this:



Quote:
Originally Posted by jazzie View Post
my vga is POWERCOLOR HD7790 TURBODUO OVERCLOCKING FROM factory

so you mean 2d voltage or 3d voltage?

i can change my voltage in CCC right ? in tab Graphic overdrive!!
No for the voltage you will need Msi afterburner or Trixx.

Install one of theses and just give the voltage a tiny increase **6 and test... They been loads reports of factory overclocked cards needing a voltage increase to make it more stable.

Sai
07-24-2013, 09:00 PM
Sai you could test this:

Using Trixx and Afterburner, none give me the option to change voltages. It is locked/gray (MSI AB) and does not even display (TriXX)
I have tried checking unlock settings in option as well.

jrweiss
07-24-2013, 11:18 PM
I made it to the final stages of the RMA process, but it seems I won't be able to do that.
Going by what it says, I need to ship them the card out of my own expense.

I've never once used postal service, much less know what to do, nor want to pay for it.
Yay.

The biggest thing that would scare me is, if I was able to send it back, they would return it saying there isn't anything wrong with it.
Most likely, once you get an RMA, they will send a different card once they receive yours. I doubt the RMA vendor does testing & repair on the card before they send out a new one.

If they sent you an RMA, they sent instructions on how to do it. Use DHL, UPS, or FedEx so you can track its progress.

eduardoturata
07-25-2013, 12:31 AM
Well, i went to the store when i get my card, i returned the card , they gave me a new card, with the new card I didn had any problems so far so i guess that the previous card was defective. I didn had any driver issue or any crash so far , on gaming the new card runs great

Please let us know how it goes since many of us are having these issues. I just got mine last sunday and i have just a few days to RMA it with the local dealer. After that i will have to international RMA it, with expenses...

My dealer offered me to take another model instead but i really liked the look of this 7790... so i wanted know if i should try to get a working one or get another model.

Are your exchanged card still running smooth? no more issues?

Thanks!

eduardoturata
07-25-2013, 12:40 AM
Using Trixx and Afterburner, none give me the option to change voltages. It is locked/gray (MSI AB) and does not even display (TriXX)
I have tried checking unlock settings in option as well.


The latest version of HIS "iTurbo" (1.4.4) worked on mine to unlock voltage control. Actually this was the only one that worked for me.

Didnt need to setup anything, the voltage control option was there from the start. This might help you configure your profiles with a more stable voltage.

Good luck!

PS: i just bought mine here in Brazil 4 days ago. Same issues driver chrashes, 2d clock popping mid-game, audio-problems, mouse cursor enlarged, PC freezes .... i simply got all issues reported for this card, but all in 1 card! lol mine must be the worst one... i hope.

MCardona81
07-25-2013, 01:10 AM
Please let us know how it goes since many of us are having these issues. I just got mine last sunday and i have just a few days to RMA it with the local dealer. After that i will have to international RMA it, with expenses...

My dealer offered me to take another model instead but i really liked the look of this 7790... so i wanted know if i should try to get a working one or get another model.

Are your exchanged card still running smooth? no more issues?

Thanks!

The new card is running fine, no driver/black screen issue, I have been playing a lot of games and the all run fine with no issues, it's been almost a month since i had to change the card. I have to be clear, they gave me another Sapphire 7790 Dual-X OC and no issues so far

eduardoturata
07-25-2013, 02:21 AM
The new card is running fine, no driver/black screen issue, I have been playing a lot of games and the all run fine with no issues, it's been almost a month since i had to change the card. I have to be clear, they gave me another Sapphire 7790 Dual-X OC and no issues so far

Thanks a lot mate! Thats exactly what i wanted to know.

Yes, sorry i didnt make it clear you are right, mine is also a Sapphiretech HD7790 DUal-X OC (1GB). I have it for less than a week and i got several driver crashes and black screens. Seems i am the worst case so far, mine also gives me audio problem (i think when the TV comes out of sleeping state, the audio driver does not recognize it) and from time to time my mouse cursor becomes pixelated and like twice or 3 times the original size... and not only when browsing, but also inside games (like in that WoW "hand" pointer... becomes a huge glove lol).

Ive read about these other issues from other 7790 owners but seems my GPU is PREMIUM and got all the bugs in one :p. Anyways, thanks a lot for the input, seems like most people who RMA to Sapphhire get their same card back? Thats what it sounded like from other users, so probably better chances to really trade the card is with the local dealer, like you did.

