Specific BIOS modding problems with an ASUS ROG G751-JY notebook

BIOS/UEFI Modding BIOS Modding Guides and Problems
47

@EthanWalker - You’re welcome!

1. Yes, never flash stock BIOS with FPT, especially Asus BIOS as it’s in a capsule too (so wrong size). But in general this will cause you to loose all NVRAM volumes, all board specifics like UUID and serial, and often LAN MAC ID too if full BIOS flash is done.
And yes, if your initial FPT backup should be used as a source file for any FPT reflash. If you want a mod BIOS, it should be made from that file, if you are flashing with FPT. To further add here, whenever it’s possible to use FPT vs AFU, this is easier and safer (in my opinion)

2. Yes - type each command one line at a time, enter after first, then onto second one. Also, do not use capitol letters in Setup or Var, I know you didn’t above, but just mention in general as this will cause error.
setup_var 0xA2 0x0
setup_var 0xA3 0x0

3. In both instances you mention above, either way BIOS Lock and SMI lock have to be disabled before you can flash.
You can fix this with AFU, or FPT f you want, now that we know /N /GAN will correct the issues brought up in this case. But yes, for sanity and safety I would suggest you first unlock the locks, then reflash the stock FPT BIOS region dump again (Either untouched, or edited to disable this locks first)
If you reflash stock FPT dumped BIOS region as-is, you will need to again redo the unlocks before you flash follow that up with the mod BIOS flash. And on that note, you need to let me make you a new BIOS mod with your FPT BIOS dump as the source if this is your intention.
I can make you two if you want, one with only the two locks unlocked, and another with mod + both locks unlocked. Or I can remake you a single mod BIOS, and that can be used after your initial stock FPT BIOS region reflash, but after that and before you can flash in this mod BIOS you will need to unlock the locks again

I will redo your mod BIOS now, using your FPT Backup as source. If you are here reading this now and you don’t see a link, please wait/refresh and I will edit link into this thread once I’m done.
Additionally, if you are reading this soon after I post it, it will take me some time (not minutes), so the wait may be an hour or more as I will have to redo BIOS mod from scratch since pre-edited BIOS (Stock I made for Svan) will not be used.

@Svan - yes, stock BIOS is in 2kb capsule. Thanks for your dump, I will check it so I can see where all the board specifics are kept in this BIOS.
Please PM me your original MAC ID if you don’t mind, thanks. Just so I can confirm it’s not also in another area you missed and didn’t change Never mind, I confirmed with EthanWalkers dump that MAC ID is only in the second padding file, along with UUID, DTS key, Serial, board ID etc

** Edit - @EthanWalker - Here is new mod BIOS, built using your original FPT backup as source
http://s000.tinyupload.com/index.php?fil…231575434765315

Flash using this command >> FPTw.exe -bios -f EthanWalker2.bin
** Please change extension .213 to .bin before flashing, sorry I forgot and FPT may error out if trying to flash as named.


@Fernando - In case you’re curious on this particular mod we’ve been discussing past few days, here is image example of my work folder, showing this took 11 interim BIOS builds to get all changes done properly without affecting padding files (add/remove etc + Fix FIT)
It was actually more than this too, maybe 6-8+ others I deleted as I sorted out which way to do what first (again). Also, a further few between these shown edits where removed as well, due to testing different edit moves or change orders that failed, so also deleted.

This is final count and required steps/process or BIOS broken if done in other moves/edits/order. I tried to name the files as edits done, so you can maybe see how the process was done USBS = AsusUSBS module near end of main volume, LAN = LANDriver near end of volume

EditProcess.png



Here is final edit compared to stock

EthanWalkerFinal2FPTBase.png



And here is change log
1. EFI RST 14.8.2.2397
2. oROM RST 14.8.2.2397 TRIM Mod
3. NVME + M.2 Mod Insert
4. All microcodes removed in main volume except updated 40661 << This cut down may not be required either, if further juggling and re-edit/check/re-edit/remove/add/check etc figured out a way around it, but I spent long time trying to get around it first time around, possibly not long enough though
5. All original microcodes remain in secondary (last) volume >> updated
6. BIOS Lock + SMI Lock Disabled

