Does anyone still get the AMD 7xxx bug where your monitor doesn't wake up from sleep?

[Obsoleet]

Interesting to see that Nvidia users are getting this also in the thread.

It appears to be a combo hardware/software issue. I didn't have the problem on 12.10whql but every driver after it started.
Still looking for a fix, it seems to me on 14.4 whql the problem happens less often but not entirely resolved.

 

[themyst]

Running an R9 290 reference MSI instead of the 280x, the problem is different now, the monitor will wake, but it won't actually post an output to the screen.

Yes, already did the basic things such as disabling link state power management, fully wiping drivers using a utility and fresh installing them, etc.

Running 14.4 catalyst drivers.

 

[h9826790]

This may be an OS issue, not necessary the driver.

Try different setting may somehow solve the problem, e.g. disable screen saver (you will put the monitor to sleep anyway)

 

[atticus14]

Well I had this problem as long as I can remember and with each driver release I try again and fail but I would like to state that so far I've been running for about a month with 14.4 RC1 with no problems. For me this was majorly annoying because none of the wake up "tricks" that were shared here ever worked for me.

I have a MSI 7950 Twin Frozr (the one with 7970 PCB) 1100/1500 via Afterburner
(i7 920 @ 3.6 on the EVGA SLI X58, Win 7 64)


I did try the AMD uninstall utility for the first time before installing the 14.4 RC1


here's a quick link to the utility, I hope it helps someone

http://support.amd.com/en-us/kb-articles/Pages/AMD-Clean-Uninstall-Utility.aspx

 

[coffeejunkee]

Running an R9 290 reference MSI instead of the 280x, the problem is different now, the monitor will wake, but it won't actually post an output to the screen.

Yes, already did the basic things such as disabling link state power management, fully wiping drivers using a utility and fresh installing them, etc.

Running 14.4 catalyst drivers.

You shouldn't have to change any settings to fix this. I only had this problem with certain older drivers and newer drivers fixed it. No need for clean install either. Thus it's not a problem with any setting, it's a problem with the driver. That said, it still doesn't explain why some people are having the problem and others don't. Other potential culprits could be the monitor itself, any overclocks, any hardware monitoring software or even a problem with the card itself.

This may be an OS issue, not necessary the driver.

Try different setting may somehow solve the problem, e.g. disable screen saver (you will put the monitor to sleep anyway)

Don't think it's an OS issue either. I used Win 7 and 8, the bad drivers had the problem, the good ones didn't.

I also see people with older AMD cards and even Nvidia cards in this thread. We need to define this problem: it affects AMD cards with Zerocore feature (HD7000 and later) only and it only occurs when the OS puts the monitor to sleep, not the whole pc. Any other problems with either the monitor or the pc not waking up from sleep belong in a different thread.

 

[james1701]

Too bad we can't get someone official from AMD to chime in on this problem. It's all over the net, but no official word on what the cause may be.

 

[Peter Nixeus]

Monitor manufacturer here. I can provide some insight from our experience since 90% of our computers are using AMD graphics.

1) The #1 problem are display cables. If they are not VESA/HDMI compliant or have a problem with the "hand shake," the monitor will not wake from sleep. Try more "well know" cable brands. For example for DisplayPort cables, other than the ones we provide with our monitors - we recommend Accell cables. Try a different cable brand since it is the cheapest way to test first - to see if it resolves the issue.

2) On our first production runs, some of our monitors did have wake from sleep issues - but was quickly resolved with a firmware update performed on the monitor's PCB/Scaler. Some time during a production of the PCB/Monitor, something minor like not updating the firmware can cause handshake/wake problems like waking from sleep or incorrect native resolutions.

3) Wake from sleep issues are more common in monitors with resolutions exceeding 1920x1200 when using a lower quality cable.

We have AMD/ATI cards ranging from 5450s to the new R-series... a few Mac Mini.. We have no wake from sleep issues for all our office monitors ranging from 1920x1080 to 2560x1600.

 

[mindbomb]

I think this problem may be related to monitor timings. I'm gonna play around with some stuff with custom resolution utility and see if that helps. Specifically, I'm gonna increase the horizontal blanking. If anyone else wants to try this, keep in mind the 165mhz pixel clock limit of single link dvi. I'm using custom resolution utility as the tool to do this.

edit:
Well, it didn't work for me.

 

[SSBrain]

I have this problem too.
My main display is a Dell U1412M, my GPU a Sapphire Radeon HD7770, using Windows 8.1 Update and latest Beta Catalyst drivers (14.6). Doesn't happen with the integrated Intel GPU.

I tried pretty much everything I could as I've written here:
http://forums.amd.com/game/messageview.cfm?catid=454&threadid=174323&enterthread=y

 

[Kodiack]

This issue is incredibly frustrating and annoying. There are several times where the AMD GPU just doesn't "wake up". I'm fortunate to have an integrated HD 4600 GPU that I also use in my system. I know that this is an AMD issue because the integrated chip wakes up just fine, while the monitor that the R9 290 pushes stays black. I can log back in using the integrated GPU, but then I have to go to the screen properties and force the system to re-detect the monitor connected to the AMD card.