Even tho i really loved the card's design and performance, i think i will choose to trade it for another model... maybe its not worth the chance of getting another/same faulty card.

Thanks for the quick feedback again, cheers

jrweiss
07-25-2013, 05:59 AM
I had to RMA 2 7750s, and got different cards back each time.

eduardoturata
07-25-2013, 06:37 AM
I had to RMA 2 7750s, and got different cards back each time.

Thats good to hear, i read other people say it otherwise.

By the way guys, that issue that happens in game, it seems that for unknown reasons the 2D clock is activated out of nowhere while we are running a 3D application.

For me it worked turning the 2D clock off.

Seems that this issue is gone here. Also disabling the 2D will force it to go back to 3D mode with no need to restart the PC, in case you got stuck in 2D mode.

I mean that "Disable 2D clocks" option available within Trixx, iTurbo, AsusTweak, Afterburner etc. Maybe "Disable ULPS" can be helpful too since it is power related (Ultra Low Power State).

I also believe this crash/recovery issue while NOT using 3D applications
have something to do with the power-saving/boost technology.

My bet is that for some reason at some point when changing between p-states, the frequency (speed) is being updated BEFORE the correspondent voltage power, leading to crash. I dunno, but for me it seems it has to do with under-voltage.

Will RMA my card tomorrow, lets see...

CoolerKing
07-28-2013, 04:56 PM
So is this really a hardware issue or a driver issue? If it's hardware, then what's the problem?

I bought my 7790 from someone who won it in a contest, but I don't know where he won it.. Will I still be able to RMA my card, does anyone know?

jrweiss
07-28-2013, 06:01 PM
You'll probably have to submit a trouble ticket to find out...

eduardoturata
07-28-2013, 09:28 PM
You'll probably have to submit a trouble ticket to find out...

How do i do that? I am looking for it but cannot find where i can open one. Thanks

pOLLYpOCKET
07-28-2013, 09:45 PM
https://support.sapphiretech.com/presentation/tickets/tickets_list.aspx

Go there and open a ticket. I'm getting my card RMA'd.

CoolerKing
07-31-2013, 08:08 AM
Instead of obtaining an RMA, I received the following instruction through the ticket system:

Increase power control to plus 10 in CCC.

Can anyone verify if this works?

eduardoturata
07-31-2013, 08:25 AM
Instead of obtaining an RMA, I received the following instruction through the ticket system:

Increase power control to plus 10 in CCC.

Can anyone verify if this works?



Ok i read the older posts and saw you back there. Well let me tell you from the experience i had with my card while i had it (i RMA it back to the online store i bought it, still waiting)

Increasing power control to "plus10%" in CCCenter MAY "fix" the problem. I am not sure how it all works, i only know its power (voltage) related, and from my POV SAPPHIRETECH should release NEW BIOS for us, with fixes for these crashes.

All solutions other than a software update, are actually only "half-fixing" the problem, using manual tweaks.... If im not wrong, 7790s from other brands had the same issue and had it fixed with new BIOS releases.


Anyway, if the "plus10% power" does NOT work for you, here is my advice (other than RMA or asking for a new bios)

Download/install any of the following softwares that lets you create/manage profiles, with different clock/voltage settings: Afterburner (MSI) ; ITurbo (HIS) ; Trixx (Sapphire) ; PrecisionX (EVGA) ; GPUTweak (ASUS).

Now you will notice you can create different profiles and you should use the factory settings, but with a small increment of voltage.

If lack of voltage is your problem , then the plus10 in CCC or the method above will fix it.

NOW if the crash problem remains, then i must say you have a problem with the card entering 2D mode when it should stay 3D or high performance mode.

To fix that, just go into AB/ITurbo/Trixx etc and check that option "Disable 2D clocks" and this should do it.

If the crash problem STILL remains , then you should also check that "DIsable ULPS" (Ultra Low Power State).

If none works, you should set a 3d clock/voltage, store it in all profiles and also check the "force constant voltage" option.

Note : You might need to "unlock voltage control" depending on the software/version you are using, and for some of them, the option is hidden/locked (just google and you find more on how to enable the voltage control)
For me, the software which worked the best was the "iTurbo" from HIS.

I had my card working fine after these tweaks, but i still had an issue with my monitor (Led TV) similar to what you have. After the computer had powered off my tv by timer, when i turned it on again (by moving the mouse), the sound (HDMI) would not come back.... and i also had some strage issue with my mouse pointer sometimes becoming twice the size it should be. Anyway, i believe the crashes are Bios related and the monitor/sound/mouse issue is Driver related. Since some issues appear on other brands cards, these are AMDs fault. Some issues might be Sapphire's aswell, and it is responsible for providing us the ways to use its product anyway.