MMTool could not be used at all, as we’ve discussed, it adds a padding file no matter what you do, however since we’ve been discussing and I keep redoing these, I actually have not tested if it still happens if using my special method of removing a few key modules and changing main ucode volume count.
Since UEFITool does similar too, I assume MMTool may also be able to be used in similar way, but only by doing the mod as I’ve used EFITool 25 here, creating all these interim builds as you recompile and check everything each move to ensure no added padding file

48



20190725_110245.jpg
20190725_110245.jpg800×500 127 KB


20190725_110517.jpg
20190725_110517.jpg800×513 126 KB



Scr00.jpg
Scr00.jpg645×577 121 KB






UPDATE:

Flashbacked to (FPT)backup I made before flashing.
biosreg2.bin = backup = currentbkup.bin from post #24
but still afraid to restart

49

In your first FPT image, with the data mismatch, what BIOS is that one? I’ve only seen similar issues when someone used AFU previously, so it’s something AFU messed up, nothing we can do.
You can reboot now, since you flashed back in your backup. Then fix keyboard LAN etc by using the first mod I sent you and AFU /N /GAN.

Once you programmer arrives you can put on clean BIOS properly, then after that you should be able to FPT flash without these issues.

50

@Lost_N_BIOS I don’t know, I did everything as you said. Disabled the locks, Flashed the biosreg.bin with FPT. Maybe I should try again or maybe that bin is corrupted ?
My LAN works, the keyboard doesn’t.

The programmer arrives today, so I don’t know if there is any point in flashing with AFU. But if it is needed, I can do it.

Update:
Received the programmer.

51

Something I don’t understand if you go to Asus forums you can find a guide how to insert NVMe Pcie module and the guy is using only AFUWINGUI 3.07.01 to backup and flash (for flash he select main bios in setup tab)

ROG G751-JY BIOS 211 NVMe Upgrade Guide Step by Step

@EthanWalker

Great picture do you know wich bios chip model is inside your laptop ?

52


I’ll find out later. As I said, I’m not good at modding, programming etc.

53

@EthanWalker - It’s as I mentioned, you used AFU previously, which messes stuff up for this.

To fix your KB do the AFU flash as mentioned >> AFU filename /N /GAN

If you want to fix with programmer now, reflash stock BIOS using stock method (from within BIOS if this system has EZ Flash)

Then dump the BIOS with programmer and send to me.

Here is software and guide for the programmer
http://s000.tinyupload.com/index.php?fil…257455007472602

[GUIDE] Flash BIOS with CH341A programmer
https://www.bios-mods.com/forum/Thread-G…341A-programmer << Same as one directly above, but with all images expanded/visible at once

@Svan - yes, that may be possible way to properly flash mod BIOS in this system using AFU without using /GAN
Only you can find out. I didn’t lead you guys into /GAN flashing, and personally always suggest against AFU unless it’s the only way or already being used etc.
BIOS region is all we edit for BIOS mod, in this instance (aside from the NVRAM issues pre-introduced by you guys flashing with /GAN) NVRAM is within the BIOS region, but the way AFU is setup it doesn’t flash them when you select BIOS region, additional checks must be set to flash NVRAM. Of which I am unsure for the GUI version, you’d have to look, but /N or /K or K* (*=numbers) are the switches for such in CMD line or DOS

54

@Lost_N_BIOS

I tried flashing the modded bios with afudos modbios.* /P and then backup again to check everything (/P = Program main bios image)

1) The MAC address is in the file in hex mod (hardware working in windows)
2) Backlight keyboard is working.
3) EFI + OROM IRST are updated in UBU
4) PCIe NVMe is present in UBU
5) The MC code is updated in HWINFO, only got this difference in UBU :

Earilier flashed with /GAN :



New one flashed with /P :




Wich tools did you use to see MAC, UUID, DTS key, Serial, board ID ? (I tried FD44 editor but seems not working with laptop bios).
If I can find in hex the MAC adress on the new backup that mean all those data are there too in the same part of the bios ?

I understand it’s a hard work on this bios. Since I found a good way to flash can you do again the mod without touching original Microcodes ? Only EFI + OROM IRST 14.8.2.2397 and NvmExpressDxe_4.

If you think that is mandatory to do the mod on the original AFUDOS backup. Tell me, I will send you a private message with a link.