Occasionally, AMD's display driver seems to just "die" and bring the entire system down with it, although that's definitely the exception. Regardless, AMD desperately needs to resolve this. To make sure they know about it, use the AMD Issue Reporting Form. I've had good luck with their feedback forms before and at one point had a QA person from AMD contact me a couple weeks before they rolled out a fix to an obscure X58 chipset/HD 7000 series stability issue, so it actually does go somewhere.

Also of note is that they do have a a dropdown option for "System Does Not Resume Correctly from Sleep/Idle Mode", so maybe it's known but not well-documented. *shrug*

 

[BrightCandle]

This issue is incredibly frustrating and annoying. There are several times where the AMD GPU just doesn't "wake up". I'm fortunate to have an integrated HD 4600 GPU that I also use in my system. I know that this is an AMD issue because the integrated chip wakes up just fine, while the monitor that the R9 290 pushes stays black. I can log back in using the integrated GPU, but then I have to go to the screen properties and force the system to re-detect the monitor connected to the AMD card.

Occasionally, AMD's display driver seems to just "die" and bring the entire system down with it, although that's definitely the exception. Regardless, AMD desperately needs to resolve this. To make sure they know about it, use the AMD Issue Reporting Form. I've had good luck with their feedback forms before and at one point had a QA person from AMD contact me a couple weeks before they rolled out a fix to an obscure X58 chipset/HD 7000 series stability issue, so it actually does go somewhere.

Also of note is that they do have a a dropdown option for "System Does Not Resume Correctly from Sleep/Idle Mode", so maybe it's known but not well-documented. *shrug*

I just want to say while I applaud your desire to report the bug properly I reported this failure to wake bug to AMD in February 2012. I repeatedly raised this and other bugs over a period of 6 months before I eventually gave up in frustration and changed to Nvidia cards. This is just how AMD support is, they will not do a thing with your bug, all of mine have been sat there for years without fixes.

You need to decide whether you can live with it or not, because AMD is most likely not going to fix it unless the media breaks a story on it.

 

[chimaxi83]

I just want to say while I applaud your desire to report the bug properly I reported this failure to wake bug to AMD in February 2012. I repeatedly raised this and other bugs over a period of 6 months before I eventually gave up in frustration and changed to Nvidia cards. This is just how AMD support is, they will not do a thing with your bug, all of mine have been sat there for years without fixes.

You need to decide whether you can live with it or not, because AMD is most likely not going to fix it unless the media breaks a story on it.

I agree with this. AMD doesn't seem to care about these kinds of "small" issues.

 

[tweakboy]

Very normal what your experiencing. gl

 

[VirtualLarry]

I've been dealing with this issue, since I set up my Lenovo H520 with IB i3-3240, 8GB DDR3-1600, and a 1TB HDD. Installed a NAXN 350W, or maybe 400W PSU, and an HIS 7790 1GB GDDR5 card. OS is Windows 8. Monitor is 1920x1200 24" IPS Auria.

Using a 10ft HDMI cable purchased from 1sale.com.

Not having any sort of display or cable issues when monitor isn't in sleep mode. Nor am I having any issues upon cold power-up.

I will note that I run BOINC in the background, which "crunches" on both the CPU and the GPU. What that does when the PC goes into monitor-sleep mode I don't know.

I noticed that when it won't wake up, the HDD light is flashing, so it's not hung, the PC is still processing in the background. However, I cannot shut it down normally by pressing the PC's power button. It seems to do nothing. Nor does the button on my keyboard that shows a power symbol.

I have to hold down the power button and force a hard shutdown. Then when I reboot after doing that, the wireless card won't activate, so I have to do a restart to get the wireless back.

Edit: Oh yes, forgot to mention, NVidia for me from now on. Done with AMD's shoddy engineeering.

Btw, I previously had two of these cards, and I was crunching with them on an ASRock 990FX Extreme 4 board with a Thuban 1045T CPU, 16GB DDR3-1600, and Windows 7. Variously on BOINC, and F@H. I had the same problem on that system. I figured it was just the system, but it has to be either the cards, or the drivers.

I will also note, that on the Lenovo i3, it previously was waking up ok last night, but one time, the screen was just black with a mouse cursor showing (didn't put monitor to sleep properly), and other times, when I woke it, all of my desktop windows had been resized to something very tiny, and were all in the upper-left corner. So the desktop workspace had been massively shrunk down for some reason. (My HDMI-connected monitor is my only display.)

Edit: This is with a fresh clean install of 14.4 WHQL drivers.

Edit: Other weirdness. Sometimes, instead of turning the monitor off, I would get a black screen with a white mouse cursor arrow that I could move around, but moving it didn't cause the screen to come back. Turning off the monitor at that point, and then turning it back on, caused the screen to go into standby again.

Anyways, I recalled having a similar issue back in Dec. of 2013, during the DC forum's F@H race, when I was using the beta drivers. I remember rolling back to the 13.12 WHQL drivers and it fixed the problem. So I tried that again.

