geekhack
geekhack Community => Keyboards => Topic started by: clynx on Sun, 30 June 2013, 19:46:05
-
When the computer wakes from sleep all LED's are on and nothing works until I unplug and plug the USB back then all is good. It does not happen all the time at least once a day though. I tried different USB ports, powered and non-powered USB hubs with the same problem. No problems when powering on and off the computer only when waking from sleep. Has anyone else had this problem? Thanks in advance for any help or ideas.
-
When the computer wakes from sleep all LED's are on and nothing works until I unplug and plug the USB back then all is good. It does not happen all the time at least once a day though. I tried different USB ports, powered and non-powered USB hubs with the same problem. No problems when powering on and off the computer only when waking from sleep. Has anyone else had this problem? Thanks in advance for any help or ideas.
are you using AMD gfx with afterburner?
-
I am using Gigabyte ATI 7870. No Afterburner. It is OC'ed though . Thanks.
-
Do you have an AM3+ motherboard? They are known to sometimes have this issue.
I'd tell you what causes it, but I'd be lying out I said I knew, sorry.
-
Yes it is an AM3+ board. I just found that the USB Filter driver is known to mess with some USB devices. I have now removed it and time will tell. At least now I will stop blaming the problem on the keyboard. Thank you very much for your ideas that allowed me to start looking in a better direction.
-
overclocking these new amds have problem with Zero core.
it interferes with it.
turn of ULPS. see if that helps.
-
Never heard of that... My computer with a 7970 has ULPS enabled, and I didn't have this problem. I can see my card's fans turn off when it idles for a while (reverse-ATX case with a modded window), and everything wakes up fine. It sounds more like a motherboard BIOS issue (up to date?) or perhaps try a different USB port? USB 2.0 instead of 3.0? Just a different port in general.
-
Bios is up to date. I have removed USB Filter Driver first and will wait and see. The problem is intermittent but it was happening at least once a day when in sleep mode for a few hours. If it happens again I will disable overclock function in Catalyst next. All of your help is greatly appreciated as with the help to find this great keyboard! Thank you.
-
... video... I ... just never &*$#ing touch a computer again. You aren't frigging qualified to install DIMMs, forget troubleshoot this issue. My gods. The level of stupidity in computers these days, just shoot them all. P.S. video driver has absolutely no effect on USB whatsoever, never has, only the sound because of the HDCP stack. You want to argue it, go hang with the kiddies on Reddit who order an i7 3820 then can't figure out why it doesn't work on their Z77 board.
TL, DR: DON'T MESS WITH THE VIDEO DRIVER. PEOPLE CLAIMING IT'S VIDEO DRIVER SETTINGS ARE BOTH WRONG AND UNQUALIFIED.
Look, the problem flat out is the chipset. Period. (And no. Unless you worked at AMD, AMI, Award or Phoenix you are not qualified to argue with me on this nor are you qualified to say it isn't.) And since multiple boards have not resolved with various BIOS modifications, it's either present in both bases (Award and AMI/Phoenix) or the chipset. But since the problem is present in 970A and 990FX, it's the SB950 chipset. Actual root cause is that USB state change is glitching out and not completing transition from D3 to D0. Given the specific symptoms I'd say it's hanging in D2 (armed for remote wakeup.)
And wouldn't you know it, errata for the SB950 basically says exactly that (http://support.amd.com/us/ChipsetMotherboard_TechDocs/49645.pdf).
1 - G3->S5->S0 also effects S5->S0+D0 transition as 48MHz PLL not available immediately on wakeup.
3 - Not this one because there's no BSoD. No BSoD, it's not this errata. The end.
5 - This is the other suspect one, despite lack of BSoD. EHCI state machine sync loss could cause a failed D3->D0. And no fix.
15 - This one is also a possibility. In a sleep state change, the primary expected effect would be the described symptoms.
18 - Nope; BIOS fix available in CIMx and state transition is pre-DMA.
19 - Yes this is the most likely one. Babble fits perfectly.
25 - Not really a possibility. It's very rare to find devices that do that.
-
TL, DR: DON'T MESS WITH THE VIDEO DRIVER. PEOPLE CLAIMING IT'S VIDEO DRIVER SETTINGS ARE BOTH WRONG AND UNQUALIFIED.
I only see one ;)
-
... video... I ... just never &*$#ing touch a computer again. You aren't frigging qualified to install DIMMs, forget troubleshoot this issue. My gods. The level of stupidity in computers these days, just shoot them all. P.S. video driver has absolutely no effect on USB whatsoever, never has, only the sound because of the HDCP stack. You want to argue it, go hang with the kiddies on Reddit who order an i7 3820 then can't figure out why it doesn't work on their Z77 board.
TL, DR: DON'T MESS WITH THE VIDEO DRIVER. PEOPLE CLAIMING IT'S VIDEO DRIVER SETTINGS ARE BOTH WRONG AND UNQUALIFIED.
Look, the problem flat out is the chipset. Period. (And no. Unless you worked at AMD, AMI, Award or Phoenix you are not qualified to argue with me on this nor are you qualified to say it isn't.) And since multiple boards have not resolved with various BIOS modifications, it's either present in both bases (Award and AMI/Phoenix) or the chipset. But since the problem is present in 970A and 990FX, it's the SB950 chipset. Actual root cause is that USB state change is glitching out and not completing transition from D3 to D0. Given the specific symptoms I'd say it's hanging in D2 (armed for remote wakeup.)
