Report and debug reports for the SXL2-50 system on the PANDA motherboard. The non-installed hardware PnP entries are now gone!

I have figured out how I can run NT4 on these SXL2 systems, but it only works with Service Pack 1 and in 1x mode. At least I can have the L1 cache enabled now.

I tried to HWiNFO32 w/NT4 on my AMI Mark V Baby Screamer board with the SXL2-66 installed, but I receive the error:

1HWiNFO32.exe - Unable To Locate DLL
2The dynamic link library WININET.dll could not be found in the specified path...

Is a particular service pack required to run HWiNFO32 on NT4? Perhaps I can copy over WININET.dll from a computer with SP6a installed?

feipoa wrote on 2020-04-19, 08:02:

I have figured out how I can run NT4 on these SXL2 systems, but it only works with Service Pack 1 and in 1x mode. At least I ca […]
Show full quote

I have figured out how I can run NT4 on these SXL2 systems, but it only works with Service Pack 1 and in 1x mode. At least I can have the L1 cache enabled now.

I tried to HWiNFO32 w/NT4 on my AMI Mark V Baby Screamer board with the SXL2-66 installed, but I receive the error:

Copy code to clipboard

1HWiNFO32.exe - Unable To Locate DLL
2The dynamic link library WININET.dll could not be found in the specified path...

Is a particular service pack required to run HWiNFO32 on NT4? Perhaps I can copy over WININET.dll from a computer with SP6a installed?

Yes, I definitely recommend latest SP. But if that won't work, you might try to copy that DLL. It might require other DLLs as well.

In case anybody else is running the original installation of NT4, HWiNFO needs the following from SP6:
crypt32.dll
msasn1.dll
shlwapi.dll
wininet.dll

HWiNFO will run, but I was unable to open the Settings box that first appears when you open HWiNFO, and I was unable to generate a report file.

Are you able to install SP6? I think the OS is also missing some important GUI features required.

No, not on my SXL2-66 system. I recall trying all the other service packs, starting with 6, then 5, etc. but the system would no longer boot after service pack installation. I only have SP6, SP5, and SP4 saved on my hard drive, so it is possible that I quit after SP6, 5, and 4 attempts.

Hello Mumak,

I saw this thread, and I'd thought I'd help.

I downloaded the latest v6.24 Installer from your website, and tested it on three different retro systems of mine with three different OSes.

First up, my Pentium MMX machine under Windows 95 OSR2.5. Everything detected properly, although the small number of sensors being detected was sad, but I guess for this hardware that's to be expected, no?

Then, I tried HWINFO32 on my Pentium III-S machine running Windows 98SE. Again, everything detected properly, but this time the dearth of sensors was more unexpected. I believe years ago when I ran HWINFO32 on my Slot 1 Pentium III system it showed a CPU temperature sensor. However, on this machine, no such sensor is displayed. Either the software is lacking, or the sensor isn't accessible under Windows. The Voodoo5 5500 in the system I know doesn't have a temperature sensor, so no qualms there.

Next, I tried HWINFO32 on my Pentium 4 (Socket 775) machine running Windows 98SE. Again, everything detected properly, but this time temperature sensors for both the CPU and GPU were being shown. Yay! I also ran HWINFO32 under XP Pro. SP3 32-bit on the same system for comparison on how the software works under both OSes (looked more or less identical to me).

Anyways, I've attached all of the logs and debug files below (which also includes the report I gathered under DOS 6.22 on my Pentium MMX machine using the latest DOS version of HWINFO from your website), and if you want an overview of each of the systems, you can click here.

the_ultra_code wrote on 2020-04-20, 07:36:

Hello Mumak, […]
Show full quote

Hello Mumak,

I saw this thread, and I'd thought I'd help.

I downloaded the latest v6.24 Installer from your website, and tested it on three different retro systems of mine with three different OSes.

First up, my Pentium MMX machine under Windows 95 OSR2.5. Everything detected properly, although the small number of sensors being detected was sad, but I guess for this hardware that's to be expected, no?

