Chkcpu wrote on 2025-03-31, 14:17:

Hi analog_programmer, […]
Show full quote

Hi analog_programmer,

Looking at the steps you used to get the patched “original.tmp” back into the BIOS BIN file, I see how this would have worked as well. Two remarks though about your procedure:

1) Using CBROM to strip down the BIOS BIN-file to contain only the bootblock and the compressed “original.tmp” module including its header, and nothing behind, is a nice alternative method for copying the compressed patched “original.tmp” over the old one in the BIN-file. No worries about shifting the compressed modules behind original.tmp to make room, and calculating how many bytes this shift must be.
Of course, just as you describe in steps 2 and 7, you have to extract and save each module before you release it from the BIN, and put them back one by one afterwards.
A nice side-effect is that each time you use CBROM to store one of the previously released modules back in the BIN-file, the checksum bytes of the BIN are recalculated. So using MODBIN to edit the sign-on message and update the BIN to get the checksums correct, as I do as the last step, isn’t necessary in your procedure.
But checking the BIN with MODBIN as a last step is always a good idea of course!

Note that this “strip down” procedure is only necessary on 128KB compressed BIOSes. Although CBROM lists all modules as being stored behind original.tmp on both 128KB and 256KB BIOS BIN-files, this is actually so on 128KB BIOSes only.
On the 256KB BIOS, original.tmp is located separately at file-offset 20000h with nothing directly behind it. All the other modules are located at offsets 0-1FFFFh and are not in the way when overwriting the compressed original.tmp module by the patched version.

2) You don’t describe editing the header of the patched original.tmp module after compression. It is very well possible that CBROM takes care of that when putting the previously released modules back into the BIN. I never tested this and edit the header of the compressed patched original.tmp before I put it back into the BIN-file, just to be sure. 😉

Happy patching!
Jan

Hi,
I'm not 100% sure about these steps, I'm just trying to recreate by memories and the notes in my thread what I've done when patched those alien BIOSes to work on my mislabeled BX-6AV2. To clarify what is the working algorithm I have to try these steps or some similar variants.

I'm sure I was not bothered to change any header for the manually compressed patched module "original.tmp" when pasting it into the "stripped" BIOS dump file. I just found the same matching starting bytes in the "stripped" BIOS file which still contained the original compressed "original.tmp" module as in the beginning of the new compressed "original.tmp" module and pasted over from the first matching byte in the "stripped" BIOS while looking at the "end area" for some leftovers from the original compressed "original.tmp" module (I think there were some which I changed to "FF" HEX-values). The BIOS for BX-6AV2 board is 256 kB in size. I decided that CBROM.EXE will take care of correct placing for all the additional modules when putting them back into the "stripped" BIOS with patched and compressed "original.tmp" in their original order and CBROM get this job done correctly. From your explanation now I see that this BIOS "stripping" procedure was unnecessary in my particular case with these 256 kB BIOS experiments.

CBROM.EXE recalculates and writes new BIOS checksum when changes some modules, but I don't know if it verifies the BIOS checksum when opens a BIOS dump file.

Hey, I got Abit AB-AX5 board, and I was wondering if you have patched BIOS for it. On your page you stated that patching was unsuccessful because of unstable 2.0 .. 2.2 voltages.
I ordered K6-2 450 AHX which according to AMD specs should run at 2.3 to 2.5V core, so I think it could be fine with the lowest official voltage on AX5 (2.5V).
What do you think? Do you still have patched BIOS with more multipliers somewhere by any chance?

Hi 0xDEADBEEF,

Yes, I have been working again on the patched AB-AX5/PX5 BIOS.
Last year I successfully patched several BIOSes for the jumperless QDI Titanium B series, and that stimulated me to look at the jumperless Abit AX5/PX5 BIOS again. 😉

Recently I found something interesting that may explain the unstable Vcore on K6-2/350 and faster AMD models. I was testing my latest patched AX5/PX5 BIOS on a PX5 board with a K6-2/400 at 2.2V Vcore. The new low Vcore logic worked fine but anytime I did a reboot after setting the CPU SOFT Menu to a CPU Speed above 300MHz, the Vcore jumped up to 3.2V! Ouch!

