boot issues New Build, 2 different motherboards, no boot after initial CMOS clear unless...

[TiredoFAsus]

I'll try to make this as short as I possibly can..

 

Main Problems:

- The Motherboard refuses to turn on after the PC goes to sleep.

- Doesn't turn back on when a bios change that requires a hard reset happens, "ie, Disabling Hyper-threading or enabling XMP and many other options",

- Makes a click sound when I press the turn on power button but doesn't boot unless I disconnect some of the peripherals and then press the power button for longer than needed "i.e. a Tap doesn't turn it on, a long press "1.5-2 seconds is needed". 

- The most success I had with turning it on is when I disconnected the Audio Interface "GoXLR Mini", then it boots with a long press. However, even with the Audio interface disconnected,  it still doesn't boot back from sleep or a bios change, I would have to manually re-boot it with a long press again.

 

 

My full setup:

 

PC Hardware: 2 x Rog Strix Z790-I, 13700K, EK Nucleus 240 AIO, 7200CL34 G.Skill, Rog Strix 4090 OC Edition, SK Hynix Platinum P41 1TB Nvme Gen4, Thermaltake SFX 1000W ATX 3.0, Asus Prime AP201 Case.

 

Peripherals: Pulsar Xlite mini & Final Mouse Tenz, Steelseries Pro TKL & Wooting 60HE, LG 27' 2k IPS Monitor, GoXLR Mini Audio Interface, Sure SM7b Mic, Sennheiser HD560s.

 

Software: Windows 11 Home & Pro tested, both Latest Builds. BIOS, ME Version, GPU BIOS Version, Chipset and everything else was updated through the motherboard's support page.

 

 

Things I've tried:

- Bios & all other software available on the support website are used to update the system.

- I've tried multiple combinations of USB ports to see if one of them was the issue.

- Double checked my installation, cables are well seated, nothing is bent too hard or squeezed in a place that it shouldn't be at.

- Disconnected the GPU and plugged in an HDMI straight into the mobo.

- Disconnected both ram stick separately.

- Disconnected chassis fan.

- Disconnected ARGB header for the AIO.

- Disconnected the Power Button from the motherboard and used the reset key on "ROG Hive" to turn it on. 

- Returned the old motherboard and ordered a new one thinking it was the issue.

- Increased CPU Core Boot Voltage to 1.5v.

- Increased some of the RAM dependent voltages as instructed by Asus's support.

- Tried different random Bios options to see if it helps.

- Did a 2 hour cinebench R23 test, and 2x memtest68 and they both passed without an issue.

- Tried a high 3500W Power strip, 2 of them.

- Tried connecting the PC straight to the socket.

- Tried using HDMI instead of DP when connecting the monitor to the GPU.

- Disabled the earth conductor pin on the receptacle to supply the strip only with phase and neutral.

 

 

More Details:

 

at some point I started with disconnecting some of the peripherals "I don't have many" and I soon found out that disconnecting the Audio Interface would yield the best chances of this system to be able to boot. Even then, waking from sleep never works, I just hear a click sound and the PC just shuts off. A bios change requiring a hard reset doesn't work either. Just to be clear, all those peripherals were working normally with the older build a month and a half ago.

 

What's more confusing is, with the new mobo the system is exhibiting the exact same problems as before. For example, after clearing the CMOS, I can with 100% success rate turn the system on with a single tap on the power button, even while all the peripherals are connected. Yet I can't log into windows 11. If I go to advanced settings in bios, I don't see an option to access trusted platform settings "TPM". Which is weird since i need it to boot windows 11 otherwise I won't be able to, the only way for the TPM option to appear is to fully shutdown the system and then reboot it. More to that, after shutting down the system, now I'm back to needing a longer press on the power button and 2-3 clicks for the PSU/Motherboard for a boot "and I can't boot it with all the peripherals connected, I'd have to start disconnecting some"

 

The PC in general works fine aside from that! Games run smoothly without any problems, everything else is good, benchmarks show no issues regarding temps or voltages whatsoever. The only problem is, it doesn't freaking boot normally or go to sleep. the PC Case itself has -ve air pressure naturally "All mesh panels", not being able to put the PC to sleep means that I will have to air dust it every 3-4 weeks. 

 

I tried with Asus's Support and I have posted in ROG forums yet it yielded nothing but utter disappointment and a waste of my time. After almost 5 weeks of emailing their support, the agent decided that I needed to RMA the board through the retailer that I bought it from, and now I'm not so sure that it was a hardware issue with the board itself, Or am I just that unlucky and got the same bad seed twice? Both boards have the same batch number yet one is produced in 11/2022 and the other in 2/2023 "new one"!

 

Any Boot Voltages that I can "SAFELY" play with? If anything comes to mind, please do share. I'm desperate here!

 

A bit of a visual explanation if anyone's interested:

 

 

[RONOTHAN##]

18 minutes ago, TiredoFAsus said:

Any Boot Voltages that I can "SAFELY" play with? If anything comes to mind, please do share. I'm desperate here!

