DenizOezmen wrote on 2023-02-06, 20:54:

Thank you, that is interesting! I've taken a quick look. On a first glance at the 0102 BIOS, I can't find any positions where th […]
Show full quote

i386 wrote on 2023-02-05, 14:10:

Yes, I remember! IIRC, ASUS BIOS for P2L97 lx2i0102.zip worked with PPRO, no shutdown problem. https://www.elhvb.com/mobokive/A […]
Show full quote

Yes, I remember! IIRC, ASUS BIOS for P2L97 lx2i0102.zip worked with PPRO, no shutdown problem.
https://www.elhvb.com/mobokive/Archive/Asus/s … 2l97/index.html
More modern BIOS-es for this board - no PPRo support. But I did not modified intermediate
P2L97 BIOS-es for PPRo support and I do not know about APWC problem in them. Anyhow,
latest BETA BIOS for P2L97 (from official suite) definitely has the APWC problem.

Thank you, that is interesting! I've taken a quick look. On a first glance at the 0102 BIOS, I can't find any positions where the code actually reads anything from port 0xB2. Maybe the handlers are still missing there?
In any case, the latest P2L97 BIOS has fewer handlers than the P3B-F. I wonder whether the generation of SCIs (which are ACPI-specific as far as I understand) might be a problem.

I've attached a test build that (aside from removing the ACPI table) disables two handlers that might be responsible for switching between SCI and SMI generation.

maxtherabbit wrote on 2023-01-28, 14:55:

I suspect you are correct about the shutdown problem, I will try both of these test versions and get back.

@maxtherabbit: Could you test this one if you find the time?

i386 wrote on 2023-02-05, 14:10:

PS. Currently I'm repairing the ASUS P2L97-DS after severe damage.

Oh, a dual board. What happened to it?

Yes of course. Unfortunately I haven't had a chance to test the previous test build you posted either. Would you still like them tested separately or just go for the newer one?

maxtherabbit wrote on 2023-02-06, 21:29:

Yes of course. Unfortunately I haven't had a chance to test the previous test build you posted either. Would you still like them tested separately or just go for the newer one?

Thank you, no hurry!

Tests of both would be great. I'm not yet sure what to expect.

DenizOezmen wrote on 2023-02-06, 20:54:

Thank you, that is interesting! I've taken a quick look. On a first glance at the 0102 BIOS, I can't find any positions where the code actually reads anything from port 0xB2. Maybe the handlers are still missing there?

Yes, very probably that no the port 0xB2 reading in BIOS code, since that BIOS very old and raw.

DenizOezmen wrote on 2023-02-06, 20:54:

I've attached a test build that (aside from removing the ACPI table) disables two handlers that might be responsible for switching between SCI and SMI generation.

It is very interesting!! Unfortunately, I have no converter now, but I'll think about SMM/SCI stuff.

DenizOezmen wrote on 2023-02-06, 20:54:

Oh, a dual board. What happened to it?

Board just caught fire under the linear stab transistor, probably due overheating (maybe some
person installed VGA card with big current on 3.5V line). I've bought board with this damage already.

I found interlayer closure between +5V and GND line. Currently I've deleted closure, cleaned cinder
and repaired print ways over and under defect place. Both transistors of linear stabilizer I've deleted
and take 3.3V power from ATX connector. Board started!! But I also found bad capacitors on it (swelling
and leakage). Currently I'm doing full recapping my board. Later, I hope to show the photos of my board.

BTW, I remember one more potential PPRo HW incompatibility problem.
PPRo has CMOS signal level 3.3V, but Pentium2/3 have 2.5V CMOS levels.
Sometimes some PPRo->Slot1 converters have pull-up resistors to 3.3V for
CMOS signals. In some boards PPro may not starting (or stop during POST)
even with right BIOS. Deleting these resistors may to solve problem.

First, thanks for your work! I'm running your P3B-F bios with P3 Tualatin 1.4s, which I installed a couple of years ago (just when you released it) and never found time to properly thank you, so thank you very much! 😀

However, I just noticed this post:

DenizOezmen wrote on 2022-09-14, 12:41:

Babasha wrote on 2022-09-13, 08:17:

I need a info are there tools/methods to customize ASUS P3B-F BIOS with custom "OEM" logo or picture?

You can do two kinds of logo-related changes to the P3B-F BIOS: One possibility is to change the EPA logo in the upper right corner. The image format is severely restricted in terms of color choices, though. You can find a short tutorial e.g. here. There are several programs that can create the necessary EPA files, but not all of them produce compatible results. I verified that the EPACoder tool mentioned in that article works for the P3B-F. (Note: You will need to create a "Version 1" EPA file, contrary to what the article implies.) An example result can be seen in the first attachment.