Hope this helps, let us know how it goes

CoolerKing
07-31-2013, 09:59 AM
Thanks for your assistance, but please allow me to respond to your original post:


I dont understand, cant you verify it yourself? Did you close the deal already, got the card? wich specific issue are you having?

Dude be more clear, you did not tell us what issue you got. Sounds like you dont even have the card yet is that it?

The reason I can't test it myself atm is that I haven't had the crash for some days now.. The fact that it's power-related might mean that it only occurs if there's heavy desktop use.. My problem was the same as everyone else here: The screen would hang, I would get a windows notice that the driver has stopped responding, and after that the videocard seemed to be locked in some kind of low performance mode until you would reboot..

I also don't have CCC installed by default.. Frankly, I've never had any use for it before.. It seems ridiculous that we would have to do our own tweaks with a tool to stop it from crashing, so I really hope that this is just a temporary solution and that they will either release a BIOS update that will fix this problem permanently, or start offering real RMA warranty.

eduardoturata
07-31-2013, 01:36 PM
Frankly, I've never had any use for it before.. It seems ridiculous that we would have to do our own tweaks with a tool to stop it from crashing, so I really hope that this is just a temporary solution and that they will either release a BIOS update that will fix this problem permanently, or start offering real RMA warranty.

Yeah i agree...

Fallen_Crow
08-01-2013, 06:52 PM
Sai and other guys, can you try this?

Resolved issues:
Resolved crash when enabling AMD CrossFireX™ on some AMD 970 platforms
Improved stability for AMD Radeon HD 7790
Resolved intermittent mouse cursor corruption when rapidly moving the cursor across window borders

It's included on the last beta driver (13.8) from AMD page.

Sai
08-03-2013, 02:50 AM
Sai and other guys, can you try this?

Resolved issues:
Resolved crash when enabling AMD CrossFireX™ on some AMD 970 platforms
Improved stability for AMD Radeon HD 7790
Resolved intermittent mouse cursor corruption when rapidly moving the cursor across window borders

It's included on the last beta driver (13.8) from AMD page.

I downloaded it yesterday. I'm going to see how things go and report back here.
I am hoping very hard this is THE fix.

EDIT: It has been over 2 days so far and I've not crashed yet.
Things are looking up but I have that type of luck where I jinx myself. Let's see how things go a bit longer.

Drtzekos
08-05-2013, 05:37 PM
I would like to ask, which uninstaller(s) are you using (@sai)? and if you still run without a crash?

MCardona81
08-07-2013, 08:05 PM
I would like to ask, which uninstaller(s) are you using (@sai)?

try to use "AMD Catalyst Un-install Utility" you can find it in the download section

Sai
08-10-2013, 02:59 AM
I would like to report, I've not had a single crash since installing the new 13.8 Beta drivers.

MCardona81
08-10-2013, 05:30 AM
I would like to report, I've not had a single crash since installing the new 13.8 Beta drivers.

Niceeeeeeee !!
you have been struggling with that issue ofr a long time !

ExzeHD7770
08-11-2013, 03:36 AM
I would like to report, I've not had a single crash since installing the new 13.8 Beta drivers.

Can u give a link for the driver?, im struggling whit this error and im driving mad whit the ingame crashes

Fallen_Crow
08-11-2013, 12:08 PM
Windows 7 Beta 13.8:

http://www2.ati.com/drivers/beta/amd_catalyst_13.8_beta.exe

Caique
08-12-2013, 12:18 AM
Hello guys, im just registered here on the forum because i'm having the same issue ("Display driver has stopped working and bla bla bla)
Please let me say that i'm from brazil,so my english may sucks,have mercy. :D
I recently purchased the HD 7790 2GB OC and i'm facing the same issues,thank god i found this thread.
I was getting frustrated with this as i buyed this card and right from the start facing issues with it.
I've already tried some things that i saw on some sites from people having this problem and none of them worked :/ (Fresh windows install,edit TDR in registry,CD Drivers,latest drivers,old drivers)
i just send a email to sapphire explaining the issue,hoping for a RMA (though i did not want to come down to this)
Its good to see that @Sai seems to have his problem solved by installing the 13.8 beta drivers. i just intalled it now and i will keep updating here if the problem persists or not. I hope Sai keep us informed about his progress too.
The last thing i didn tried yet was increasing the power control by 10%.If all of this does not work i don't know what to do :/
Lets test now these drivers...