Earlier, I already found that this jumpy Vcore was accompanied by a reset to Default values in the CPU SOFT MENU, and a “CPU is unworkable or has been changed.” Message on the POST screen.
But the recent discovery was that the jumpy Vcore and CPU is unworkable message could be avoided by setting the CPU Operating Speed to User Define and the Speed Error Hold option to Disabled.
Now my K6-2/400 kept running fine at 400MHz and 2.2V Vcore, even after several warm or cold reboots. 😀

So it appears that the unstable Vcore is caused by some logic in the BIOS, and is not a hardware but a software issue, and that can be fixed!!
I’m now looking into this “Speed Error Hold” logic, to see how that can be changed.
I will send you a patched BIOS to test when I have a working fix for this issue.

Cheers, Jan

Sweet!
I have seen “CPU is unworkable or has been changed.” message too, happens when I switch to different CPU. I think there is some "check if user is doing something too stupid" logic in BIOS that prints it 😉
I am excited to hear you are working on it.
This board is pretty cool, 4 pci + 4 isa, built in usb and ATX form factor that makes it compatible with modern ATX cases.

Is there any hope of the QDI P5I430TX-250 (Titanium I) receiving a patched bios like this? Seems very similar to the Titanium IB.

Chkcpu wrote on 2025-03-30, 12:36:

About CBROM V1.30C, I selected this 1999 version because it worked best on all 1997/1998/1999 socket 5/7 Award BIOSes. For 2000+ and Slot 1 BIOSes I do use a later CBROM version, usually V2.07.

I found CBROM version 1.30 (seems like it's slightly older than your attached version 1.30C) and currently I have to deal with an 128 kB Award BIOS ver. 4.50 from 1997 for a socket 3 mobo (this is not a mistake). It seems like this particular BIOS is not much compatible with CBROM 1.30 or 1.30C versions from 1999. Are there any other (older) suitable CBROM versions, which I can try?

Also I have MODBIN versions 4.50.36 from 1993 and 4.50.80C from 1999, but nothing in between them. Which versions of MODBIN should I look for (for the time period 1994 - 1998)?

Hi Chkcpu ,
I have a Compaq Camaro motherboard (MVP4) with the compaq bios. It works very well, but it's lacking in options. ¿Is it posible to unlock options such as disable com and lpt ports, disable onboard audio, enable or disable external cache, assign irq to pci devices?
link to retroweb :
https://theretroweb.com/motherboards/s/mitac- … igon-pwa-camaro

I'm not using the Mitac bios because in my tests the compaq is more stable and can achieve better oveclocking results.

sizzlinbeef wrote on 2025-04-14, 19:30:

Is there any hope of the QDI P5I430TX-250 (Titanium I) receiving a patched bios like this? Seems very similar to the Titanium IB.

Hi sizzlinbeef,

The P5I430TX-250 board comes in both a SpeedEasy and a non-SpeedEasy version. I understand you have the SpeedEasy version without the JC, JV, and JS jumpers?

I don’t know if these early Titanium I boards use a DC-DC Converter IC like the Raytheon RC5041M to control the Vcore voltage, or if a discrete circuit with resistors is used.
Can you make a high-res photo of your motherboard so I can see what kind of circuit is used?

These earlier boards often use linear regulators with discrete components to control the voltage to the CPU. If this is the case on your board, then only a hardware mod of an interposer can get you Vcore voltages below 2.8V.

Cheers, Jan

analog_programmer wrote on 2025-04-14, 23:27:

Chkcpu wrote on 2025-03-30, 12:36:

About CBROM V1.30C, I selected this 1999 version because it worked best on all 1997/1998/1999 socket 5/7 Award BIOSes. For 2000+ and Slot 1 BIOSes I do use a later CBROM version, usually V2.07.

I found CBROM version 1.30 (seems like it's slightly older than your attached version 1.30C) and currently I have to deal with an 128 kB Award BIOS ver. 4.50 from 1997 for a socket 3 mobo (this is not a mistake). It seems like this particular BIOS is not much compatible with CBROM 1.30 or 1.30C versions from 1999. Are there any other (older) suitable CBROM versions, which I can try?

