No crashes with updated version, full automatic report is done:

The peripherals section in all versions reports "Keyboard controller: N/A", means it could not be identified or not detected at all?

I also wonder why -v (verbose) log actually produces smaller report. Seems the PnP sections are missing, and nothing really "verbose" is present 😀

Thanks for the feedback, I'm happy that this is resolved now.
"Keyboard controller: N/A" means it couldn't be detected.
Verbose mode is to produce the report file silently without showing the GUI/progress.

Greetings,

Here is the fresh update for IBM 5160, running with 1.5 version.
Now the overall performance is better than in version 1.2. The drawing speed is the same, for example it takes 18 sec from program run till welcome about window appears. The longest step at this moment is ISA Bus scan, it takes 3 min. 1 min 45 sec takes LPT scan and 1 min 25 sec takes drive detection.
What was fixed in compare to previous run:
- now it can detect BIOS manufacture
- it can now properly detect MFM drive
Issues:
- math co-processor speed still shows wrong data - 102,7 Mhz
- no ISA cards was detected
- no FDD was detected
- it crashes when I try to run Videoadapter section. See video here.

Ready to test next versions )

Susanin79 wrote on 2021-11-29, 13:25:

Greetings, […]
Show full quote

Greetings,

Here is the fresh update for IBM 5160, running with 1.5 version.
Now the overall performance is better than in version 1.2. The drawing speed is the same, for example it takes 18 sec from program run till welcome about window appears. The longest step at this moment is ISA Bus scan, it takes 3 min. 1 min 45 sec takes LPT scan and 1 min 25 sec takes drive detection.
What was fixed in compare to previous run:
- now it can detect BIOS manufacture
- it can now properly detect MFM drive
Issues:
- math co-processor speed still shows wrong data - 102,7 Mhz
- no ISA cards was detected
- no FDD was detected
- it crashes when I try to run Videoadapter section. See video here.

Ready to test next versions )

Thanks for the detailed report! According to the video you posted it seems it's crashing at "Checking: Trident, IIT". Is that right?

Posting v1.51 to hopefully fix 8087 FPU clock and display corruption on the IBM EGA adapter:
[version pulled as there's a newer one]

Oh boy. I've always hoped you would do this. Thanks!

I am in possession of an original IBM PC/AT decked out with a bunch of expansion cards. I can test the program on it later this week.

Ok. Version 1.5 raport:
- CMI8330 PnP is now correctly detected ( I think this is the most detailed description I have seen),
- Cirrus Logic VGA now correctly connected to ISA bus (instead of MCA),
- when UM9008F is plugged in program still hangs at: Checking: GUS [port 210]

Strange behavior: when I'm checking Mainboard Info -> Bus Type: ISA -> ISA and other devices first time after starting the program
it shows both sound card and ethernet as: "NE2000 or compatible Ethernet Adapter" respectively at ports 0220 and 0300,
but exiting this screen (level up to Bus info) and returning, makes the sound card item disappear (only the ethernet card remains).
Of course sound card is still in Bus Type: ISA PnP

I'm attaching debug+report files for both situations (with and without ethernet card).
I can also check with RTL8019 ISA ethernet (in PnP mode and NE2000 compatible jumper configurable mode) and another sound card, if it will be useful.

Mumak wrote on 2021-11-29, 14:34:

Thanks for the detailed report! According to the video you posted it seems it's crashing at "Checking: Trident, IIT". Is that right?

Yes, exactly right after the Trident IIT each time. Will try new version tomorrow. I need to use my 386 bace PC as a "proxy" to move files into XT. I decided to keep this XT as possible authentic as I can, so no any CF to IDE adapters is installed )

zyga64 wrote on 2021-11-29, 17:22:

Ok. Version 1.5 raport: - CMI8330 PnP is now correctly detected ( I think this is the most detailed description I have seen), - […]
Show full quote

Ok. Version 1.5 raport:
- CMI8330 PnP is now correctly detected ( I think this is the most detailed description I have seen),
- Cirrus Logic VGA now correctly connected to ISA bus (instead od MCA),
- when UM9008F is plugged in program still hangs at: Checking: GUS [port 210]

Strange behavior: when I'm checking Mainboard Info -> Bus Type: ISA -> ISA and other devices first time after starting the program
it shows both sound card and ethernet as: "NE2000 or compatible Ethernet Adapter" respectively at ports 0220 and 0300,
but exiting this screen (level up to Bus info) and returning, makes the sound card item disappear (only the ethernet card remains).
Of course sound card is still in Bus Type: ISA PnP

I'm attaching debug+report files for both situations (with and without ethernet card).
I can also check with RTL8019 ISA ethernet (in PnP mode and NE2000 compatible jumper configurable mode) and another sound card, if it will be useful.

Thanks! This has given me sufficient information to filter out the false NE2000 adapter. I could also add reporting of the valid NE2000 adapter's MAC address in the next version.

