Thanks for confirming, and thanks for agreeing to test it all! I know you have programmer, so testing a failure is easy to fix, but I wasn't sure if you'd want to waste the time on a bunch more test BIOS that may all be fail.

Looking at all this here in a little bit!

Edited my post, for some further information.

It's not a waste of time. It is a research project , and as in every research some things fail while others success.

It is a really interesting work for me.

That's how I'd feel too! I'll check the edit again in a bit, see if you get the chip numbers or not. I was thinking some part of the success BIOS transfer may be needed, while other parts may not.
The Kaby modules you found, those are probably needed or at least maybe needed to be edited to match/add in kaby sections, but some may not be needed from that BIOS. Hopefully some of the ones I note with HDMI mentioned are not necessary to transfer over.

For the HDMI tests though, for now, I only plan to compare where I find HDMI, then transfer those sections into Frankenstien as test 1. Test 2, edit those particular modules where it looks feasible with the OG's differing data leaving most of the original code in place.
Test 3, maybe a combo of both of those methods.

Can you please reupload G20CI-to-G20CB-1402-BIOS-VBIOS1058-GOP1079-unlicensed.bin I get certificate error on that page linked. Never mind, got it!

So, it is a ASM1442K from AsMedia on my G20CB board. Don´t know what is on the G20CI.

An overview can be found here.

It is a simple level shifter. But it has Enable/Disable. Unfortunately I can´t find a pinout to check if En/Dis is driven correctly Datasheet found. We have to think about that PCIe to the ASM1442K is working correcty, but the chip itself is disabled, due to a wrong enablepin connection to the GPIO(?) of PCH... -> Checked and is working as it should.

Thanks! So that leaves us at finding the other boards chip to confirm it's different, correct? Then, if it's not, the issue lies in some other thing unrelated to the chip itself.

There is a I2C interface on that chip, for some configuration i think.

It may be another problem than the chip. Till now i found nothing

I took an oszilloscope and did some measurements.

- Original G20CB Bios with Skylake: Visible PCIe data and I2C communication is visible to and from ASM1442 chip. Some initialization parameters. After that, screen comes online.

- Modified G20CI Bios with Skylake/Kabylake: neither PCIe data nor I2C communication.

AND NOW I discovered that G20CI does NOT have HDMI onboard!!!!! That I could have checked before… I am an idiot!!!

But will it be possible to implement it??????


Okay... one issue is eliminated.

Lets head on to the temperature issue, before implementing new features.

Read something, that it is possible that a BIOS is reading the Tj max temperature.

So I booted to Windows, launched Coretemp… indeed. Tj is 100°C. For some reason, Tj max is read as CPU temperature. As a result ASUS Software is reading that Value too.

Coretemp reads out everything correct from the cpu using DTS.

I assume, that something has changed with that SIO chip, that may use the PECI interface (one-wire interface to read-out temps from the cpu). On a G20CI there is a NCT5579 while I have a NCT5568.

So we need to find way to add inHDMI related stuff, not find and edit/carry over settings etc. Good method of testing to figure that out!

I need more time! I'll try to look into all of this soon as I can, but knowing your progress speed lately you may figure it out much quicker than I get into trying to narrow it down

Take the time you need.

Meanwhile I played with the Super-IO issue, replaced the driver and some related PEI and DXE drivers, resulting in non booting bios.

I think I need a dissassembler to disassemble some things...

I wish I could find assembler/disassembler 101 guide!
Did you ever test this Franken BIOS putting back in your other updated vBIOS from G20CB original vBIOS, since it would tie into having HDMI vs the other not, maybe that's good test to run if you have not already.
Or is current vBIOS updated from your original board already?

The Frankenstein BIOS has the VBIOS from G20CB integrated from the beginning. There is much more missing. Especially the I2C configuration and the PCIe mapping to what ever PCIe port. And, to complete it all, an port entry in the DMI table.

Well that is good to know already original vBIOS. I hope to have time to dig around tomorrow night!

Tried my first disassembling tests... they failed.

They failed in that matter, that uefitool is hanging up itself replacing a patched module.

Currently I am trying to get the SuperIO chip configured correct. I believe, that is the most important thing. I suspect, that the fans are not working correct, too...

You can support the 7 and 8 generation CPU by porting H170I or Z170I BIOS. But some of the functions may have an impact!
I succeeded in lighting this board over 8400!