ExzeHD7770
08-12-2013, 12:29 AM
well.. i try the latest beta drivers and go worst, now i have blue screen :(. Im going to cd drviers and hope RMA replace my card, but i dont know...

Jaraju
08-12-2013, 09:58 AM
Downloading the new drivers now! Hopefully this fixes the issue once and for all! *fingers crossed*

Caique
08-14-2013, 06:23 PM
Well i like to report that it has been days and not a single crash happened after installing the 13.8 beta drivers.
i just got the reply from saphire and they said to test on another machine to see if the problem persist -_-.... Well i dont have another system to test :/
Also i forgot to mention that this problem only ocurs when doing normal stufff on windows.
If a month pass and the problem don't appear again, im gonna call this fixed. If not i think RMA is the only way :/

PS: my english sucks, have mercy :D

Jaraju
08-19-2013, 10:31 AM
I would also like to report that there has not been a single driver crash after installing the 13.8 beta drivers. :)

Devvy
08-22-2013, 03:07 PM
well.. i try the latest beta drivers and go worst, now i have blue screen :(. Im going to cd drviers and hope RMA replace my card, but i dont know...

Having the same issue with both 13.8 beta drivers.. Not sure if I should RMA or not when 13.4 and 13.5 beta2 are atleast somewhat stable for me..

Caique
08-22-2013, 03:14 PM
Having the same issue with both 13.8 beta drivers.. Not sure if I should RMA or not when 13.4 and 13.5 beta2 are atleast somewhat stable for me..

13.8 Beta2 is up,give it a try,who knows? good luck :D

Devvy
08-22-2013, 05:37 PM
13.8 Beta2 is up,give it a try,who knows? good luck :D

I tried beta2 drivers the day it went up and still get a blue screen within 5-10mins doing normal desktop stuff..

pOLLYpOCKET
08-30-2013, 10:44 PM
Hi guys, just wanted to update my experience. If you notice, I posted in the beginning of this thread. I tried all the fixes including voltage control, windows timing fixes, beta drivers, and everything. Nothing ever worked.

Well I'm happy to report that I got my card RMA'd and have not had one issue with the replacement card at all. So if you're still having issues, get it replaced. It was really fast turnaround time for me. I had a replacement back within the week.

Devvy
09-06-2013, 04:24 PM
Tried the new 13.10 beta drivers and still blue screen.. Gone back to 13.5beta2, requested an RMA but still hesitant about filling it out..

jrweiss
09-06-2013, 07:33 PM
Why hesitant? Trying 1 set of new drivers should be enough... If the card isn't working, try another one!

MCardona81
09-11-2013, 04:52 AM
Hi guys, just wanted to update my experience. If you notice, I posted in the beginning of this thread. I tried all the fixes including voltage control, windows timing fixes, beta drivers, and everything. Nothing ever worked.

Well I'm happy to report that I got my card RMA'd and have not had one issue with the replacement card at all. So if you're still having issues, get it replaced. It was really fast turnaround time for me. I had a replacement back within the week.

glad to hear that, i had to return the card too, with the new one , no issues

Devvy
09-19-2013, 10:49 PM
Switched to the 13.9 driver this morning and can say I haven't had a single problem that I had using beta drivers.. No crashing, no blue screens, and even jumped up 13-15fps in FFXIV

*Edit*
I jinxed myself.. Was working fine all morning and now it just blue screened while watching a youtube vid in fullscreen.. Restarted and now my core clock is stuck at 300 and won't move >_>

MCardona81
09-20-2013, 12:40 AM
Switched to the 13.9 driver this morning and can say I haven't had a single problem that I had using beta drivers.. No crashing, no blue screens, and even jumped up 13-15fps in FFXIV

*Edit*
I jinxed myself.. Was working fine all morning and now it just blue screened while watching a youtube vid in fullscreen.. Restarted and now my core clock is stuck at 300 and won't move >_>


I had the same problem with the 13.9 i install it and the gpu clock got stuck at 300mhz at 100% activity, so i swtiched back to the 13.4 and the gpu clock issue has gone , but now the gpu-z and the catalyst control center says that i have the 13.9 version installed

jrweiss
09-20-2013, 01:19 AM
ALWAYS uninstall the old drivers & CCC AND clean the Registry with CCleaner or Driver Sweeper or similar before installing new drivers!