So far, in a day and a half almost, no problems waking from monitor sleep mode, even with BOINC crunching on the GPU in the background.

Also, the desktop windows don't get resized down to tiny windows when it restores, the whole desktop stays put.

 

[birthdaymonkey]

Yup, it's ridiculous. I have been a bit of an ATI/AMD fan boy for a long time, but I can't use their products as long as they have this serious problem.

 

[james1701]

The reports of it being the USB pass through in the monitor did not fix my problem.

http://en.community.dell.com/support-forums/peripherals/f/3529/t/19567748.aspx

 

[ZGR]

I'm getting this as well with the R9 290. After the computer sleeps for awhile, the monitor and computer wake up, but the monitor displays a black screen. I have to do a force shutdown to reboot.
This doesn't happen every time, but it is kind of annoying. The motherboard still reports its booted into windows.

Heck, this even happens on my Macbook Pro when it goes into hibernation on OS X. Not a problem on Windows, but I will get a similar black screen bug which has been around OS X for years. Entirely different cause, but similar issue.

I wonder if I can still use the keyboard during these periods. Anyone know how to force an AMD card into different power states with a .bat file? I can do this with an NVIDIA card using NVIDIAInspector, but I am not familiar with a similar utility for AMD.
If I could force the card into 3d mode/high performance mod and then back to 2d/idle, I wonder if that could wake it up. Or a script to end the AMD display driver and restart it.

 

[BrightCandle]

This sounds like the bug I had with the 4870X2, 5970 and the 2x7970's. Its one of a range of power saving bugs I raised with AMD many years ago. It only for me happened on a particular monitor however, I didn't have the problem on the Dell 2410 but on an earlier IPS screen which worked perfectly on the Nvidia card before it but had problems waking from sleep with AMD. Never got a response from AMD, apparently its still there in their latest cards.

You can try raising a bug with AMD support although be aware that I have raised over 40 bugs with them over the last few years and they didn't fix any of them. I would say learn to live with it, change the monitor or change the GPU.

 

[mkmitch]

I spent a month trying to get 2 different 290x's to work. I finally figured out how to solve this problem, my evga 780 is due to arrive tomorrow. Had nothing but Radeon cards since forever. Never again.

 

[spaceman]

i hate to say it im having same prob w. 7870 tahiti le

 

[f1sherman]

I'm getting this as well with the R9 290. After the computer sleeps for awhile, the monitor and computer wake up, but the monitor displays a black screen. I have to do a force shutdown to reboot.
This doesn't happen every time, but it is kind of annoying. The motherboard still reports its booted into windows.

Heck, this even happens on my Macbook Pro when it goes into hibernation on OS X. Not a problem on Windows, but I will get a similar black screen bug which has been around OS X for years. Entirely different cause, but similar issue.

I wonder if I can still use the keyboard during these periods. Anyone know how to force an AMD card into different power states with a .bat file? I can do this with an NVIDIA card using NVIDIAInspector, but I am not familiar with a similar utility for AMD.
If I could force the card into 3d mode/high performance mod and then back to 2d/idle, I wonder if that could wake it up. Or a script to end the AMD display driver and restart it.

Need to run this in a devcon.exe batch file, but with admin rights (else you get no devices found).
I don't know how to do this on Windows 8.1
Proly needs another 3rd party tool

C:\Programs\DEVCON\devcon.exe restart *VEN_1002
pause

Oh and the shorctut file needs to be in C:\ProgramData\Microsoft\Windows\Start Menu

 

[MeldarthX]

Most don't want to hear it; I've had this issue on a couple different systems due to cheap cables.; I've seen people try everything but change the cable; it is a frustrating situaton. If you have changed your cables I feel for you even more.

Its one of those situations that happens on both sides; though people don't seem to want to admit it. Also a cable that worked for a couple years can degrade and no long work as well or at all. I've have that issue also.....We've had a monitor manu come in and say a lot of it comes down to cheap cables they have seen.......

 

[Maximilian]

Occasionally yes. And the bug where the display driver stops working but recovers again on boot.

I thought I was buying into a mature solid thing with a 280x, the design is years old after all!

 

[SSBrain]

Most don't want to hear it; I've had this issue on a couple different systems due to cheap cables.; I've seen people try everything but change the cable; it is a frustrating situaton. If you have changed your cables I feel for you even more.

Its one of those situations that happens on both sides; though people don't seem to want to admit it. Also a cable that worked for a couple years can degrade and no long work as well or at all. I've have that issue also.....We've had a monitor manu come in and say a lot of it comes down to cheap cables they have seen.......

I tried using several different cables, including the one which came with the monitor, which was new. Displayport or DVI, it won't matter: the problem still occurs.

 

[f1sherman]

Occasionally yes. And the bug where the display driver stops working but recovers again on boot.

I thought I was buying into a mature solid thing with a 280x, the design is years old after all!

Bold part is what's annoying here.
For a while, I had the working theory that it's due to my undervolting... nope.

C:\Programs\DEVCON\devcon.exe restart *VEN_1002

should restart GPU, but the batch file needs admin rights.

ANYONE knows how to do this?