The second possibility is adding a full-screen picture that replaces the POST screen. I wasn't aware of that until now (because it's not really documented), but the P3B-F also seems to support this feature. ASUS probably reserved it for certain mainboards, seeing that they only released their Logo tool for a few boards, but at least it worked on the P3B-F for one of the example images in the archive (see second attachment).

Could you kindly elaborate more on that? So a P3B-F usually don't have a full screen logo by default? Mine had it from the factory - IIRC it was all blue and reminiscent of that Windows 95 install picture with mouse and CDs on it. When I installed the modified BIOS, I was no longer seeing it. Is it possible to get that logo back? Or was it overwritten? It's not that I'm worried too much about it, but It'd be nice to have it back. 😀

I want to show the photos of my P2L97-DS, that I mentioned earlier

Before repairing

After repairing

I hope to run it with two CPU's.

i386 wrote on 2023-02-08, 16:43:

I want to show the photos my P2L97-DS, that I mentioned earlier

That hurts!

Zoomer wrote on 2023-02-07, 10:56:

First, thanks for your work! I'm running your P3B-F bios with P3 Tualatin 1.4s, which I installed a couple of years ago (just wh […]
Show full quote

First, thanks for your work! I'm running your P3B-F bios with P3 Tualatin 1.4s, which I installed a couple of years ago (just when you released it) and never found time to properly thank you, so thank you very much! 😀

However, I just noticed this post:

DenizOezmen wrote on 2022-09-14, 12:41:

Babasha wrote on 2022-09-13, 08:17:

I need a info are there tools/methods to customize ASUS P3B-F BIOS with custom "OEM" logo or picture?

You can do two kinds of logo-related changes to the P3B-F BIOS: One possibility is to change the EPA logo in the upper right corner. The image format is severely restricted in terms of color choices, though. You can find a short tutorial e.g. here. There are several programs that can create the necessary EPA files, but not all of them produce compatible results. I verified that the EPACoder tool mentioned in that article works for the P3B-F. (Note: You will need to create a "Version 1" EPA file, contrary to what the article implies.) An example result can be seen in the first attachment.

The second possibility is adding a full-screen picture that replaces the POST screen. I wasn't aware of that until now (because it's not really documented), but the P3B-F also seems to support this feature. ASUS probably reserved it for certain mainboards, seeing that they only released their Logo tool for a few boards, but at least it worked on the P3B-F for one of the example images in the archive (see second attachment).

Could you kindly elaborate more on that? So a P3B-F usually don't have a full screen logo by default? Mine had it from the factory - IIRC it was all blue and reminiscent of that Windows 95 install picture with mouse and CDs on it. When I installed the modified BIOS, I was no longer seeing it. Is it possible to get that logo back? Or was it overwritten? It's not that I'm worried too much about it, but It'd be nice to have it back. 😀

It was overwrited.

Babasha wrote on 2023-02-09, 05:50:

It was overwrited.

Thanks! In that case (if somebody's interested and it's THAT rare) I may try to find my backups and somehow extract the image so that it can be put back to the modded bios with that Asus utility.

DenizOezmen wrote on 2023-02-06, 20:54:

Thank you, that is interesting! I've taken a quick look. On a first glance at the 0102 BIOS, I can't find any positions where th […]
Show full quote

i386 wrote on 2023-02-05, 14:10:

Yes, I remember! IIRC, ASUS BIOS for P2L97 lx2i0102.zip worked with PPRO, no shutdown problem. https://www.elhvb.com/mobokive/A […]
Show full quote

Yes, I remember! IIRC, ASUS BIOS for P2L97 lx2i0102.zip worked with PPRO, no shutdown problem.
https://www.elhvb.com/mobokive/Archive/Asus/s … 2l97/index.html
More modern BIOS-es for this board - no PPRo support. But I did not modified intermediate
P2L97 BIOS-es for PPRo support and I do not know about APWC problem in them. Anyhow,
latest BETA BIOS for P2L97 (from official suite) definitely has the APWC problem.

Thank you, that is interesting! I've taken a quick look. On a first glance at the 0102 BIOS, I can't find any positions where the code actually reads anything from port 0xB2. Maybe the handlers are still missing there?
In any case, the latest P2L97 BIOS has fewer handlers than the P3B-F. I wonder whether the generation of SCIs (which are ACPI-specific as far as I understand) might be a problem.

I've attached a test build that (aside from removing the ACPI table) disables two handlers that might be responsible for switching between SCI and SMI generation.

maxtherabbit wrote on 2023-01-28, 14:55:

I suspect you are correct about the shutdown problem, I will try both of these test versions and get back.

@maxtherabbit: Could you test this one if you find the time?

i386 wrote on 2023-02-05, 14:10:

PS. Currently I'm repairing the ASUS P2L97-DS after severe damage.

Oh, a dual board. What happened to it?