The only ones I can think of are VCCIN and the PLL voltages, but the only reason to touch these are for either high memory overclocks (7000+) and cold bug removal when on LN2. They shouldn't affect your ability to boot at ambient. If you care, increasing VCCIN helps the ability to boot at low temps but also reduces the max memory speed, I'd want to leave it below 2V. The PLL voltages sweet spot hard and vary per chip, so you'd have to try each of them, increasing until it works. I'd want to leave those below ~1.1V, though most chips wouldn't scale that high anyway and the Strix board might not even let you set it above that anyway (my Unify-X tops out at 1.135V for all of them). 

 

Main thing I'd be thinking for what to try would maybe be a different BIOS revision. Latest does not always mean best, and from the bit of time I remember browsing the Maximus threads back when I had a Z690 Apex said there was a couple of BIOS revisions that were pretty terrible out there. It might be worth reverting to a version or two back and see if your issues starts working. Fast Boot I've also seen cause some weird issues where it doesn't want to reboot, but that shouldn't cause it to not start in general. Another thing that might be worth checking depending on how easy it is to get at is to check the CMOS battery voltage, some boards can be very particular about wanting it at a certain voltage level and will give some very strange behavior otherwise (Luumi's SR-3 Dark, for instance)

[Whatisthis]

Take the board out of the case with only  cpu, ram, boot drive and power connected. Start it by shorting the power pins with a screw driver of jumper. See what it does then. Sometimes the case / standoffs can cause word things. It will also eliminate the peripherals issue. If it boots fine, start plugging things back in one at a time until the issue comes back.

[TiredoFAsus]

7 hours ago, RONOTHAN## said:

 

As for BIOS revisions, I've tried the one that comes with the board which is 7xx, then I updated into 8xx, afterwards they dropped another update while I was troubleshooting the first board "904" and its the latest. They are all exhibiting the same problems. I've also tried to disable fast boot both in bios and windows power plan settings. As for checking the CMOS, the battery is probably buried somewhere under the heatsinks, it's not impossible to reach but I'd like to keep the motherboard intact incase I wanted to mark it as another DOA and ship it back to Amazon.

 

Funny you mentioned the Apex Z690. I purchased one last year, and thank God while I was waiting for the 12900KS to arrive, I saw that the batch of my apex is the same one that everyone was having problems with. I immediately returned it, cancelled all other orders and decided to just wait for the platform to mature a bit.

[RONOTHAN##]

26 minutes ago, TiredoFAsus said:

Funny you mentioned the Apex Z690. I purchased one last year, and thank God while I was waiting for the 12900KS to arrive, I saw that the batch of my apex is the same one that everyone was having problems with. I immediately returned it, cancelled all other orders and decided to just wait for the platform to mature a bit.

The one I bought was from that batch as well. The guy I bought it from (there's no way I'm spending $700+ on a motherboard, it was second hand) said it was one of the good ones. He lied, and it was worse than a Z790 Aorus Elite for memory overclocking even with the 13th gen memory controller. It wasn't as bad as some of them, I could get it to boot 7600MT/s, I just couldn't get it to be stable at anything above 6600 with M die and 7200 with A die, while the same CPU and RAM on the Unify-X did 7000 with M die and 7800 with A die (on the most recent BIOS 8000 looks possible as well). The BIOS had all the different voltage settings you need scattered across 4 submenus (to be clear, this is a complaint I have with the other Maximus boards since Z170, it's just way worse on 13th gen cause you actually need all those hidden voltages), and was just overall a mess of a board. That board was just a mess, the only thing redeemable about it was Turbo VCore being an actually good OC utility and a bit more stable than Dragon Power I have with the Unify. 

 

You didn't miss out on much. That board was a nightmare and I sold it along with my kit of M die to a friend (he wasn't planning on memory overclocking, so it was actually good for him, better than the Aorus Elite he would've had otherwise) within two weeks of getting it. 

 

29 minutes ago, TiredoFAsus said:

As for BIOS revisions, I've tried the one that comes with the board which is 7xx, then I updated into 8xx, afterwards they dropped another update while I was troubleshooting the first board "904" and its the latest. They are all exhibiting the same problems. I've also tried to disable fast boot both in bios and windows power plan settings. As for checking the CMOS, the battery is probably buried somewhere under the heatsinks, it's not impossible to reach but I'd like to keep the motherboard intact incase I wanted to mark it as another DOA and ship it back to Amazon.

Looking at photos, you might just have to take the grill off the IO cover in order to get at it. Can't tell for certain though. 

 

TBH at this point, if you don't want to check the clear CMOS, I'd just return it and get something else. The Z790 Edge is a very well reviewed board and is a bit cheaper as well. 

[TiredoFAsus]

46 minutes ago, RONOTHAN## said:

 

Yea thankfully I did dodge a bullet with that apex, and that platform as a whole. Yet I certainly ate another one with this! Funny thing is, when I returned the 1st Z790I Strix, I was so frustrated from the issue, Asus's Customer Service and their forums. While I was looking at other ITX board offerings from different manufacturers, I did see the edge yet somehow my brain said let's give Asus another chance! 

