Rampage V Extreme Persistent Wake from Sleep Problems

[nonverbal]

Since I first owned this board (2014) I have had consistent wake from sleep problems with all OSs including Windows 8/10 and various LInux distros and different hardware configurations (e.g., Titan X SLI, 32GB of RAM, different SSDs). Specifically, 1 out 5 times after the computer has gone to sleep it will not wake-up and resume without requiring a hard reboot (i.e., holding down the power button). This happen regardless if I am running stock settings or overclocked and despite tweaking myriad voltages such as System Agent and memory.

Every time the system refuses to return from sleep the Qcode E1 is displayed and the PCIEX16_4 LED is blinking. Generally the system appears to return partially (there is system activity) but the display and USB connections are inactive and I have to hard restart and am greeted by the failed overclock BIOS screen, even if the settings are default and not overclocked. Only one PCIE slot 1 is populated with a Titan X.

A few BIOS versions ago I had persistent problems with some USB ports not working, but that seems to have resolved. I'm wondering if anyone has any advice or if this may just be an RMA issue either. Besides this issue with my system, it is stable and boots and behaves under load completely fine (surviving stress tests).

Thank you for any advice you can provide.

[Arne_Saknussemm]

Hi autoturk

Sorry I had Haswell in my post but we're talking Haswell E X99...different platform...
1.3 VCCSA on that platform is at the high end of safe I think...though too much can be as bad as not enough sometimes

Maybe start a new thread?

[autoturk]

Arne Saknussemm wrote:
Hi autoturk

Sorry I had Haswell in my post but we're talking Haswell E X99...different platform...
1.3 VCCSA on that platform is at the high end of safe I think...though too much can be as bad as not enough sometimes

Maybe start a new thread?

Thank you! I'll go ahead and start a new thread.

[GoatHumper]

Do you have similar "safe ranges" for me? 😄

I've set 1.35v for the DRAM, and that should be OK, but haven't touched the VCCSA yet - I'm thinking 1.1v would be my max (currently at ~0.8v)...

Thoughts?

[Arne_Saknussemm]

Keep under 1.2 VCCSA 1.1 is fine...just bump to 1.0v and then 1.05.....1.1...

[GoatHumper]

Arne Saknussemm wrote:
Keep under 1.2 VCCSA 1.1 is fine...just bump to 1.0v and then 1.05.....1.1...

I just did. It's been less stable these past couple of days that I've messed with the voltages than before where I had just set Rampage Tweak to Mode 1.

We'll see how this pans out. If not, I'll simply stick to Mode 1 and live with the disappointment. Why allow a Motherboard/CPU combo to support a memory configuration that's not stable without at least stating as much? 😞

Cheers...

[GoatHumper]

After a few relapses, I've noticed that the LED display shows "E1" which according to the manual corresponds to "S3 Boot Script execution" ... I wonder if there's a means for debugging what instruction it's hanging on...

Also, there's a blinking yellow light right by the LED, from the PCI Express I think... I'm afraid I didn't check to see which slot it was. I have a 1080 TI on PCIE_X16_1, NVME drives on both PCIE_X16/X8_3 and PCIE_X8_4, with a PCIE x1 sound card (old but gold SB X-Fi Titanium Fatal1ty) on PCIE_X4_1. I'll be more mindful the next time it happens to identify which device might be causing issues. At this point, my money is on the sound card since it's the older device.

Cheers!

[badmojout]

GoatHumper wrote:
After a few relapses, I've noticed that the LED display shows "E1" which according to the manual corresponds to "S3 Boot Script execution" ... I wonder if there's a means for debugging what instruction it's hanging on...

Also, there's a blinking yellow light right by the LED, from the PCI Express I think... I'm afraid I didn't check to see which slot it was. I have a 1080 TI on PCIE_X16_1, NVME drives on both PCIE_X16/X8_3 and PCIE_X8_4, with a PCIE x1 sound card (old but gold SB X-Fi Titanium Fatal1ty) on PCIE_X4_1. I'll be more mindful the next time it happens to identify which device might be causing issues. At this point, my money is on the sound card since it's the older device.

Cheers!

When you see the E1 code, how did you wake the computer? *By moving the mouse or kitting a key? *If so, try this the next time. *Wake up your computer and when everything starts up but you get a black screen, just wait. *Make yourself a sandwich, grab a drink but just wait it out. *I suspect it’ll eventually resume and load into Windows. *Give it a good 15-30 minutes before you think of doing a hard reset or power off. *If my hunch is correct, it should boot into Windows well before the 15 minute mark. *

[GoatHumper]

badmojout wrote:
When you see the E1 code, how did you wake the computer? *By moving the mouse or kitting a key? *If so, try this the next time. *Wake up your computer and when everything starts up but you get a black screen, just wait. *Make yourself a sandwich, grab a drink but just wait it out. *I suspect it’ll eventually resume and load into Windows. *Give it a good 15-30 minutes before you think of doing a hard reset or power off. *If my hunch is correct, it should boot into Windows well before the 15 minute mark. *

Well... first off: it happens regardless of whether I wake it up via keyboard, mouse, or power button.

Second, the reason I use suspend is precisely so I don't have to wait "too long" to get the computer back up from "shutdown" (i.e. cut down on power consumption, while also cutting down on booting delays).

Waiting longer than 1 minute for the box to come back up from suspend is a waste of time - I'm better off just rebooting at that point.

So thanks for the suggestion, but no thanks 🙂

I'm hoping someone will chime in and suggest a BIOS voltage tweak that might do the trick (Arne? You still there? :D).

Cheers!

[Arne_Saknussemm]

GoatHumper wrote:
Arne? You still there?

Sort of :o...sorry..seem to have missed some posts...

Waking from sleep problems are often RAM related so my first thought is more of the same...either relax timings or bump voltage(s) a bit

If the RAM is unstable and you are getting BSODs or having to hard reset you might also run into OS corruption which feeds back to greater instability

Maybe run "sfc /scannow" in cmd as admin to check OS is OK

[GoatHumper]

Arne Saknussemm wrote:
Sort of :o...sorry..seem to have missed some posts...

Waking from sleep problems are often RAM related so my first thought is more of the same...either relax timings or bump voltage(s) a bit

If the RAM is unstable and you are getting BSODs or having to hard reset you might also run into OS corruption which feeds back to greater instability

Maybe run "sfc /scannow" in cmd as admin to check OS is OK

A few points: the timings are already relaxed and the voltages are already high (1.1v for VCCSA, 1.37v for DRAM, etc... the DRAM says it's meant to work with 1.35 so I'm reluctant to go much higher than that). Also, I'm not running windows so O/S corruption isn't in play (or, at least, shouldn't be). The box has so much RAM that swapfiles are off on all O/S instances, so that's another variable off the table.

It catches my eye that it's the same device every time. Perhaps there's a way to increase the voltage for the SouthBridge (?) or PCI-e bus to give those devices a boost when resuming from S3? Which settings should I look at for that?

Thanks!