Progress! This build allows Acronis True Image 2011 to boot! It locked up before, so I suspect gutting the ACPI tables was the trick

Still the same behaviour when running the shutdown utility in DOS however.

EDIT to add: Windows 2000 setup now launches too without having to manually change HAL

I apologize that I start with a post off topic.... I'm playing with an ASUS P3V133 trying to get a 16bit ISA controller to work(CA9342 no BIOS no PNP)
the disk attached is detected by the mainboard but cannot be accessed, I would like to understand if there is hope to make it work or if because of the PCI Bridged Isa it is not possible at all...
Could it be that mod firmware helps?
thanks if you can give me some guidance!

fumik wrote on 2023-02-16, 21:00:

I apologize that I start with a post off topic.... I'm playing with an ASUS P3V133 trying to get a 16bit ISA controller to work( […]
Show full quote

I apologize that I start with a post off topic.... I'm playing with an ASUS P3V133 trying to get a 16bit ISA controller to work(CA9342 no BIOS no PNP)
the disk attached is detected by the mainboard but cannot be accessed, I would like to understand if there is hope to make it work or if because of the PCI Bridged Isa it is not possible at all...
Could it be that mod firmware helps?
thanks if you can give me some guidance!

You need to disable the all internal HW in motherboard BIOS SETUP, that were enabled on you card and have
same HW resources, as internal - COM, LPT, FDC, IDE. Else you'll get conflict with internal and external controllers.
Unfortunately, not all board can turn off internal IDE completely. Often, only interrupt from controller will be disabled.
I've got very good result on FIC PA-2011 motherboard with 16-bit MFM controller (its registers model almost as ATA).

i386 wrote on 2023-02-17, 14:22:

You need to disable the all internal HW in motherboard BIOS SETUP, that were enabled on you card and have same HW resources, as […]
Show full quote

fumik wrote on 2023-02-16, 21:00:

I apologize that I start with a post off topic.... I'm playing with an ASUS P3V133 trying to get a 16bit ISA controller to work( […]
Show full quote

I apologize that I start with a post off topic.... I'm playing with an ASUS P3V133 trying to get a 16bit ISA controller to work(CA9342 no BIOS no PNP)
the disk attached is detected by the mainboard but cannot be accessed, I would like to understand if there is hope to make it work or if because of the PCI Bridged Isa it is not possible at all...
Could it be that mod firmware helps?
thanks if you can give me some guidance!

You need to disable the all internal HW in motherboard BIOS SETUP, that were enabled on you card and have
same HW resources, as internal - COM, LPT, FDC, IDE. Else you'll get conflict with internal and external controllers.
Unfortunately, not all board can turn off internal IDE completely. Often, only interrupt from controller will be disabled.
I've got very good result on FIC PA-2011 motherboard with 16-bit MFM controller (its registers model almost as ATA).

thanks for the answer!, yes then it's probably my case, disabling everything on the mainboard; serial and parallel are assigned those of the isa controller but hdd and floppy do not work, I can move the head with sstor but not access the data... In fact what I would relly like to do is read my old files from an MFM disk with controller wd1006, but unfortunately I can not even from a normal ide... 😀)

i386 wrote on 2023-02-17, 14:22:

You need to disable the all internal HW in motherboard BIOS SETUP, that were enabled on you card and have same HW resources, as […]
Show full quote

fumik wrote on 2023-02-16, 21:00:

I apologize that I start with a post off topic.... I'm playing with an ASUS P3V133 trying to get a 16bit ISA controller to work( […]
Show full quote

I apologize that I start with a post off topic.... I'm playing with an ASUS P3V133 trying to get a 16bit ISA controller to work(CA9342 no BIOS no PNP)
the disk attached is detected by the mainboard but cannot be accessed, I would like to understand if there is hope to make it work or if because of the PCI Bridged Isa it is not possible at all...
Could it be that mod firmware helps?
thanks if you can give me some guidance!

You need to disable the all internal HW in motherboard BIOS SETUP, that were enabled on you card and have
same HW resources, as internal - COM, LPT, FDC, IDE. Else you'll get conflict with internal and external controllers.
Unfortunately, not all board can turn off internal IDE completely. Often, only interrupt from controller will be disabled.
I've got very good result on FIC PA-2011 motherboard with 16-bit MFM controller (its registers model almost as ATA).

But I did further tests with controller this time PCI (directly no onboard bios)and everything works normally, so the onboard controller would seem actually disabled
I suspect the problem is on the implementation of PCI bridged ISA that introduce some sort of limitation on this mobo.......mah....😀

fumik wrote on 2023-02-18, 10:36:

But I did further tests with controller this time PCI (directly no onboard bios)and everything works normally, so the onboard controller would seem actually disabled
I suspect the problem is on the implementation of PCI bridged ISA that introduce some sort of limitation on this mobo.......mah....😀

PCI bridge doesn't matter at all for you problem! PCI card used PNP mechanism and it tunes
to it own resourses, no conflicts with ISA.