 

To be fair, I'm not really leaning that much into the board being the main culprit. As I said, how on earth did I manage to get 2 boards, with 4 months of manufacturing dates apart that have the same issue? While I didn't find any problems that resemble mine with this board, I found some DOA from customer reviews, yet a DOA could mean so many things.

 

Could you elaborate more on PLL voltages? I've had my fair share of bios tweaks in the past, yet PLLs are something that I never got into or needed to adjust.

[RONOTHAN##]

1 minute ago, TiredoFAsus said:

Could you elaborate more on PLL voltages? I've had my fair share of bios tweaks in the past, yet PLLs are something that I never got into or needed to adjust.

So there are 5 different PLLs, and IIRC ASUS boards let you set boot and runtime versions of each (though that might be exclusive to the Maximus boards). There's one for the CPU as a whole, one for the Ring, one for the System Agent, one for the E cores, and one for the memory controller. Under extreme cold conditions, setting these a bit higher can help keep the system booting (this is usually what LN2 mode sets), while at ambient setting each of these can help stabilize memory settings above 7000MT/s. At ambient, my 13700K likes 1.035V on the CPU, Ring, and SA and 0.990V for the MC. The E core setting I left on auto because it doesn't really do anything if you aren't trying to max out your E cores which I can't be bothered to do. These only kinda became relevant on 12th gen, it could help Ring, CPU, and memory overclocking a bit as it slightly reduced core voltage requirements, but it helps a lot more on 13th gen as it can be the difference between getting say DDR5 6800 working and DDR5 7600 working. 

 

I've got no idea what you might need to set them to, I've never used a 13th gen chip on LN2 and forget what the go-to LN2 mode settings for them are, but every chip is different so you might just need to experiment with them a bit. Again, I'd try to keep them under ~1.1V, and know that they only go in 0.015V increments by Intel spec (these voltages are generated by the CPU itself), so don't bother trying to set them in 0.001V increments like at least my Apex would let me set. 

[TiredoFAsus]

aight mate, thanks a lot. I'll give it a try and report back. 

[TiredoFAsus]

10 hours ago, RONOTHAN## said:

So there are 5 different PLLs, and IIRC ASUS boards let you set boot and runtime versions of each (though that might be exclusive to the Maximus boards). There's one for the CPU as a whole, one for the Ring, one for the System Agent, one for the E cores, and one for the memory controller. Under extreme cold conditions, setting these a bit higher can help keep the system booting (this is usually what LN2 mode sets), while at ambient setting each of these can help stabilize memory settings above 7000MT/s. At ambient, my 13700K likes 1.035V on the CPU, Ring, and SA and 0.990V for the MC. The E core setting I left on auto because it doesn't really do anything if you aren't trying to max out your E cores which I can't be bothered to do. These only kinda became relevant on 12th gen, it could help Ring, CPU, and memory overclocking a bit as it slightly reduced core voltage requirements, but it helps a lot more on 13th gen as it can be the difference between getting say DDR5 6800 working and DDR5 7600 working. 

 

I've got no idea what you might need to set them to, I've never used a 13th gen chip on LN2 and forget what the go-to LN2 mode settings for them are, but every chip is different so you might just need to experiment with them a bit. Again, I'd try to keep them under ~1.1V, and know that they only go in 0.015V increments by Intel spec (these voltages are generated by the CPU itself), so don't bother trying to set them in 0.001V increments like at least my Apex would let me set. 

Unfortunately PLL voltages did not work, if i go as high as 1.1 on some, the system RGB/LEDs will only turn on and the boot fails. I might try windows 10 just to see if the whole TPM and other secure boot options are the reason. 

[TiredoFAsus]

Just to update this thread;

 

Upon testing the power supply the first time, the technician said it was working well and in order. The other technician however had another concern which was the 5V rail on that unit, he said you pull as much as 1000W on this unit without it failing, but it's initial supply of current be either too high for the motherboard or too low which in both cases will emulate a failed boot. He carried on with his testing and the power supply's 5V rail indeed was having some sort of incompatibly issue with the board or the peripherals, he didn't know exactly what was causing the immediate shutdown and wanted to do more testing on his time without even charging me for. I told him that I'll let him do so only if Amazon rejected this as a return.

 

Amazon's support on the other hand had already asked me to pack the motherboard and ship it back for a full refund lol, when I explained to them next day how their support almost ended up costing them money and that I took it to a technician and paid out of my own pocket to make sure the board is the problem, they ended giving me an exception on my 2 month old Thermaltake power supply that I got from them plus a $50 promotional credit. Which is awesome, still does not replace my wasted time, yet that was extremely generous of them!

 

On the other hand, the first technician that checked on the system "with the old motherboard" ended up costing the retailer the price of that motherboard in new condition. They cannot ship the old one back as "DOA" apparently, since it was the PSU's fault all the way, Plus now its an open+used motherboard. 

 

Now I have an Asus Tough 1000W Atx3.0 and it's working perfectly fine, no failed boots, no need to disconnect any of the peripherals, computer can go in and out of sleep instantly. Thankfully everything is working now! Thanks RONOTHAN## for your input.