Asus Hero VIII Alpha Wake Problems Q-Code 5d

Barney Craggs

Reputable
Feb 10, 2016
3
0
4,510
Hey all. Me again.

There have been a few discussions around boot and wake problems but as yet I can't find anything the same as mine so reaching out.

The build has been broadly working fine for a week or so. Benchmarks, stress tests and such all going fine. BUT the occasional crash or a long time on sleep led to a power up error along the lines of "Power surge detected, hit F1 to head off to the BIOS and fix things..." I've now managed to replicate this consistently. Just set the machine to sleep and on wake the screens don't sync up, the keyboard lights don't flash. Also the yellow/orange DRAM_LED stays lit and the Q-Code rocks a 5d - which according to the manual is "reserved for future.."

A full power cycle, a quick trip into BIOS without changing anything and it all boots fine. Also turning the machine fully off doesn't seem to cause this problem.

I thought perhaps there was a problem with the DRAM itself - I do have 2x8Gb kits installed and whilst they are exactly the same model, timings and such there is a "revision" difference on the back of each pair. BUT these have been fine and very stressed tested for days. Tried swapping slots, singles, pairs, orders... all with same effect, the machine posts a 5d after sleep. On that I can't see the problem actually being the DRAM.

BIOS is 1402 (cleared CMOS just to be sure), no overclock or XMP, Windows 10 64bit and all the latest drivers.

I'm stumped. Anyone suggest something?

---

MORE information 22Feb

Tried at 1402 & 1302 bios all with cleared CMOS (i.e. stock).

- 1 module in A2, then A1 (same for all 4 sticks) 8Gb
- 2 modules in A2/B2, then A1/B1 (same for each pair) 16Gb
- 4 modules (fully populated) 32Gb

Modules are all Corsair Vengeance 2666 LPX (CMK16GX4M2A2666C16) bought as 2x8Gb (twice). Given large amount of stress testing for a stable uptime they seem fine. But something is odd at power up.

Machine boots perfectly from cold. BUT a sleep or restart (effectively same thing, no?) results in the post "Surge detected..." message and a 5d.

It's a brand new box (the only old component is a keyboard and a secondary drive). PSU is a EVGA 750W G2 jobbie so more than enough grunt.


Is there some odd bios setting that might need tweaking to get a stable boot for these modules or should it just work? I also presume whatever drivers or issues there are in Windows should have no bearing on this problem.

Any thoughts?
 
Solution
This sounds like a C state problem. I have heard of this happening with G1s (in particular the 80+ Bonze) but not G2s. What is the exact model of your PSU, is it a 80+ Bronze? If it is a C state issue then you could try setting your C State to C4 to verify or turn off the Hibernate Modes on Windows. One last trick you could try is turning the ECO Mode off on the power supply.

That all said, newer power supply such as your should not have this problem. Have you tried contacting EVGA? They usually have staller support.

1 (888) 880-3842

If you think it is a software issue then this post might shed some light...

This may also help...
This sounds like a C state problem. I have heard of this happening with G1s (in particular the 80+ Bonze) but not G2s. What is the exact model of your PSU, is it a 80+ Bronze? If it is a C state issue then you could try setting your C State to C4 to verify or turn off the Hibernate Modes on Windows. One last trick you could try is turning the ECO Mode off on the power supply.

That all said, newer power supply such as your should not have this problem. Have you tried contacting EVGA? They usually have staller support.

1 (888) 880-3842

If you think it is a software issue then this post might shed some light...

This may also help...

http://forums.evga.com/Second-1300w-g2-power-supply-still-the-same-problemConclusion-Found-m2035550.aspx

Not sure if this helps, I could be wrong as that is a good power supply.
 
Solution
Hey. Thanks for the response. So the PSU is a brand new (like I've just replaced it in the last half an hour) with an EVGA SuperNova G2 750W (Gold) so I'm going to punt the problem isn't that alone. From that link it looks like some C state problem so I'm off to look up sleep modes supported by m/b and such,