Then, I tried HWINFO32 on my Pentium III-S machine running Windows 98SE. Again, everything detected properly, but this time the dearth of sensors was more unexpected. I believe years ago when I ran HWINFO32 on my Slot 1 Pentium III system it showed a CPU temperature sensor. However, on this machine, no such sensor is displayed. Either the software is lacking, or the sensor isn't accessible under Windows. The Voodoo5 5500 in the system I know doesn't have a temperature sensor, so no qualms there.

Next, I tried HWINFO32 on my Pentium 4 (Socket 775) machine running Windows 98SE. Again, everything detected properly, but this time temperature sensors for both the CPU and GPU were being shown. Yay! I also ran HWINFO32 under XP Pro. SP3 32-bit on the same system for comparison on how the software works under both OSes (looked more or less identical to me).

Anyways, I've attached all of the logs and debug files below (which also includes the report I gathered under DOS 6.22 on my Pentium MMX machine using the latest DOS version of HWINFO from your website), and if you want an overview of each of the systems, you can click here.

Thank you for the results! Unfortunately they don't contain sensor data (you probably didn't open the Sensors window when you captured the reports), so I cannot check why the P-III machine is lacking the sensors. If you provide such files, I will check that in detail and see if there's anything to be fixed.

Mumak wrote on 2020-04-20, 07:55:

the_ultra_code wrote on 2020-04-20, 07:36:

Hello Mumak, […]
Show full quote

Hello Mumak,

I saw this thread, and I'd thought I'd help.

I downloaded the latest v6.24 Installer from your website, and tested it on three different retro systems of mine with three different OSes.

First up, my Pentium MMX machine under Windows 95 OSR2.5. Everything detected properly, although the small number of sensors being detected was sad, but I guess for this hardware that's to be expected, no?

Then, I tried HWINFO32 on my Pentium III-S machine running Windows 98SE. Again, everything detected properly, but this time the dearth of sensors was more unexpected. I believe years ago when I ran HWINFO32 on my Slot 1 Pentium III system it showed a CPU temperature sensor. However, on this machine, no such sensor is displayed. Either the software is lacking, or the sensor isn't accessible under Windows. The Voodoo5 5500 in the system I know doesn't have a temperature sensor, so no qualms there.

Next, I tried HWINFO32 on my Pentium 4 (Socket 775) machine running Windows 98SE. Again, everything detected properly, but this time temperature sensors for both the CPU and GPU were being shown. Yay! I also ran HWINFO32 under XP Pro. SP3 32-bit on the same system for comparison on how the software works under both OSes (looked more or less identical to me).

Anyways, I've attached all of the logs and debug files below (which also includes the report I gathered under DOS 6.22 on my Pentium MMX machine using the latest DOS version of HWINFO from your website), and if you want an overview of each of the systems, you can click here.

Thank you for the results! Unfortunately they don't contain sensor data (you probably didn't open the Sensors window when you captured the reports), so I cannot check why the P-III machine is lacking the sensors. If you provide such files, I will check that in detail and see if there's anything to be fixed.

NP. Here you go - a new report and debug file, in each instance I opened up the "Sensors" menu.

Also, I was wondering - would it be possible to add an "Add shortcut to desktop" option at the end of the installation process, before you hit the "Finish" button? Most other monitoring utilities have such a feature, while HWINFO doesn't. It's not the end of the world if it never gets one, but without one, I have to go to the Start Menu or go to where the executable is located and copy/make a shortcut and put it on the desktop. An inconvenience, for sure, but I'd think an easy one to solve.

the_ultra_code wrote on 2020-04-20, 08:07:

Mumak wrote on 2020-04-20, 07:55:

the_ultra_code wrote on 2020-04-20, 07:36:

Hello Mumak, […]
Show full quote

Hello Mumak,

I saw this thread, and I'd thought I'd help.

I downloaded the latest v6.24 Installer from your website, and tested it on three different retro systems of mine with three different OSes.