Also I have MODBIN versions 4.50.36 from 1993 and 4.50.80C from 1999, but nothing in between them. Which versions of MODBIN should I look for (for the time period 1994 - 1998)?

Hi analog_programmer,

These socket 3 Award v4.50(G) BIOSes are usually uncompressed, so CBROM won’t work on these. But you write that it is a 128KB BIOS from 1997 and this puts this BIOS into the compressed BIOS era.
Can you upload a dump of this BIOS here, so I can take a look at this enigma? 😉

About MODBIN v4.50.80a, I’ve used it successfully on all uncompressed and compressed Award BIOSes from the 1994-1999 time period.

Jan

Chkcpu wrote on 2025-04-15, 14:23:

Hi sizzlinbeef, […]
Show full quote

Hi sizzlinbeef,

The P5I430TX-250 board comes in both a SpeedEasy and a non-SpeedEasy version. I understand you have the SpeedEasy version without the JC, JV, and JS jumpers?

I don’t know if these early Titanium I boards use a DC-DC Converter IC like the Raytheon RC5041M to control the Vcore voltage, or if a discrete circuit with resistors is used.
Can you make a high-res photo of your motherboard so I can see what kind of circuit is used?

These earlier boards often use linear regulators with discrete components to control the voltage to the CPU. If this is the case on your board, then only a hardware mod of an interposer can get you Vcore voltages below 2.8V.

Cheers, Jan

Yes, it's the SpeedEasy version.

Links didn't seem to work, pictures are attached here.

another

Thanks for the clear pictures!

First thing to notice are the two power transistors TIP120 and TIP102 with a big heatsink, and the absence of inductor coils. This means that this board has a linear power regulator circuit for the CPU, instead of the more efficient switching power regulator as used in later socket 7 boards, like the QDI Titanium B series.
Handling power for a Pentium-MMX/233 is usually the upper limit for these linear regulators.

Secondly, comparing your pictures with the Titanium I picture of the non-SpeedEasy version on TRW, I notice that the ICs U14, U22, and U13 are now present. Also 3 transistors, near the heatsinks of the two power transistors are added on your board.
Clearly these components are for the BIOS SpeedEasy control of the 11 missing jumpers!
U14 and U22 are DM7406’s and house 6 open collector buffers each, probably used as electronic switches to replace the jumpers.
U13 is an LS174 with 6 D flip-flops to latch the data for the electronic switches. But I see no DC-DC converter, normally associated with switching power regulators, for wider control of the CPU’s Core Voltage.

I was hoping to see a switching mode power regulator on this Titanium I SpeedEasy version, but due to the hardware constrains of the linear regulator circuit, modding the Titanium I BIOS for additional Vcore support is unfortunately not possible.

Regards, Jan

Chkcpu wrote on 2025-04-15, 14:39:

Hi analog_programmer, […]
Show full quote

Hi analog_programmer,

These socket 3 Award v4.50(G) BIOSes are usually uncompressed, so CBROM won’t work on these. But you write that it is a 128KB BIOS from 1997 and this puts this BIOS into the compressed BIOS era.
Can you upload a dump of this BIOS here, so I can take a look at this enigma? 😉

About MODBIN v4.50.80a, I’ve used it successfully on all uncompressed and compressed Award BIOSes from the 1994-1999 time period.

Jan

Hi, Jan!

The BIOS is for FIC 486-GVT-2 motherboard and I got it from TRW site - "beta BIOS (Y2K fix) + VGA BIOS" after I read the thread with the topic about the BIOS updates for this very same motherboard. As you can see, this 128 kB Award BIOS ver. 4.50 contains optional VGA ROM which is redundant for the FIC 486-GVT-2 motherboard and our fellow forum member scorp (Necroware's YT channel owner) just cut the 128 kB "3276gn1.bin" BIOS dump file in half leaving only the second part with the "core BIOS". But there are some hardware problems as seen in the comments left from other users who have tried it.

So I decided to try something else - to remove the redundant optional VGA ROM by using CBROM.EXE. The only problem is that all of the oldest CBROM versions that I have (1.30, 1.30C and 1.32) "see" this optional VGA ROM as "other" instead of "vga". Here is the example with CBROM ver. 1.30:

Anyway, I proceeded with option VGA ROM removal by CBROM:

And the final result is this:

I also reproduced the same procedure using CBROM versions 1.30C and 1.32 and they gave me exactly the same binary matching BIOS dump files.