mucormucedo
10-11-2013, 08:50 AM
I have a HD 7750 2GB DDR3, I have it for 10 months, tried Windows 7 and 8 x64, all the new drivers, last one 13.11, TDR delay, voltage, memory clocks over/undurclock, CCleaner and Driver Sweeper, same error. Crashes are just in games, some games give errors more often than others. I got some game crashes in Ubuntu but I can't be sure if it is the video card(Dota 2 and Don't Starve) both are native linux clients. If I emulate, for example, Mass Effect I have no problems but in Windows the crash will come for sure in minutes. I played SF IV today with no problems in training stage, when I played arcade, immediately crashed. I did FurMark stress test for hours, no crash. I had more problems with Windows 8 than W 7, crashes were more often. Not sure what to do about "Display driver amdkmdap stopped responding and has successfully recovered" is it hardware error or driver error, still a mistery for me.

colinscrap
10-11-2013, 10:34 AM
Forgive my intrusion; I read a lot of the thread (but not all) and am an amateur on tech subjects; however, I wondered if any of you ever considered that your CPU is faulty?

The reason I say this is that when discussing issues on my first PC build in 10 years, almost 2 months ago (a Z87 i7 4770 Sabertooth system), I noticed some slight damage to the pins on my first motherboard CPU base as I put it together; and asked online about it, and noticed a few other people with similar problems; and it turns out that you can have some of the pins shorting out as they are "not all used" (as one techie put it); but that begs the question whether some of them are used only when you draw more power through the CPU when you pump more graphics through your new more expansive graphics card?

Or maybe its just getting hot under the collar and needs some new thermal paste.
Just a thought.

But I did think it was the PSU at reading the post at the head of the first page; but that (PSU issue) was discounted...

N.B. Even new boards have driver updates; I did have to update a bunch of drivers on my M/B and its chipsets (about 16 updates in the first month); so some issues may be to do with updates to your old board/s (but I guess you tried that already)?

Did you consider a strip down and de-dust and re-build from scratch like it was a new PC?

mucormucedo
10-11-2013, 03:22 PM
I did a clean up, new paste for the CPU, also saw this 1857 at my video card, but from this pictures it looks normal
http://eu2.sapphiretech.com/mc/images/prods/1682/11202-13_HD7750_2GBDDR3_HDMI_DVI_VGA_PCIE_C01_6348349021 09102116_600_600.jpg
I allready bought a new Power supply for nothing but I'm looking for new things.

Nismosis
10-23-2013, 02:53 AM
Hey guys, same problem here. Except my card stops responding until I turn off my pc.

I have found that it has something to do with the overclocking software. Even if I adjust the fan speed a little it will fail within a few minutes. If I don't run the software and keep it at default settings I have zero problems.
I'm thinking maybe the software conflicts somehow with my motherboard possibly, my motherboard is a few years old. the pcie port is version 2.0.
Any idea's?

Decebalus
10-24-2013, 07:21 PM
Hi everyone, i found something that works for me, no need to tweak the clocks or power, just disable windows aero or do as in the link below.
(i was getting vertical lines with the noise glitch BSOD(not actualy blue tho) )

Better explained here.

http://www.winmatrix.com/forums/index.php?/topic/36304-kb-2670838-windows-7-platform-update-causes-bsod-or-disables-aero/

mucormucedo
11-16-2013, 04:07 PM
I replaced my 7750 with an 4650 and i still got the error. I tested RAM, no errors. Don't know what to think.

jrweiss
11-16-2013, 08:40 PM
I don't see your hardware layout anywhere... New power supply "for nothing"?

https://www.sapphireforum.com/showthread.php?28641-Troubleshooting-Before-Posting-Read-This-First-!!-%28-Or-Threads-May-Get-Deleted-!!%29

Did you do a full driver and Registry cleanup, and install the correct driver for the 4650?

http://support.amd.com/en-us/download/desktop/legacy?product=legacy2&os=Windows%207%20-%2064

mucormucedo
11-16-2013, 09:17 PM
AMD Athlon(tm) X4 740 Quad Core Processor,AssRock FM2A55M-DGS, Radeon HD 4650 1 GB DDR2 128 b, A-Data 8 GB DDR3, W7-64b. I tried W8, I had 80% more crashes. I tested the memory and it's OK. I get crashes in Ubuntu, but I cant be sure of the cause. I did see a BIOS update today for my MB have to try that, and maybe Windows 8.1.