Thanx in advance.

55


I got it, that’s why I decided to go with FPT and Programmer. I trust you. (heart emoji + beer + beer + beer + beer)

So,
I downloaded the stock Bios from Asus Support

Step 1: Tried to flash it within Bios (Easy Flash)
Status: FAILED
Reason: Build date is too old !



Step 2: Flashing with Asus WinFlash tool.
Status: FAILED
Reason: The Bios file is older than the current Bios version.


Step 3: Flashing using WinFlash + Command Prompt.
Details: *Opened CMD + Administrator Privileges; Command 1: cd C:\Program Files (x86)\ASUS\WinFlash; Command 2: Winflash /nodate
Selected the desired Bios, then Update/Flash.
Auto Reboot.
Easy Flash started.
Erasing…100% Programming…100% Verifying… 100%
Status: Completed BUT I don’t know. Not sure if it was flashed/downgraded correctly.

What’s new:
*The keyboard’s LEDs still not working.
*Touchpad ‘‘Scroll’’ doesn’t work. (cannot scroll pages)


I wanted to skip this step and go to ‘‘dumping’’ Bios with the programmer but then I decided to wait for your instructions.
Reason: If I got it right: Re-flashing stock BIOS should have brought the keyboard to life.
It didn’t, so this means that Something went wrong… Right ?

56

@Svan that is normal if using AFU and not using all the proper switches for what you want flashed, with AFU You must be specific (unlike FPT - with FPT region = entire region byte for byte)

To program BIOS with AFU and not using /GAN >> AFU /P /B /N /K

/B = BootBlock - This often contains the uCodes in AMI BIOS, but not always
/N - NVRAM
/K - Non-Critical blocks (K* where * =number) - This like NVRAM sort of, sometimes the keyboard backlight control can be in NVRAM or K2 - Sometimes this may need removed from above command line, if it gives you error

You can see MAC, UUID Serial etc with UEFITool NE (right click module, view as hex) or if it’s not in a compressed module you can view entire BIOS via hex and find it.
All depends on where it’s located. In this instance it’s in a non-compressed padding file module, so you can view either way directly.
It’s all in that same block/area as MAC, so you’re looking right at all the info when you see the MAC.

Why would you want old microcodes? Yes, I can do, but it’s a huge pain for this BIOS, I’ll probably have to rebuild 20+ BIOS again and may not find the correct way to do it safely, remember the space is limited and you added larger files already and I removed microcodes to accomodate.
If this way easy to mod BIOS = no problem, could do in one quick edit.

No, I do not think, nor suggest AFU used at all here, so I’d never suggest anything AFU is mandatory. I’m unsure what you mean by that.


@EthanWalker - I hate when Asus does that! Looks like you walked around it properly though, so beer your way too
However, I am not 100% certain WinFlash flashes all things properly either. Since BIOS is messed up by AFU already, EZ Flash may not have either (Since it leaves your NVRAM in place and other things)

You can dump BIOS any time you want with your programmer, just be sure to uncheck Erase/blank in the “Auto” Function before you connect to the chip.
This only needs done once on each version, then it will be like that from default. This way nothing gets erased until you mean to erase it.

Reflashing stock BIOS doesn’t help anything, in regards to your NVRAM. This was broken by AFU, so first needs fixed by AUF (AFU /N /GAN)
Then all will be fixed, at least in regards to your keyboard. This does not mean all is fixed in regards to BIOS, that would only be fixed by EZ Flash, whatever AFU is messing up I think.
Fix your stuff with AFU /N /GAN and then dump the BIOS with programmer

After this, I have no idea what your goal with all this process is anymore, so you’ll need to explain and then wait more.
This applies to your both in a way All this unnecessary flashing is unnecessary I know you had to fix stuff once initial /GAN messed it up, but redoing with /N /GAN fixes this, so after that I am confused why anything further is being attempted.
I know you wanted to play with your programmer EthanWalker, and I know we discussed trying to fix what AFU messed up via EZ Flash, but that didn’t work and you carried on with yet another non-ideal flash method too.
Best thing to do for now is fix it via AFU, get backup with your programmer and let me check for you to confirm it’s OK backup so you know what version software works good and you learn how to use etc.
Then since EZ flash doesn’t work, either wait for next BIOS or we can change dates I suppose (more unnecessary editing for no real reason)

57

@Lost_N_BIOS

Ok don’t do nothing if it’s too complicate. Now I understand how you got the space required for the PCIe NVMe. So the big question where are the microcodes in the bios ?
I will try /B and /K since I restored NVRAM already and flashed main bios image.