First up, my Pentium MMX machine under Windows 95 OSR2.5. Everything detected properly, although the small number of sensors being detected was sad, but I guess for this hardware that's to be expected, no?

Then, I tried HWINFO32 on my Pentium III-S machine running Windows 98SE. Again, everything detected properly, but this time the dearth of sensors was more unexpected. I believe years ago when I ran HWINFO32 on my Slot 1 Pentium III system it showed a CPU temperature sensor. However, on this machine, no such sensor is displayed. Either the software is lacking, or the sensor isn't accessible under Windows. The Voodoo5 5500 in the system I know doesn't have a temperature sensor, so no qualms there.

Next, I tried HWINFO32 on my Pentium 4 (Socket 775) machine running Windows 98SE. Again, everything detected properly, but this time temperature sensors for both the CPU and GPU were being shown. Yay! I also ran HWINFO32 under XP Pro. SP3 32-bit on the same system for comparison on how the software works under both OSes (looked more or less identical to me).

Anyways, I've attached all of the logs and debug files below (which also includes the report I gathered under DOS 6.22 on my Pentium MMX machine using the latest DOS version of HWINFO from your website), and if you want an overview of each of the systems, you can click here.

Thank you for the results! Unfortunately they don't contain sensor data (you probably didn't open the Sensors window when you captured the reports), so I cannot check why the P-III machine is lacking the sensors. If you provide such files, I will check that in detail and see if there's anything to be fixed.

NP. Here you go - a new report and debug file, in each instance I opened up the "Sensors" menu.

Also, I was wondering - would it be possible to add an "Add shortcut to desktop" option at the end of the installation process, before you hit the "Finish" button? Most other monitoring utilities have such a feature, while HWINFO doesn't. It's not the end of the world if it never gets one, but without one, I have to go to the Start Menu or go to where the executable is located and copy/make a shortcut and put it on the desktop. An inconvenience, for sure, but I'd think an easy one to solve.

Thank you. Unfortunately I don't recognize any HW monitoring chip on that system. Are you sure it was reported in the past?

Mumak wrote on 2020-04-20, 08:16:

the_ultra_code wrote on 2020-04-20, 08:07:

Mumak wrote on 2020-04-20, 07:55:

Thank you for the results! Unfortunately they don't contain sensor data (you probably didn't open the Sensors window when you captured the reports), so I cannot check why the P-III machine is lacking the sensors. If you provide such files, I will check that in detail and see if there's anything to be fixed.

NP. Here you go - a new report and debug file, in each instance I opened up the "Sensors" menu.

Also, I was wondering - would it be possible to add an "Add shortcut to desktop" option at the end of the installation process, before you hit the "Finish" button? Most other monitoring utilities have such a feature, while HWINFO doesn't. It's not the end of the world if it never gets one, but without one, I have to go to the Start Menu or go to where the executable is located and copy/make a shortcut and put it on the desktop. An inconvenience, for sure, but I'd think an easy one to solve.

Thank you. Unfortunately I don't recognize any HW monitoring chip on that system. Are you sure it was reported in the past?

Well, on a Slot 1 Pentium 3, not a Socket 370 Tualatin. Guess I can look into it. I mean, you can see CPU temp in BIOS, so the CPU should have a sensor on it. But, I'm guessing not all sensors that are accessible in-BIOS are accessible in an OS.

If I have time, I'll set up a Slot 1 testbench and see if I can get a sensor reading if I can't find anything online.

Ok.

1. I486SX is detected correctly, logfile attached
2. I have tested again 486DX2-66 on FIC 486-GVT-2 with 2 VLB I/O cards (QD6500 and Promise PDC20230C), but sadly both make program crash. I have taken photo of both adapters, and screenshot when system freezes. Promise was shot twice – firs time the screen was corrupted, second time I have turned L2 cache WB enable, and system crashed too, but this time without corrupting the screen. Logfile and photos attached.

Regards

P.S.