I did Uninstall the driver from Device Manager. Driver Sweeper seems to do more harm than good.

jrweiss
11-16-2013, 10:24 PM
If you're getting crashes in 2 different OSes with 2 different GPUs, it's likely a hardware problem (bad PCIe slot, Power Supply voltage) or BIOS incompatibility rather than drivers. The BIOS update is a good move.

Uninstalling AMD drivers isn't good enough. If you don't like Driver Sweeper, try CCleaner instead to clean the Registry after you purge the installation folder. You may want to re-install the OS from scratch.

What power supply? Can you see any change/instability in the 12V output under load (CPUid HWMonitor or similar)? Are CPU and GPU temperatures OK?

mucormucedo
11-17-2013, 07:21 AM
I hava Nox Urano SX 500 PSU with 21A/ 15A/ 36A, THe voltage is fine and so are the temperatures, I recorded with TechPowerUp GPU-Z and there is no connection between the crash and voltage or temperature. The BIOS update says " Improve system compatibility " and it is a month old it doesn't say what it does.

I used CCleaner for registry clean. I did not try a fresh install for a while.

jrweiss
11-17-2013, 08:27 PM
AFAIK, GPUz does not monitor the 12V input, only the 0.8-1.2V GPU voltage (nor does it monitor CPU parameters). There is more on the card (and in the computer) that relies on the 12V line than the GPU itself...

36A on the 12V rail is reasonable. However, I am a bit worried about their "85** efficiency" claim without 80** certification. Neither do they claim to meet ATX or EPS specs. It is possible they could not meet the voltage stability requirements...

Rubenoa
11-18-2013, 11:39 AM
Also, remember that the certificates 80 plus and others, are based on Max. peaks output, neither of them are on sustained/continuos values. Iīm personally against these certifications, as doesnīt tell us anything about the real values.

mucormucedo
11-18-2013, 12:31 PM
I tried a fresh install of Windows 8.1 x64, not even 1 min passed after I started Dota 2(I did it multiple times in just a few minutes), and driver stopped, I just installed latest BIOS and latest AMD beta driver 13.11.

Edit: I replaced it again, my 7750, that gives the error in just a few seconds, with the 4650 and no errors so far.

jrweiss
11-18-2013, 10:30 PM
Are you using the legacy driver for the 4650 or the same as for the 7750? If the 4650 runs on the new driver, I'd request an RMA for the 7750.

jrweiss
11-18-2013, 10:33 PM
Also, remember that the certificates 80 plus and others, are based on Max. peaks output, neither of them are on sustained/continuos values. Iīm personally against these certifications, as doesnīt tell us anything about the real values.
But the certifications also have voltage stability thresholds, so they should apply at any level.

With 36A 12V peak, there is plenty of headroom even if continuous load is only half that...

Rubenoa
11-19-2013, 11:51 AM
But the certifications also have voltage stability thresholds, so they should apply at any level.

With 36A 12V peak, there is plenty of headroom even if continuous load is only half that...
Negative, the certification only achieve max output loads...

jrweiss
11-19-2013, 07:11 PM
EPS12V Power Supply Design Guide V2.92, Voltage Regulation Spec:

6.5 Voltage Regulation
STATUS Required
The power supply output voltages must stay within the following voltage limits when operating at steady state and dynamic loading conditions...
**12V -- 11.4-12.6V [**/-5%]
. . .
The load transient repetition rate shall be tested between 50 Hz and 5 kHz at duty cycles ranging from 10%-90%.
ATX Spec 2.2:

4.1.4 Voltage Tolerances
Tolerance for the motherboard power rails should comply with the values listed in Table 6.
. . .
**12VDC ą 5 %
The ATX Power Supply Design Guide v2.01 further states (sec 3.2.1):

The DC output voltages shall remain within the regulation ranges shown in Table 2 [same range as stated above] when measured at the load end of the output connectors under all line, load, and environmental conditions.
Also, Sec 3.2.5.1 describes a "typical" load as ~50% of rated load, so any compliant PSU should be able to run at 50% rated load continuously.

Most reputable PSU mfgrs state they are compliant with the ATX and/or EPS12V specs. So if there is no actual ATX or EPS12V "certification", then just change my original statement to
"But the standards also have voltage stability thresholds, so they should apply at any level."