And wouldn't you know it, errata for the SB950 basically says exactly that (http://support.amd.com/us/ChipsetMotherboard_TechDocs/49645.pdf).
1 - G3->S5->S0 also effects S5->S0+D0 transition as 48MHz PLL not available immediately on wakeup.
3 - Not this one because there's no BSoD. No BSoD, it's not this errata. The end.
5 - This is the other suspect one, despite lack of BSoD. EHCI state machine sync loss could cause a failed D3->D0. And no fix.
15 - This one is also a possibility. In a sleep state change, the primary expected effect would be the described symptoms.
18 - Nope; BIOS fix available in CIMx and state transition is pre-DMA.
19 - Yes this is the most likely one. Babble fits perfectly.
25 - Not really a possibility. It's very rare to find devices that do that.
Everyone just needs to calm down..... lol
(http://t.qkme.me/3oqduj.jpg)
-
There are a couple settings I would try:
Try a different USB port, preferably one that is native and not an extra chip on the board.
Take a look at the your power settings. If you are on power saving or balanced, it may be putting USB devices to sleep and they aren't waking. I had is on my intel system and it wasn't just the keyboard but the mouse as well.
-
Never heard of that... My computer with a 7970 has ULPS enabled, and I didn't have this problem. I can see my card's fans turn off when it idles for a while (reverse-ATX case with a modded window), and everything wakes up fine. It sounds more like a motherboard BIOS issue (up to date?) or perhaps try a different USB port? USB 2.0 instead of 3.0? Just a different port in general.
usually the ULPS has problems if you used an OC software that's NOT the Overdrive from AMD.
The Overdrive from AMD supports all the power states.
But Afterburner does not, and will crash the **** if the amd zero core is engaged.
-
... video... I ... just never &*$#ing touch a computer again. You aren't frigging qualified to install DIMMs, forget troubleshoot this issue. My gods. The level of stupidity in computers these days, just shoot them all. P.S. video driver has absolutely no effect on USB whatsoever, never has, only the sound because of the HDCP stack. You want to argue it, go hang with the kiddies on Reddit who order an i7 3820 then can't figure out why it doesn't work on their Z77 board.
TL, DR: DON'T MESS WITH THE VIDEO DRIVER. PEOPLE CLAIMING IT'S VIDEO DRIVER SETTINGS ARE BOTH WRONG AND UNQUALIFIED.
Look, the problem flat out is the chipset. Period. (And no. Unless you worked at AMD, AMI, Award or Phoenix you are not qualified to argue with me on this nor are you qualified to say it isn't.) And since multiple boards have not resolved with various BIOS modifications, it's either present in both bases (Award and AMI/Phoenix) or the chipset. But since the problem is present in 970A and 990FX, it's the SB950 chipset. Actual root cause is that USB state change is glitching out and not completing transition from D3 to D0. Given the specific symptoms I'd say it's hanging in D2 (armed for remote wakeup.)
And wouldn't you know it, errata for the SB950 basically says exactly that (http://support.amd.com/us/ChipsetMotherboard_TechDocs/49645.pdf).
1 - G3->S5->S0 also effects S5->S0+D0 transition as 48MHz PLL not available immediately on wakeup.
3 - Not this one because there's no BSoD. No BSoD, it's not this errata. The end.
5 - This is the other suspect one, despite lack of BSoD. EHCI state machine sync loss could cause a failed D3->D0. And no fix.
15 - This one is also a possibility. In a sleep state change, the primary expected effect would be the described symptoms.
18 - Nope; BIOS fix available in CIMx and state transition is pre-DMA.
19 - Yes this is the most likely one. Babble fits perfectly.
25 - Not really a possibility. It's very rare to find devices that do that.
This might be the amd ulps problem that nearly everyother person with AMD cards run into..
-
And wouldn't you know it, errata for the SB950 basically says exactly that (http://support.amd.com/us/ChipsetMotherboard_TechDocs/49645.pdf).
1 - G3->S5->S0 also effects S5->S0+D0 transition as 48MHz PLL not available immediately on wakeup.
That one sounds like a strong contender for the culprit. Interesting that they make distinction between USB 2.0 and USB 1.1 behaviour...
OP - I'm guessing that the Pure is USB 2.0, so do you have a USB 1.1 keyboard you could test with?
Try a different USB port, preferably one that is native and not an extra chip on the board.
Actually, if it's a chipset problem, an extra chip's port might work... try everything!
-
Well it has been several days now after I removed the "USB Filter Driver". I have had no more LED Keyboard freeze issues. It also could be because of a different USB port too, either way success! I hope this discussion may help others. I really want to thank everyone who replied to my question. I never would have figured this out. Thank you very much.
-
Well it has been several days now after I removed the "USB Filter Driver". I have had no more LED Keyboard freeze issues. It also could be because of a different USB port too, either way success! I hope this discussion may help others. I really want to thank everyone who replied to my question. I never would have figured this out. Thank you very much.
Yeah, that'd flag it up as Errata 19 as usual triggered by low quality engineering and design. (That's just about EVERY SB950 board though.) Or could be pin-level defect in the SB950. My guess is anything else on the same port will have similar issues - especially since keyboards are very coarse by comparison.