Sorry for size of attachment.
Next time I will lower the resolution

Thanks for the results! Please try to disable the "Check Keyboard Controller and Super IO" option if that will fix the crash.

Back again

I did some tests with “Check Keyboard Controller and Super IO” option enabled, and disabled.
Testbed remains the same (FIC 486-GVT-2 + Cirrus Logic CL-GD5428 VLB card + 486DX2-66), except IO cards.
I have tested following cards
- Octek VL-Combo VLB Promise PDC20230C
- VLSIOVX VLB Promise PDC20230C
- DC2000VL-2 VLB Promise PDC20230B
- QD6500A VLB Vision QD6500-119
- PTI-227B V2 ISA Winbond W83757F
- IDE-PLUS-V4L ISA GoldStar Prime 2C
All cards behaved in the same manner, when Keyboard Controller and Super IO” option was disabled –thing was fine; when I had enabled this option – program crashed on peripherals screen.
I wonder whatever it is the fault of the program , or the motherboard. Unfortunately I don’t think I have another working VLB board at the moment, so I cannot test it.
Logs included, though I don’t think, they will be of much use.
Regards

P.S. I have found a website wits some information about old VLB controllers (some datasheets and some register info).
http://www.ryston.cz/petr/vlb/vlbidechips.html

phantom_pl wrote on 2020-04-21, 07:04:

Back again […]
Show full quote

Back again

I did some tests with “Check Keyboard Controller and Super IO” option enabled, and disabled.
Testbed remains the same (FIC 486-GVT-2 + Cirrus Logic CL-GD5428 VLB card + 486DX2-66), except IO cards.
I have tested following cards
- Octek VL-Combo VLB Promise PDC20230C
- VLSIOVX VLB Promise PDC20230C
- DC2000VL-2 VLB Promise PDC20230B
- QD6500A VLB Vision QD6500-119
- PTI-227B V2 ISA Winbond W83757F
- IDE-PLUS-V4L ISA GoldStar Prime 2C
All cards behaved in the same manner, when Keyboard Controller and Super IO” option was disabled –thing was fine; when I had enabled this option – program crashed on peripherals screen.
I wonder whatever it is the fault of the program , or the motherboard. Unfortunately I don’t think I have another working VLB board at the moment, so I cannot test it.
Logs included, though I don’t think, they will be of much use.
Regards

P.S. I have found a website wits some information about old VLB controllers (some datasheets and some register info).
http://www.ryston.cz/petr/vlb/vlbidechips.html

Thanks for the exhaustive testing!
I don't think this is related to the cards, but mainboard. As I have already explained in the HWiNFO for DOS thread, detecting such systems requires probing of various ports, which can cause troubles on some systems where these ports are used for something else which collides with the test method.
I will add an automatic exclusion of SIO check for your mainboard model in the next build.
BTW, Do you know what SIO chip is on that mainboard ?

As far as I know, this board doas not have any SIO chip, but relies on external cards.
It uses VIA chipset VIA VT82C486A, and has an additional chip VIA VT82C482.

Then I think something in the SIO detection is conflicting with the VT82C486A. Those early chipsets used ports all over, unfortunately I wasn't able to find a datasheet for the VT82C486A (only for the 496G).

Quick Update

I managed to revive another 486 VLB board, that I thought was dead - Soyo Sy-025K2, based on SiS chipset SIS 85C471.
I tested it with "Check Keyboard Controller and Super IO" option Enabled, and with the same boards as above, except VLSIOVX VLB Promise PDC20230C board, that refused to cooperate:-(.
Pros: Nothings crashes, report is generated without problems
Cons: Still no detection at all 🙁
Logs attached.
Regards

Thanks. With "no detection at all" you mean detection of the IDE controller cards? This is currently not implemented, I checked the IDE documentation and only some of the PDC controllers could be theoretically detected. Let me add this and I will post a new test build.
I have also checked the SIS 85C471 datasheet, but can't see anything there that could allow a detection of this chip.