The 80** standard/certification requires a certain efficiency demonstrated at 20, 50, and 100% rated load (https://en.wikipedia.org/wiki/80_PLUS), so it is reasonable to assume the PSU can attain the rated load. I have not been able to find the original test protocol that describes the length of test for the certification; both 80plus.org and epri.org seem to have disappeared. However, Plug Load Solutions (http://www.plugloadsolutions.com/80PlusPowerSupplies.aspx), the redirect site for 80plus.org, cites a test protocol they claim to use for 80** testing, current as of April 2012 (http://www.plugloadsolutions.com/docs/collatrl/print/Generalized_Internal_Power_Supply_Efficiency_Test_ Protocol_R6.6.pdf). Steady-state operation is defined as no less than 15 minutes (Sec 4.3):

Prior to power measurements, the UUT shall be allowed to operate at each load point for at least 15 minutes in order to allow the power supply to reach a steady state of operation. A steady state of operation has been reached if the total input power reading over two consecutive five-minute intervals does not change by more than ą 1%
Therefore, an 80** certified PSU should be able to operate at 100% load for at least 15 minutes.

Since the OP's crashes occur during transients (e.g., "gives the error in just a few seconds"), and not after a prolonged time at or near max rated load, your peak vs continuous load argument is weak.

mucormucedo
11-20-2013, 08:51 AM
4650 runs on the 13.4 wich is the latest for that model.. The 7750 runs on the latest 13.11, but the error is the same since 12.8 driver, tried all drivers since then.

Rubenoa
11-20-2013, 11:32 AM
Since the OP's crashes occur during transients (e.g., "gives the error in just a few seconds"), and not after a prolonged time at or near max rated load, your peak vs continuous load argument is weak.
Negative again, as I have the test plenty of people did at the forum and other sources,to some of those 80 certified PSUs and the sustained values whre not near to the ones they specified as Max. Peaks, again, you are totally wrong and again you confuse Max. Peaks with sustained values when talking about intensity.
Those 15 minutes are just max. test time, and they take the maximum value they reach prolonged on an overall of a max.15 minutes (said in those links you posted).
Also, they only talk about Max. Peaks on intensity, no words about any sustained values or continuous.....
Conclusion, with no haste, you donīt know what you are talking about mate.

JPtjee
11-21-2013, 12:06 PM
Hi guys,

I have exactly the same problem. I have a 7790 card that produces the same errors in the log, but the difference is that my PC is crashing when this happens. So I only see "No input" on the screen. I tried to install the beta driver and to turn off the Overdrive feature of AMD, but still no luck. Before, when Overdrive was enabled, the screen crashes directly after starting a game at CoD 4 Demo or rFactor. When I disabled Overdrive, the problem was gone IN GAME, but when I'm watching a movie for example and click the mouse button after watching, the problem still occur. Also when I play some music, the problem occurred.

I don't know what to do anymore. I contacted AMD. They advised me to rollback to 13.4 and send me a general page where I can find the problem and more then 10 possible solutions. What I can do more is:
- Rollback to 13.4
- Updating Windows with the latest patches (not included in regular Windows Update updates)
- Updating BIOS
- Send the card back to the store and change it for a Nvidia 650 Ti

Do you have any tips?

My setup:
AMD FX-6300
ASRock 970 Pro3 R2.0
ASUS HD7790-DC2OC-1GD5
Be quiet! Pure Power L8 430W CM
Kingston ValueRam 8GB DDR3-1600 CL11
Seagate Barracuda 7200.14 1TB

Thank you!

jrweiss
11-21-2013, 09:35 PM
Negative again, as I have the test plenty of people did at the forum and other sources,to some of those 80 certified PSUs and the sustained values whre not near to the ones they specified as Max. Peaks, again, you are totally wrong and again you confuse Max. Peaks with sustained values when talking about intensity.
Those 15 minutes are just max. test time, and they take the maximum value they reach prolonged on an overall of a max.15 minutes (said in those links you posted).
Also, they only talk about Max. Peaks on intensity, no words about any sustained values or continuous.....
Conclusion, with no haste, you donīt know what you are talking about mate.
Yes, I am sure you are the only one on this forum who knows everything there is to know about continuous vs peak power. I'm also sure there is only one line of PSUs (the one you use) that can sustain stable voltage continuously above 50% of its rating. So, I'll defer to your superior knowledge.

jrweiss
11-21-2013, 09:39 PM
. . .
What I can do more is:
- Rollback to 13.4
- Updating Windows with the latest patches (not included in regular Windows Update updates)
- Updating BIOS
- Send the card back to the store and change it for a Nvidia 650 Ti

Do you have any tips?