(For AFU i was just talking about the backup instead of the downloaded original bios from ASUS to mod it. But it’s not important now )

Greetings.

Update : ok fixed that was /B (microcodes are in boot block) ---- /K worked but got bugs in bios settings (restore defaults settings doing nothing and missng options in menus). Restored from original backup.

To resume flash modded bios (modded from original bios from ASUS) : AFUDOS modbios.bin /P /B

P = Program Main bios image (For all main modifications in bios mod)
B = Program Boot block (For Microcodes updates in bios mod)

This file is only for G751JY with CPU I7-4710HQ !!!

58



And this is the result:


I think I will go for Afu Gan and flash the last mod you sent me. This programmer is driving me crazy. I was afraid to brick the board till now… but now I can’t even connect this thing. Found a way to keep it Powered but the software says it is not connected, device manager also does not see it. Tried from 1.17 version to 1.31. None of them detects this piece.

59

@Svan - nothing is too complicated for me, in regard to this BIOS, I was only saying “it is complicated” and then further edits for no reason were for no real reason.
Microcodes are stored in two volumes in the BIOS.

@EthanWalker - sorry, that image explains nothing to me, except maybe you have a short somewhere or that USB port is faulty/loose etc. I have no clue, you don’t say what’s going on? You aren’t trying to use the same system to connect the programmer to that you are trying to dump/reprogram are you?
It’s ard to tell anything from a single limited image. In case you are, you can’t do that, two systems are needed here, one working system to run programmer on and then the board you are recovering.
Try some other system to program with, or use another USB Port. Or, it’s possible that you have something connected backwards too? Don’t connect the cable until you have the programmer plugged into the USB Port and are ready to do some dumping or flashing.
If it always cuts out like that, once cable is connected to board but not before, then you have something connected wrong. Show me images of the chip on board with nothing connected to it, so I can see where pin1 is, then show me image of your cable connected to it, and show me a few images from different angles of how you have the cable connected to the programmer.

Device manager has nothing to do with any of this. LED will remain light up on the programmer, unless you have a short.

60

@Lost_N_BIOS Solved with another programmer that works great.
Mine is defected.

biosdumb.zip (139 KB)

61

@EthanWalker - please explain what’s going on, what happened, what did you do etc? Or did you just mean your programmer itself was a bad sample (DOA, shorted on arrival etc)?

62



The programmer was defected. I cannot say what it had but the Power LED was not working/lightning when fully plugged into Usb…it showed up red only when half way plugged
but even so, it was not detected by the PC ( and no, it wasn’t the same system )
I asked the Amazon for a replacement.

I found another programmer in a local repair shop, it wasn’t for sale, I borrowed it.

When I connected it, the PC reacted with sound and notification, as if connecting any other device.
The C341A software also detected it (connected status on down-right-corner).


I did everything as instructed:
Unchecked the Erase and Blank options before connecting the programmer. Restarted the software and then I opened again the Settings, to verify and to be sure that those two options were unchecked. Exit the software.
Removed the CMOS battery on the patient motherboard.
Connected the clip to Bios chip ( detected in software as Winbond 25X40C )
Connected the programmer to Usb ( detected by the Windows 7 and by the CH341A_c.exe software )
Read.
Verified.
Saved.
Opened in the Hex editor - No FF’s.

*made saves from different versions of Ch341 software (1.17; 1.18; 1.29; 1.34) then opened them in HxD -> and did a Data comparison analysis: all of the saves/*bin are identical.

Thanks for your time and patience.

biosdumb.zip (139 KB)

63

Sounds like a bad short inside the USB connector, hopefully they replace for you without too much hassle. Cool that someone let you borrow a programmer, that’s nice of them!