I've opened the BIOS the MODBIN ver. 4.50.63A (from 1997) and everything is OK, except that I noticed one redundant setting "Onboard C000h video ROM scan: Enable" which I changed to "Disabled" (as seen in other Award 4.50 BIOSes for motherboards without integrated video):

I have a spare 128 kB Winbond W29EE011 chip, so I hope I'll find time during the weekend to test my modified version. I'm attaching my modified 128 kB BIOS version of "3276gn1.bin" with removed optional VGA ROM in a zip-file, if you want to see what could be possibly wrong with it. Mind that I also made some minor BIOS menu settings tweaks and changed the BIOS message string:

All the best!

Chkcpu wrote on 2025-04-16, 17:46:

I was hoping to see a switching mode power regulator on this Titanium I SpeedEasy version, but due to the hardware constrains of the linear regulator circuit, modding the Titanium I BIOS for additional Vcore support is unfortunately not possible.

Regards, Jan

Well thank you regardless for taking the time and putting this to rest.

I haven’t gotten this one yet, but I think I jmay have ust joined the soft config crowd with this one? Hard to see in grainy picture.

Sis 540 motherboard project

I’m curious what your thoughts on the bios for this board is Jan.

Most interested in more clock settings in bios. And your patches of course 😀

MarmotaArmy wrote on 2025-04-15, 12:57:

Hi Chkcpu , I have a Compaq Camaro motherboard (MVP4) with the compaq bios. It works very well, but it's lacking in options. ¿ […]
Show full quote

Hi Chkcpu ,
I have a Compaq Camaro motherboard (MVP4) with the compaq bios. It works very well, but it's lacking in options. ¿Is it posible to unlock options such as disable com and lpt ports, disable onboard audio, enable or disable external cache, assign irq to pci devices?
link to retroweb :
https://theretroweb.com/motherboards/s/mitac- … igon-pwa-camaro

I'm not using the Mitac bios because in my tests the compaq is more stable and can achieve better oveclocking results.

Hi MarmotaArmy,

I have no knowledge of the Compaq BIOS and I don’t know how to decompress or decrypt such a BIOS. So I’m unable to see if it has hidden options that can be unlocked.

However the MiTAC Award v6.00PG BIOS for this nice Camaro board has all the bells and whistles to tune the MVP4 chipset. And it has K6-2+/III+ support as well.
I believe you should be able to reach the same stability as with the Compaq BIOS, but you will need to experiment with all these BIOS options. 😉
Start with the Setup Defaults and work from there. The Award BIOS usually uses optimal timings, so you may have to relax the memory timing a notch to reach the same settings that the Compaq BIOS uses, especially when overclocking.

Happy tinkering!
Jan

analog_programmer wrote on 2025-04-16, 22:03:

Hi, Jan! […]
Show full quote

Chkcpu wrote on 2025-04-15, 14:39:

Hi analog_programmer, […]
Show full quote

Hi analog_programmer,

These socket 3 Award v4.50(G) BIOSes are usually uncompressed, so CBROM won’t work on these. But you write that it is a 128KB BIOS from 1997 and this puts this BIOS into the compressed BIOS era.
Can you upload a dump of this BIOS here, so I can take a look at this enigma? 😉

About MODBIN v4.50.80a, I’ve used it successfully on all uncompressed and compressed Award BIOSes from the 1994-1999 time period.

Jan

Hi, Jan!

The BIOS is for FIC 486-GVT-2 motherboard and I got it from TRW site - "beta BIOS (Y2K fix) + VGA BIOS" after I read the thread with the topic about the BIOS updates for this very same motherboard. As you can see, this 128 kB Award BIOS ver. 4.50 contains optional VGA ROM which is redundant for the FIC 486-GVT-2 motherboard and our fellow forum member scorp (Necroware's YT channel owner) just cut the 128 kB "3276gn1.bin" BIOS dump file in half leaving only the second part with the "core BIOS". But there are some hardware problems as seen in the comments left from other users who have tried it.