My setup:
AMD FX-6300
ASRock 970 Pro3 R2.0
ASUS HD7790-DC2OC-1GD5
Be quiet! Pure Power L8 430W CM
Kingston ValueRam 8GB DDR3-1600 CL11
Seagate Barracuda 7200.14 1TB
I'd look at the PSU first. What is the 12V rating of that 430W PSU? Have you tried monitoring 12V voltage under load? CPUid HWMonitor can monitor voltages, and GPUz can put the GPU under load with its rendering test.

Rubenoa
11-22-2013, 02:16 AM
Yes, I am sure you are the only one on this forum who knows everything there is to know about continuous vs peak power. I'm also sure there is only one line of PSUs (the one you use) that can sustain stable voltage continuously above 50% of its rating. So, I'll defer to your superior knowledge.
Iīm not going to bother about answering you on this, up to you if you prefer kid manners, mister.

@JPtjee, what about that Asrock motherboard bios? You got the latest available? And for sure, review that PSU, we got in the forums plenty of cases with the same PSU, different GPU models and after changing that PSU, most of the problems were gone.

Regards"

JPtjee
11-26-2013, 05:39 PM
Hi guys! It looks like it works fine now. We fixed it to completely turn off AMD Overdrive and AMD Cool 'n Quiet in the BIOS. Thank you for your time!

Rubenoa
11-27-2013, 03:00 AM
Good to hear, thanks for posting back!

mucormucedo
01-14-2014, 07:00 PM
I have solved my problem(HD 7750), after 1 year, reading and trying all from new power supply to registry entries, and I returned the card to warranty and it was a problem, I got a new card now.
So, a lot of fixes are, well, just garbage, get a new one, don't wait like me. If you tried to fix it and still doesn't work, than it is a hardware problem. Don't waste your time and nerves.

Drtzekos
05-05-2014, 09:20 AM
I am facing the same problem.

A week ago I upgraded from 6790 to 7790 (2GB OC Sapphire). At start, i did not remove any driver, I just installed the new one over the old.. instantly I experienced very low fps in games that did not require a lot of performance (Like Dota-Warcraft3 or LoL).

Of course then I removed ALL drivers with their uninstallers and re-installed the new drivers.. Then although the performance improved in to normal levels, in random times during gaming (every 5-10') I would get a flicker for less than half a second which was more than annoying..

I removed again all drivers, this time using a driver-sweeper in safe mode and re-installed the new drivers . Since the "cleaned/sweeped" install I am getting the "Display driver stopped responding and recovered" message every half an hour two times one ofter the other.. (I did this 3 times already using 3 different sweepers) The odd thing is that it never happened during gaming but during browsing (with G Chrome). I didn't try to see if I am also still having this flickering during games..

Yesterday night after un/installing I also updated G Chrome (although it does not seem to be the problem) and went to sleep..

Later today I am planning to try

1) this --> http://support.microsoft.com/kb/2665946
2) Then if it does not work, i will try to increase a bit (1-2%) the voltage if possible (it was suggested in an other forum)
3) Increase a bit the minimum clocks (as the problems appears to occur only on "idle")
4) Format my pc and re-install everything from start
5) Rma my new GPU (i still have the old, I just changed because this one has **1 DVI port)

My configuration is:
OS: Windows 7 64Bit Professional
PS: 600W OCZ 80Plus
CPU: AMD FX 6100
MB: Asrock 970 Extreme 3
RAM: 2x 4 GB DDR3 Kingston @1600MHz
HDD: OCZ Vertex 3 120GB
Monitor: 3x Dell 2312HM (DVI/DVI/DP) ** one Toshiba TV (HDMI)

Hello again,

well after about 8-9 months, and after trying all different versions of AMD drivers I decided to RMA my card. I shipped it to computeruniverse.net (german eshop, where I bought it), and they filed the RMA procedure. In 5-6 weeks I had a "new" card, (not 100% if it was new, or repaired, but there was no cover for the card box, like a bulk version, and the bag was opened, all the parts or adaptors were new and included in the box). The new had a bit bigger hardware id, but the same bios version and it was not marked as revision xx, but it seems like the old one.

I already have the new 7790 for 2-3 weeks, till now I got no BSOD, no flicker, no non-responsive driver or stucking at browsing (it happened often durring loading of pages that required some graphics, like flash player etc).

I did no format or uninstall from drivers, between the two 7790 cards. I just replaced one with the other.

And now everything works perfectly, and as a result I would say it was an obvious hardware failure..