Z390-F BIOS updates (1902, 1903, 2004) all causing POST failures due to memory speed

[xeromist]

I'm not sure if you checked all of the secondary timings for discrepancies but I could definitely see a situation where a bug could cause one of the secondary timings to be waaaay off and cause instability. It's not unprecedented as I remember a situation exactly like that and after engineers found the bug a new BIOS release allowed the memory to work again. I just can't remember if that was something the end users were able to see or not.

Anyway, I'm glad you found some working memory.

[Billycar11]

I did try significantly loosening the timings on my 4 x 8GB, 3000Mhz, 16-18-18-38 sticks using BIOS 2004 to no avail. It's specifically the memory frequency setting that is the problem. Setting anything above 2400Mhz when all four DIMM slots are in use causes instability and POST failure on any BIOS after 1802.

Sadly I got impatient and decided to just replace my 4 x 8GB 3000Mhz sticks with a set of 2 x 16GB 3200Mhz sticks. I can now use BIOS 2004 without issue at rated memory speed which just validates my original suspicions about the nature of the instability. If I were to add two more of the same sticks I'm certain the issue would happen again.

Like I said the instability on 1902, 1903 and 2004 was caused by filling all four DIMM slots and trying to use anything higher than 2400Mhz. On BIOS 1802 this problem didn't exist.

ah man now asus will never care about 1 guy now im going to be SOL lol im not buying new ram i just spent $1800+ on a dumb strix 3080 ti that has its own issues

[dingo99]

Just to add another data point:

XI Formula here, running 64GB on mismatched sets of 16GBx2 DIMMs (one pair with Samsung chips, the other with Hynix), at 3200MHz 16-18-18-36. I've had perfect stability across all BIOS releases (just updated to 2004).

So, whatever might be happening is more complex than just having 4 DIMMs running at XMP speeds. Those experiencing problems definitely should try xeromist's suggestion of comparing ALL secondary timings between BIOS versions.

[ionised]

ah man now asus will never care about 1 guy now im going to be SOL lol im not buying new ram i just spent $1800+ on a dumb strix 3080 ti that has its own issues

I got the impression that they weren't even convinced this was an issue so I had no other choice unfortunately. I managed to sell my old 4 x 8 GB sticks on eBay for close to what I paid for the new 2 x 16GB set I haven't taken much of a hit luckily.

[lordoftears]

I have the same issue. My computer works perfectly with 1902 but it does a recursive reboot before POST with 1903 and 2004.
If I activate TPM in 1902, I experience the same issue.

My computer is equipped with 2 x 16Gb CMK32GX4M2B3200C16 Corsair RAM, but they are not either version 4.31 or 5.39, that are indicated in QVL list.

I contacted the ASUS assistant and they told me that this could be the cause.

I should try a different kind of memory or wait for a further BIOS update.

[Jesseinsf]

I have an Asus Z390 motherboard with a core i9 9900k clocked at 4.9Ghz all cores, and I am using Corsair Vengeance Pro RGB 128GB RGB (32GB per stick) at 3200Mhz. I have not experienced any issue with BIOS 2004.

What is crossing my mind with your issue is that any BIOS after 1802 may be targeting certain Asus Z390 motherboards. The only change after BIOS 1802 was enabling Resizable BAR. My hunch is the Resizable BAR may have caused a conflict with your RAM model and/or the memory controller. The only way to test this theory would be to try different RAM sticks at 3000Mhz and 3200Mhz.

[Billycar11]

I'm not sure if you checked all of the secondary timings for discrepancies but I could definitely see a situation where a bug could cause one of the secondary timings to be waaaay off and cause instability. It's not unprecedented as I remember a situation exactly like that and after engineers found the bug a new BIOS release allowed the memory to work again. I just can't remember if that was something the end users were able to see or not.

Anyway, I'm glad you found some working memory.

this is the case i put all the timings in manually saved it to the old bios saved it as a cmo then restored it on the new bios and i all works.
i finally have resizable bar on my 3080ti strix
strange that just enabling xmp on the new bios will ruin it all

[lordoftears]

this is the case i put all the timings in manually saved it to the old bios saved it as a cmo then restored it on the new bios and i all works.
i finally have resizable bar on my 3080ti strix
strange that just enabling xmp on the new bios will ruin it all

Hi, I would like to try this, saving the timings on the *.CMO file.
Which are exactly the parameters to be set in manual mode? I mean, which menu items are on the BIOS exactly? Sorry, but I would like to be sure about that before trying.

[test_lab]

I'm also experiencing RAM speed related issues with my Z390-F. My BIOS is at 2004. But in my case i get BSOD'S and freezing when i disable XMP. So the other way around. I tried two different RAM kits. May i ask to have a look at my thread?

https://rog.asus.com/forum/showthrea...(XMP-disabled)

[lordoftears]

In the end, I tried replacing my RAM kit with a fully compatible, one listed on the ASUS website (32Go, 2x16Go, DIMM DDR4 G.Skill RipJaws V PC4-25600 (3200 Mhz) (Nero), F4-3200C16D-32GVK ). Again, the same issue.

I also tried to export a txt file with the settings of bios 1902 and 2004, to compare them, but all the default settings are identical.

I suspect that behind some "auto" setting there is something different among the two bios that make my computer get stuck at POST.

So, I am still with 1902 bios.