So I decided to try something else - to remove the redundant optional VGA ROM by using CBROM.EXE. The only problem is that all of the oldest CBROM versions that I have (1.30, 1.30C and 1.32) "see" this optional VGA ROM as "other" instead of "vga". Here is the example with CBROM ver. 1.30:

The attachment CBROM130_option_VGA_ROM.jpg is no longer available

Anyway, I proceeded with option VGA ROM removal by CBROM:

The attachment CBROM130_option_VGA_ROM_removal.jpg is no longer available

And the final result is this:

The attachment CBROM130_option_VGA_ROM_removed.jpg is no longer available

I also reproduced the same procedure using CBROM versions 1.30C and 1.32 and they gave me exactly the same binary matching BIOS dump files.

I've opened the BIOS the MODBIN ver. 4.50.63A (from 1997) and everything is OK, except that I noticed one redundant setting "Onboard C000h video ROM scan: Enable" which I changed to "Disabled" (as seen in other Award 4.50 BIOSes for motherboards without integrated video):

The attachment MODBIN.jpg is no longer available

I have a spare 128 kB Winbond W29EE011 chip, so I hope I'll find time during the weekend to test my modified version. I'm attaching my modified 128 kB BIOS version of "3276gn1.bin" with removed optional VGA ROM in a zip-file, if you want to see what could be possibly wrong with it. Mind that I also made some minor BIOS menu settings tweaks and changed the BIOS message string:

The attachment 3276gn1_M-mynovga.zip is no longer available

All the best!

Hi analog_programmer,

I’ve looked into these 486-GVT-2 BIOSes, and found that FIC was up to its usual tricks again. 😉
With many of their Socket 7 BIOSes, they often made a small (Y2K) fix, years after the last BIOS update and slashed a 1999 date on it without updating the BIOS core to the 1999 standard. So these “BIOS updates” were in essence still a 1996/1997 BIOS with the limitations of that time.

I found they did the same with this “1997” 3.276GN1 BIOS. This BIOS is internally still an Award 4.50G 1994/1995 BIOS, and is completely uncompressed.
That’s why cutting off the first half of this 128KB BIOS file, that only contains the VGA ROM, worked without problems on the 486-GVT-2 board, as scorp demonstrated.

As the CBROM tool only works on the compressed Award BIOS, it can’t be used on this uncompressed 3.276GN1 BIOS. That’s why you saw a very strange CBROM ouput when you attempted to release the VGA ROM.
Because the BIOS is uncompressed, there is no LZH header to be found and it is funny to see that CBROM tried to make something of the beginning of the file and indicated the “IBM VGA Compatible” string as original filename of this “other” component. 😀

When comparing your 3276gn1_M-mynovga.bin with the original 3276GN1.BIN, I found that the first 32KB containing the VGA ROM was completely scrambled in your modded BIOS. Luckily the rest of the BIOS was not touched by CBROM, when you attempted to release the VGA ROM.

So your “mynovga” BIOS will probably work normally, however the VGA ROM is not in the way and you only had to Disable the "Onboard E000h video ROM scan:” in MODBIN.
This will effectively cause the onboard VGA ROM to be never called, so it can’t interfere with the VGA BIOS on the graphics card.
Note that the "Onboard C000h video ROM scan:” setting in MODBIN can be left Enabled.

Happy Easter,
Jan

Chkcpu wrote on 2025-04-20, 18:35:

Hi analog_programmer, […]
Show full quote

Hi analog_programmer,

I’ve looked into these 486-GVT-2 BIOSes, and found that FIC was up to its usual tricks again. 😉
With many of their Socket 7 BIOSes, they often made a small (Y2K) fix, years after the last BIOS update and slashed a 1999 date on it without updating the BIOS core to the 1999 standard. So these “BIOS updates” were in essence still a 1996/1997 BIOS with the limitations of that time.

I found they did the same with this “1997” 3.276GN1 BIOS. This BIOS is internally still an Award 4.50G 1994/1995 BIOS, and is completely uncompressed.
That’s why cutting off the first half of this 128KB BIOS file, that only contains the VGA ROM, worked without problems on the 486-GVT-2 board, as scorp demonstrated.