Looks like a good dump to me, probably… However, this is not the BIOS chip, BIOS will be 8MB and this is 512KB dump. When you mentioned the chip ID I thought that looked odd, but I didn’t look it up, until I checked your attached file
Ahh yes, this is your vBIOS, see image below Then, try again, this time with correct chip, it should have 64 in the chip ID

Dump=vBIOS.png

64


Ouch, shame on me

Found it.



So, I couldn’t connect. Spent almost 2 hours doing this…no success.

● If I connect the programmer to the USB with NO clip attached to the chip - the PC detects the programmer and it stays connected.

● When connecting the programmer with the clip attached to the Bios chip - the Pc detects the programmer (usb connected sound), and then it disconnects the programmer (like it was ejected from the usb)
The Run LED is flashing strangely and the programmer’s usb plug is getting hot. Short circuit symptoms.

If I take the clip from the Bios chip - the PC detects the programmer again and it keeps connected.
With the clip attached to the chip, the programmer keeps disconnecting from the PC.

I verified the clip, all the pins are Ok, I verified the clip when attached to the chip, all the pins are in place on the Bios chip.
Verified if the clip is connected correctly, with the red wire on the dot side on chip / pin 1 on the programmer.

I even tried twice to connect to the vBios chip (just to check), it works, the software detects the chip - everything is ok.

65

Yes, that’s more like it For W25Q64FV use software 1.30 or 1.34 and choose ID W25Q64BV

Yes, hot means something is backwards! Double check your connections again. Make sure you are connecting cable correctly to chip, red wire top pin1 on chip (bottom left corner) and be sure red wire goes into programmer in the middle slot, red wire on lever side.
If you are unsure, please upload image of how you are connecting to the chip, and a few shots of how you have the cable in the programmer.

Make sure you are not accidentally touching any of those tiny resistors by the BIOS chip. The two black 0402 above Winbond look pretty close to the legs there, so may be accidental contact. Tape them off with electric or polyimide/kapton tape if you have, then we can rule that out
Also, remove CMOS and main battery, and main power supply if you did not already, then try again. Then test adding each of those back one at a time, sometimes all must be removed, sometimes all must be connected.
However, this may have nothing to do with the above, since both the symptoms you mention (hot and disconnect) = short somewhere

In general, I have USB programmer plugged in already, and then connect the cable to the chip once I am ready (often many times until it connects properly). I’m not sure if any of that matters, but sounds like you doing it different than I do, so mentioning in case it helps.
I have shorted chips this way too, where they get hot like you mention, not kill the chips or anything I just mean in general when connecting the clip it may look correct and be correct, but still short. remove it and re-clup carefully again.
Usually I’d say out of 100 clips, I get the following made up example of results >> 50% = no chip detected, re-clip / 10% = Short / 40% = Success after many attempts of previous short or no detection
I find it best to hold the clamp down to the board, spread open already, so that the plastic is setting on/against the PCB, then slowly release pressure so it clamps to chip. This may not be possible on your setup due to those 0402 resistors, tape those off I bet that is where it’s shorting

66

@Lost_N_BIOS

The 1st programmer I had - has been returned to Amazon due to the defected plug.
The 2nd I returned to the local service center (I kept the SOIC8 clip)
Now I bought another programmer (without clip because I have one). The programmer works with no problems: no half plugging, no weird flashing leds etc.

Anyway I couldn’t get the job done. The same problem, the same symptoms.
:black_circle: The notebook has no battery, I removed it 2 years ago.
:black_circle: The CMOS battery has been removed since the first test with the programmer.
:black_circle: I tried to connect the clip in many ways.
:black_circle: Atached the clip THEN connected the programmer / Connected the programmer THEN attached the clip in various positions.
:black_circle: Plugged the programmer into 1 notebook then I tried on a desktop.
:black_circle: My paranoia told me that I should try to plug the programmer into a USB 2.0 instead of 3.0 and 3.1, so I did that.
:black_circle: Applied kapton tape on the resistors.
:black_circle: Tried not to breathe during the working process.
:black_circle: And the final method that IMHO had to be done from the beginning, tried to attach the clip looking through magnifying glass and verifying each pin on the chip/clip.

Nothing worked for me. It makes me sad because now, the only way is to take off the chip from the board, which I don’t really want to do.

Sorry for the quality, shaking hands… It is not so easy to hold the phone and attach the clip at the same time