Mumak wrote on 2021-11-29, 16:23:

Posting v1.51 to hopefully fix 8087 FPU clock and display corruption on the IBM EGA adapter:
HWI16_15.ZIP

Here is the fresh tests. FPU clock now is fixed. Display test still fall with the different error right after Weitek. See video of crash.

Evgeny.

Susanin79 wrote on 2021-12-01, 14:01:

Mumak wrote on 2021-11-29, 16:23:

Posting v1.51 to hopefully fix 8087 FPU clock and display corruption on the IBM EGA adapter:
HWI16_15.ZIP

Here is the fresh tests. FPU clock now is fixed. Display test still fall with the different error right after Weitek. See video of crash.

Evgeny.

Thanks! Happy to see the FPU clock correct 😀 Is "Checking: Weitek" the last item you can see correct during the video test?

Posting v1.6 to:
- Hopefully fix display corruption during graphics card scan on EGA
- Remove phantom NE2000 devices on some systems
- Show NE2000 MAC address
[version pulled as there's a newer one]

Mumak wrote on 2021-12-01, 14:35:

Thanks! Happy to see the FPU clock correct 😀 Is "Checking: Weitek" the last item you can see correct during the video test?

Yes, the last one before the program crash.

Hi. Just for info or suggestion, in -Virtual 386 DOSBox- you can replace the S3 VGA Crash by 'unknown' or a message like : ' please try with HWINFO.EXE '.
I know this is not for a virtual environment.

Very good i will test it on my Tandon 286 8MHz, 1mb ram, 3mb xms module, isa ethernet, SB16 CT2940 and TVGA9000A 512kb graphics running on xtide cf 1GB with Dos 6.22.

Nice to see support for pre-386 systems. I tested v1.51 on my older IBM PS/2-s a few days ago.

On the Model 30 (8530-002) everything seems to work except "Video Information" - screen goes blank after checking for Matrox MGA. I can still quit the program by pressing "Esc" a few times and then the DOS prompt looks okay, but when i run the program again the display is corrupt (partially shifted to the right).

Model 30-286 (don't know the exact model, it's a bare motherboard) gives a fatal error when checking for GUS on port 250. Also, the "System Information" shows wrong memory sizes, the machine has 1Mb (4x256Kb SIMMs). Not sure if the machine type is correct or not, AFAIK the models 30 and 25 used the same motherboards.

The program seems to be lacking the MCA adapter detection part on Model 50 (8550-021), it shows the bus type as MCA but no adapters are listed. Otherwise seems to be working OK.

andunn wrote on 2021-12-02, 09:04:

Nice to see support for pre-386 systems. I tested v1.51 on my older IBM PS/2-s a few days ago. […]
Show full quote

Nice to see support for pre-386 systems. I tested v1.51 on my older IBM PS/2-s a few days ago.

On the Model 30 (8530-002) everything seems to work except "Video Information" - screen goes blank after checking for Matrox MGA. I can still quit the program by pressing "Esc" a few times and then the DOS prompt looks okay, but when i run the program again the display is corrupt (partially shifted to the right).

Model 30-286 (don't know the exact model, it's a bare motherboard) gives a fatal error when checking for GUS on port 250. Also, the "System Information" shows wrong memory sizes, the machine has 1Mb (4x256Kb SIMMs). Not sure if the machine type is correct or not, AFAIK the models 30 and 25 used the same motherboards.

The program seems to be lacking the MCA adapter detection part on Model 50 (8550-021), it shows the bus type as MCA but no adapters are listed. Otherwise seems to be working OK.

Thanks! I'm analyzing those reports to see how can I fix/improve things. Just one question - on the Model 30 are you sure the video adapter problem starts right after "Checking: Matrox MGA-I, -II, -64"? Is it possible that there could be something else shown afterwards?

Posting v1.61 in a attempt to fix the PS/2 issues you've seen (except the hang at GUS check).
BTW, "IBM PS/2 model 25, 286-10, LW-type 37" and "IBM PS/2 model 30, 286" share the same PC Model IDs, the difference is in BIOS date. "06/28/89" should be the Model 25 according to sources, though not 100% sure about this.
[version pulled as there's a newer one]

Mumak wrote on 2021-12-02, 09:24:

On the Model 30 are you sure the video adapter problem starts right after "Checking: Matrox MGA-I, -II, -64"? Is it possible that there could be something else shown afterwards?

Took a video clip of the screen with my phone and indeed it shows "MediaVision MVV452" before the screen goes blank. Couldn't see that before, it's only a few frames in the video and too fast to see with bare eyes.

Mumak wrote on 2021-12-02, 10:30:

Posting v1.61 in a attempt to fix the PS/2 issues you've seen (except the hang at GUS check).

Model 30 still has the blank screen issue, wierdly the image corruption on next program launch does not appear every time.
Model 30-286 memory amounts are fixed and everything else seems to be OK too.
Model 50 MCA bus info is still not working, when I hit "Enter" an empty "MCA Slots Configuration" screen appears for a fraction of a second and then goes back to the "Bus Info" screen.