As the CBROM tool only works on the compressed Award BIOS, it can’t be used on this uncompressed 3.276GN1 BIOS. That’s why you saw a very strange CBROM ouput when you attempted to release the VGA ROM.
Because the BIOS is uncompressed, there is no LZH header to be found and it is funny to see that CBROM tried to make something of the beginning of the file and indicated the “IBM VGA Compatible” string as original filename of this “other” component. 😀

When comparing your 3276gn1_M-mynovga.bin with the original 3276GN1.BIN, I found that the first 32KB containing the VGA ROM was completely scrambled in your modded BIOS. Luckily the rest of the BIOS was not touched by CBROM, when you attempted to release the VGA ROM.

So your “mynovga” BIOS will probably work normally, however the VGA ROM is not in the way and you only had to Disable the "Onboard E000h video ROM scan:” in MODBIN.
This will effectively cause the onboard VGA ROM to be never called, so it can’t interfere with the VGA BIOS on the graphics card.
Note that the "Onboard C000h video ROM scan:” setting in MODBIN can be left Enabled.

Happy Easter,
Jan

Hi! And thank you very much for your detailed analysis of this Award BIOS, Jan!

I suspected that something's wrong with all the "garbage" given by CBROM when reading the contents of the BIOS (CBROM /D). But for the other "ballast" left when VGA option ROM was removed by CBROM I thought - maybe this is some useful code from another option ROM left. Also I had doubt, that these old versions of CBROM are the main culprit for the "garbage" displayed (because of this now I have a pretty big collection of CBROM and MODBIN versions 😀 ).

Oh, my goodness... These FIC's tricks with BIOS core versions are sick! Now I'm wondering why FIC's motherboards are so highly valuable, if this company had such a problems with their BIOS updates, with the information in their user manuals (for example the manual for this mobo lacks any settings for other CPUs except for intel and Cyrix 486) and their technical support in general. Ok, the quality of their production seems to be above the average level, but this doesn't justify their high reputation in terms of all the other problems caused by their bad technical support.

So, basically scorp's 64 kB BIOS no-VGA BIOS is Ok regardless of "Onboard C000h video ROM scan: Enable" setting is left untouched. It turns out that this is just a very bad beta BIOS update for the 486-GVT-2 board (maybe it works more stable for some of the other models boards from compatibility list given). I'll update my 486-GVT-2 to BIOS version 3.07G if I manage to dig up some 64 kB compatible BIOS chip as it currently has some OTP-ROM with BIOS version 3.03G. Or maybe I'll left this board as is and get rid of it, 'cause I really don't like it for some unknown reason (just a feeling) from the moment I got it in my hands.

You saved me a lot of lost time for tinkering with this FIC 486-GVT-2 BIOS and motherboard, Jan!

P.S. Unfortunately I'm now restricted from post editing by some forum's code for overusage of "Edit" button which is not a cool thing and I can't edit my previous post with the messed BIOS attachment. I still have another option to remove the messed BIOS zip-file from the forum's DB, so I'll just remove it without any editing in the post with explanation for the reasons why I'm removing the broken BIOS attachment.

analog_programmer wrote on 2025-04-21, 00:35:

Oh, my goodness... These FIC's tricks with BIOS core versions are sick! Now I'm wondering why FIC's motherboards are so highly valuable, if this company had such a problems with their BIOS updates, with the information in their user manuals (for example the manual for this mobo lacks any settings for other CPUs except for intel and Cyrix 486) and their technical support in general. Ok, the quality of their production seems to be above the average level, but this doesn't justify their high reputation in terms of all the other problems caused by their bad technical support.

Because FIC made very good motherboards, generally very stable, and back then (early -mid 1990's) they would respond to support emails.
added: the 06/16/97-VT82C486A-214L2000-00 BIOS (Y2K fix) + VGA BIOS is not from a real 486-GVT-2, is for the 486-GAC-V which due to same chipset TRW decided to include it in the 486-GVT-2.
There never was any video on a 486-GVT-2, not sure why they added it for it without some major notations. Also the board has no VRM so no 3.3v cpu's DX4's, back then Intel and AMD DX and DX-2 used same settings iirc