Maximus IX Hero won't boot, Q-code 00, without power off/on

[soccastar001]

I have a 7700K in a Maximus IX Hero. I tested all the components and put the system together and it worked great for a few days then on the 3rd or 4th day after use it started not booting when powered on. I can see the machine is getting power. All the LEDs are on, the fans and pump is running, but it won't boot and I get the Q-code 00. What's very odd is that if I switch the PSU off (or unplug it) and then plug it back in it will boot without issue and be used the rest of the evening without issue, even if I restart or shutdown and turn it back on. It's only when it has been sitting for 6+ hours that I have this issue.

It also appears to be somewhat related to the CPU OC I applied. When I set the CPU back to base clock in the BIOS it booted fine and I didn't experience this issue. Of course I have a 7700K and a custom water loop so I want to be able to OC, I'm just not sure what precisely is causing the issue. I had the CPU at 4.9GHz and 3.1v and it was stable and passed stress tests but I had this issue. I lowered it down to 4.7GHz and still had the issue so I figured I needed to resolve it another way because at 4.6 or lower that is barely an OC and not what I should be getting from my 7700K.

Any experience or help would be great.

[Chino]

Seems like an unstable overclock. A stable system shouldn post without hiccups.

[soccastar001]

I feel like there is something more going on. When I don't have this issue I was able to run an 8-hour overnight stress test on the CPU at 4.9 with Realbench without errors. And like I said, the problem persisted all the way down to a 4.7 OC. I should be able to OC a 7700K more than 100MHz over boost clock in this system, it is using a custom loop with triple radiator. The problem is really sporadic and only when the machine has been sitting.

[Findolin]

I feel like there is something more going on. When I don't have this issue I was able to run an 8-hour overnight stress test on the CPU at 4.9 with Realbench without errors. And like I said, the problem persisted all the way down to a 4.7 OC. I should be able to OC a 7700K more than 100MHz over boost clock in this system, it is using a custom loop with triple radiator. The problem is really sporadic and only when the machine has been sitting.

I have the same problem and created a topic about this a few days ago.
I noticed that the problem is gone when setting the RAM clock to 2933Mhz (G.Skill Trident Z RGB 3600Mhz). Changing the CPU clock won't change anything. Even at 5Ghz everything runs fine (with the lower RAM clock)
I tried different BIOS settings, tuned the VCCIO und System Agent voltage and even replaced the RAM. Nothing would help. Same with the new BIOS version that came out some days ago. So i think this is definitely a BIOS issue. Right now i am going with the lower RAM clock. Not very satisfying, but at least it will boot every time.

[soccastar001]

Thanks for the reply, I thought I was going crazy!

It is disattisfying to lower memory clock but a drop from 3200 to 2933 shouldn't be too much of a hit.

[Chino]

I feel like there is something more going on. When I don't have this issue I was able to run an 8-hour overnight stress test on the CPU at 4.9 with Realbench without errors. And like I said, the problem persisted all the way down to a 4.7 OC. I should be able to OC a 7700K more than 100MHz over boost clock in this system, it is using a custom loop with triple radiator. The problem is really sporadic and only when the machine has been sitting.

Realbench is great for testing CPU stability not so much for DDR4.

Thanks for the reply, I thought I was going crazy!

It is disattisfying to lower memory clock but a drop from 3200 to 2933 shouldn't be too much of a hit.

If you're running DDR4-3200, then your memory overclock is probably unstable. Use HCI Memtest or Google Stressapptest for testing DDR4 stability.

[soccastar001]

I was finally able to resolve the issue. It looks like the XMP profile for my RAM (Corsair Vengeance DDR4-3200 16-18-18-36) was the culprit. Specifically the automatic values for VCCIO and SA. With some fine tuning I settled upon 1.2 for VCCIO (which is lower than the automatic value the BIOS was setting it at) and 1.225 for SA (higher than the automatic value). This took quite a bit of finagling, especially since the edits had to go in two different directions than the BIOS/XMP profile was setting them at automatically. Once I stumbled upon these values though the machine has been booting without issue for a few days now. Thanks for all the advice.

[Brutaka]

I'm getting this same issue while running an i7-6700k (4.6 OC), asus rog IX hero, and Corsair Vengeance DDR4-3200 (16-18-18-36). I've been trying to tune the VCCIO and SA voltage values nut nothing seems to be the sweet spot. Any advice would help.

Also I copied exactly how this YouTuber did her OC'ing. Even though the IX is an upgrade by a bit, I'm certain it should be the same between the two.

https://www.youtube.com/watch?v=gZe26lv7lJ4&t

The only things I've changed from the video is

DRAM Voltage: 1.31
VCCIO: 1.225
System Agent: 1.25

I'm still getting hangs when I turn on my PC. However if I were to click the reset button, the PC will restart itself and continue as if nothing happened. So why do I have to press two buttons to actually get into windows?

It's odd how my PC will work if I set the VCCIO and system agent voltage to auto while setting my dram frequency to 1600. That's half the speed of 3200. Why does 3200 not work?

[Menthol]

Brutaka,
Fine tuning voltages is required when overclocking and every CPU system is unique so copying from a YouTube video while may seem like a good stating point will require tuning and testing
If the voltages you show are correct your DRAM voltage is to low for 3200mhz, 1.35 is minimum for that speed and may require 1.36 to 1.38 to be stable

[Brutaka]

Brutaka,
Fine tuning voltages is required when overclocking and every CPU system is unique so copying from a YouTube video while may seem like a good stating point will require tuning and testing
If the voltages you show are correct your DRAM voltage is to low for 3200mhz, 1.35 is minimum for that speed and may require 1.36 to 1.38 to be stable

Thanks for the reply. I thought I wouldn't have any support on this issue. I'll definitely try this at home. Sucks to not have a PC to boot up properly. :/

To be completely honest, I wouldn't know how to do all this voltage setup by myself. I'm more of a visual learner and have NO idea how to make the voltage the "correct" voltage. I keep on seeing color changes with the collage, like yellow to purple and such. I'm assuming it means I'm getting to a high point where I need to be cautious. Another thing that keeps getting me is when I change the collage, let's say vccio to 1.225, the voltage it's outputting will say 1.227 and I'll think "hey, I need to round it up" and chance the vccio value to 1.23. I'm pretty certain that's not the correct way but I'm still a noob when it comes to voltage tuning.

Also, what happens if I raise the dram voltage to 1.4? Would that be to much and cause problems? I'm just not sure why one wouldn't raise the voltage. My only assumption is raising it would be like revving a car and having it run constantly, thus wearing it down.

Edit #1: There's a couple of changes I've made. First was the DRAM Timing Control and setting the Maximus Tweak to Mode 1. Next was changing the VCCIO to 1.1625 and System Agent to 1.15. I haven't changed my DRAM voltage, though it's set to 1.31. So far I haven't had any issues. Though I've noticed if I set my PC to "stock" settings in the BIOS, my PC will take 14-16 seconds to boot to the OS. When I do this custom profile, it takes 24-27 seconds. Not sure why.