X370 Taichi sleep issues

Was this Windows install a fresh install?

danielocdh wrote: I have had this mobo for 1 or 2 months now, I didn't have the need to put it to sleep before so I never tried it. Some days ago I wanted to configure sleep and I discovered 2 big issues. 1 when disabling SMT on bios and then trying to sleep, the computer sleeps but then isn't able to wake up, I believe the error codes on the mobo are sometimes 1F and sometimes 0D 2 with SMT enabled it can sleep and wake up fine, the problem is that if I shut it down after it woke up the system will take extra time to shutdown. Without sleeping windows closes and goes to black screen then the system shuts down after 1-3 seconds, after sleeping this takes 2 minutes 20 seconds I have tried without improvement: Clearing the CMOS Bios 2.40 and 3.00 Always using default settings (except for the SMT in case 1) Different mouse, keyboard and monitor Balanced, high performance and AMD ryzen balanced power plans ASRock page chipset drivers and latest AMD site chipset drivers Clean boot 2 old Nvidia video drivers and the most recent one (used ddu before all installs) My specs: Ryzen 1800X, liquid cooling Corsair h110i Asus GTX 1080ti Ram crucial 4x16gb ECC @2400 dual channel Power supply EVGA Super Nova 850 P2 NVME Samsung evo 960, 2xSSD, 2xHDD, 1DVDRW Windows 10 x64 fully updated Can anyone reproduce these issues? Any ideas on how to figure out what the problem is? Thanks

I tried disabling SMT with my X370 Killer SLI/ac board with 1700X, went into Sleep mode, came out of Sleep and did not have normal operation of the PC, nothing displayed on the monitor.

Your description of "isn't able to wake up..." is very lacking in details, so not at all helpful. I'll describe what happened in my case. The PC fans, etc starts, but no signal to the monitor. Pressing Reset causes the PC to restart normally.

Checking the Windows logs at the time I woke the PC (since log entries are time stamped), I see an entry stating the OS was started. So this seems to be related to the video driver having an issue with Windows when waking from Sleep, an issue I have seen before with an Intel P67 chipset board. Event log entries saying the video driver did not respond within the expected time period was the main clue. Myself and other owners of the board (Asus) never found a fix. But that issue was not related to disabling the Intel equivalent of SMT, Hyper Threading. The issue there was supposedly related to the P67 chipset itself, and if that is true in this situation, unless a CPU microcode update from AMD can fix it, there most likely won't be a fix. But there are so many possible causes of Sleep issues, I'm just glad when it works on a PC.

The good news here is, I have no reason to disable SMT, and won't. The PC Sleeps and wakes fine with SMT enabled. Wake from Sleep issues are very common, and since Sleep is a Windows feature, it is mainly a Windows issue. The video driver manufacture may have responsibility, if there driver is not meeting Windows 10 standards. Still seems weird disabling SMT has this side affect, but most wake from Sleep issues are unusual.

I see you tried different video drivers, so you may be aware that the board is starting normally, and the issue is primarily related to the video source. But we don't know what happens when you wake your PC from Sleep.

The second situation of Windows taking a long time to shutdown after waking from Sleep could be any number of Windows related things, which I don't get on my Ryzen PC.

danielocdh wrote: Strangely I lost some unimportant files while testing this

That there portends maybe an issue with one of your drives.

Which could, from your explanations, be THE issue you're having there.

danielocdh wrote: Thanks a lot parsec, this kinda is good news. What I meant by not being able to wake up is what you described, fans start but there is no video signal and the mobo shows the error codes I mentioned. The first 2 times I tested this was after a couple of hours of have been using the system, and I remember that there was no log on the event viewer that matched the time of the sleep/fail to wake up, but maybe I missed it, I'll look for it when I test this more. I know it's uncommon to want to disable SMT but I have seen some improvement on some benchmarks and games, I still use some old programs and games that only use single core, I could live with it on anyway I guess. I expect to test this more tomorrow to try to solve the long shutdown delay. Strangely I lost some unimportant files while testing this so I'll need to disconnect all my drives and use other spare drive to install windows/test.

POST codes shown on the Dr Debug display are not errors, unless the board fails to POST. Then the last POST code that remains on the display is the number of the POST procedure that failed, and a clue to what went wrong. Waking from Sleep is not a typical starting a PC situation, so what the POST codes mean at that time is not quite the same. My X370 board does not have a Dr Debug display   so I cannot check what I get in that situation.

I've only seen the video failure after waking from Sleep happen with video cards, and it's too bad non-APU Ryzen processors don't have an iGPU, that could be enabled with a second monitor. The P67 board I had did not support the Intel iGPU, which is strange but that's how that board was. I cannot say that an iGPU is immune from this problem, but I've just never seen it. I never saw a message that the OS was started on the P67 PC, which was using Windows 7.

In your case, do either of those codes remain on the display, I assume one of them does? 0d can be a problem with the video card, which we think is related to the lack of video when waking from Sleep. 1F is a memory problem, but if that code just appeared and then disappeared, then it was probably shown since Sleep uses the memory to store the Windows "image" when the PC was put to Sleep.

I tried disabling SMT, going into and out of Sleep, and I still did not get any video, and I use two monitors with one video card, so it seems the video output does not matter. I again did not get any delay during Shutdown, so that seems to be unique to your PC for some reason. Such as a potentially bad drive.

Edited by danielocdh - 28 Aug 2017 at 3:31pm