I want to clarify, you really disabled onboard IDE and FDC in chipset SETUP?
Your need to disable FDC controller and "IDE controller ->both" in chipset setup
(or PNP/PCI SETUP, or INTEGRAL PERIPHERALS). Disabling HDD and FDD in the standard
CMOS SETUP just will disconnect int 13 service from this devices.

Anyhow, not all boards can completely disable internal ATA controller. It can be checked,
if to boot OS from external controller (SATA/SCSI or so) and to see pci devices listing.
Or better dmesg reports from FreeBSD.

i386 wrote on 2023-02-18, 13:28:

PCI bridge doesn't matter at all for you problem! PCI card used PNP mechanism and it tunes to it own resourses, no conflicts wit […]
Show full quote

fumik wrote on 2023-02-18, 10:36:

But I did further tests with controller this time PCI (directly no onboard bios)and everything works normally, so the onboard controller would seem actually disabled
I suspect the problem is on the implementation of PCI bridged ISA that introduce some sort of limitation on this mobo.......mah....😀

PCI bridge doesn't matter at all for you problem! PCI card used PNP mechanism and it tunes
to it own resourses, no conflicts with ISA.

I want to clarify, you really disabled onboard IDE and FDC in chipset SETUP?
Your need to disable FDC controller and "IDE controller ->both" in chipset setup
(or PNP/PCI SETUP, or INTEGRAL PERIPHERALS). Disabling HDD and FDD in the standard
CMOS SETUP just will disconnect int 13 service from this devices.

Anyhow, not all boards can completely disable internal ATA controller. It can be checked,
if to boot OS from external controller (SATA/SCSI or so) and to see pci devices listing.
Or better dmesg reports from FreeBSD.

yes, I disabled ide and fdc in setup
ok, you are right, booted win98 from external controller and the mobo ide controller is still visible, also report a conflict...
maybe I can recover a mobo without onboard IDE controller to keep fiddling....
thanks for your patience!! 😀

To get this thread back on track - I can report there is no observable difference between p3bf_no_acpi_tables and p3bf_no_ACPI_tables_no_A0_A1_handlers.

Both builds allow Win2k setup and Acronis True Image 2011 to load without issue, but both also suffer from the same shutdown bug.

I have also noticed that pressing the soft power button on the case at a DOS prompt will trigger the bug as well. If you simply press and release the button the machine freezes. To make it shut off you must hold it down for 4 seconds.

maxtherabbit wrote on 2023-02-18, 21:26:

To get this thread back on track - I can report there is no observable difference between p3bf_no_acpi_tables and p3bf_no_ACPI_tables_no_A0_A1_handlers.

Both builds allow Win2k setup and Acronis True Image 2011 to load without issue, but both also suffer from the same shutdown bug.

I have also noticed that pressing the soft power button on the case at a DOS prompt will trigger the bug as well. If you simply press and release the button the machine freezes. To make it shut off you must hold it down for 4 seconds.

Problem with shutdown happens with PPro only? Other CPU are OK?

Sorry for the absence. Glad to see that others chimed in and helped, though.

maxtherabbit wrote on 2023-02-18, 21:26:

To get this thread back on track - I can report there is no observable difference between p3bf_no_acpi_tables and p3bf_no_ACPI_tables_no_A0_A1_handlers.

Both builds allow Win2k setup and Acronis True Image 2011 to load without issue, but both also suffer from the same shutdown bug.

I have also noticed that pressing the soft power button on the case at a DOS prompt will trigger the bug as well. If you simply press and release the button the machine freezes. To make it shut off you must hold it down for 4 seconds.

Thanks for the tests! Thinking about this, it kind of makes sense. Pressing the power button will make the chipset generate an SCI or SMI, which in turn causes the same BIOS handler routines to be called that would be triggered by an APM shutdown command via int 15h.

The problem lies not within writing to/reading from port 0xB2 itself. If that was the case, the board wouldn't post at all. It's something within the handlers that causes the hang. Possibly not easy to track down.

i386 wrote on 2023-02-19, 08:32:

maxtherabbit wrote on 2023-02-18, 21:26:

To get this thread back on track - I can report there is no observable difference between p3bf_no_acpi_tables and p3bf_no_ACPI_tables_no_A0_A1_handlers.

Both builds allow Win2k setup and Acronis True Image 2011 to load without issue, but both also suffer from the same shutdown bug.

I have also noticed that pressing the soft power button on the case at a DOS prompt will trigger the bug as well. If you simply press and release the button the machine freezes. To make it shut off you must hold it down for 4 seconds.

Problem with shutdown happens with PPro only? Other CPU are OK?

yes