Seconary IDE channel issue in intel 865G pentium 4 chipset dealing with windows 98SE

edited February 2020 in Hardware

I just recently found out a pentium 4 tower I have , has drivers for all the hardware for windows 98 , but of course it has issues based on how it is installed.

2 things can happen based on how windows 98 SE is installed :

1) If installed with acpi disabled and force apm mode using setup switches , windows 98SE will freeze on the boot screen, if the secondary IDE channel is enabled. Googling this secondary IDE channel issue , came up with a possible solution of disabling acpi and installing with apm mode only, which doesn't work on this computer and causes the freeze issue.

2) If installed with acpi enabled , windows 98SE boots fine but has a resource conflict dealing with the entire IDE controller. If the secondary IDE Channel is disabled in the bios, the IDE controller will work correctly and resource conflict will go away.

I believe Windows 98SE might have some bug in its IDE controller driver, which is probably causing this issue. I am wondering if there are any updated IDE drivers for windows 98SE, that might fix this issue?

Googling this issue it seems like other people has this same issue dealing with the IDE controller on this same chipset. So If anyone can help me with this issue that would be great. There seems to be no solution of this issue online anywhere.

Comments

  • edited February 2020

    If you can get it to boot with the channel disabled, do that and then try installing the intel inf installer. Some versions might call it the chipset installation utility or something like that. That should load the chipset drivers it needs. Then maybe you will be able to re-enable that channel. There are some here: https://www.philscomputerlab.com/intel-chipset-drivers.html

    Another thing to look at, maybe first, is to check for swollen capacitors on the system board. 865 chipset motherboards would probably be right in that time frame. So in that case, maybe it's not software and drivers, but instead a hardware problem causing the lock up.

  • edited February 2020

    @jafir said:
    If you can get it to boot with the channel disabled, do that and then try installing the intel inf installer. Some versions might call it the chipset installation utility or something like that. That should load the chipset drivers it needs. Then maybe you will be able to re-enable that channel. There are some here: https://www.philscomputerlab.com/intel-chipset-drivers.html

    Another thing to look at, maybe first, is to check for swollen capacitors on the system board. 865 chipset motherboards would probably be right in that time frame. So in that case, maybe it's not software and drivers, but instead a hardware problem causing the lock up.

    I already tried the chipset drivers , still does the same thing. Yes this computer does have capacitor failure , but it is the filter capacitors for the cpu only. This secondary ide channel issue goes back to 2004 , when this computer was made, on what I could find online, so that makes the capacitors not the problem. Also windows XP works perfectly on it as well. Also all drivers are installed for windows 98SE on the towers, so that isn't the problem either.

  • I've got 2 or 3 Dell Optiplex GX270, which I'm pretty sure have the same chipset, and windows 98 works fine. I did have to search out drivers for other stuff, because win98 was never officially supported by Dell. I'll fire one up when I get home and see what resources are being used by what devices when win98 is running. Perhaps that will help in selecting alternate IRQ or whatever is conflicting.

  • edited February 2020

    @jafir said:
    I've got 2 or 3 Dell Optiplex GX270, which I'm pretty sure have the same chipset, and windows 98 works fine. I did have to search out drivers for other stuff, because win98 was never officially supported by Dell. I'll fire one up when I get home and see what resources are being used by what devices when win98 is running. Perhaps that will help in selecting alternate IRQ or whatever is conflicting.

    I forgot to say , I have another computer as well with the same identical chipset, and it does the same thing. So I am guessing the issue must involve the bios , or some other hardware on the motherboard. So based how the hardware is configured by the bios, is probably the issue.

  • Sometimes you can go into device manager and change the resources used by PCI devices. If you can figure out what's conflicting, maybe you can choose options that are better than what the plug and play BIOS is choosing on it's own.

  • @jafir said:
    Sometimes you can go into device manager and change the resources used by PCI devices. If you can figure out what's conflicting, maybe you can choose options that are better than what the plug and play BIOS is choosing on it's own.

    What resources would you need to change? as the tower gives no resource conflict errors at all , when the secondary IDE channel is disabled.

    I also forgot that I actually have another 3rd tower different brand, with the same chipset and this time it doesn't have the issue at all, Secondary ide channel works fine, and this tower also has bad capacitors in the same spot as the tower that does have the issues.

    Its so odd that 2 towers that are different have the same problem, but a 3rd tower that is also different and the same chipset has no problems at all.

  • I’d compare with the system that doesn’t have issues.

    Are there cards in the system or is everything onboard? If possible perhaps try removing cards or moving them to different slots.

  • @jafir said:
    I’d compare with the system that doesn’t have issues.

    Are there cards in the system or is everything onboard? If possible perhaps try removing cards or moving them to different slots.

    there are no cards in the system at all

    Figured out what the resource conflict is , When both ide channels are enabled, windows 98 says there is a conflict with a I/O memory range, that deals with the DMA controller.

    I need to figure out if on the other tower that works if it using any range that is the same for the DMA controller .

  • I changed the I/O memory range setting for the controller , and it still has the same problem. That it still gives a conflict if the secondary channel is enabled, but I found something odd though, that gets the channel to work and I see the cd drive in windows 98SE , but if you reboot windows 98SE won't boot.

    on the IDE controller , go into properties and as long as the secondary channel is disabled and set to primary channel enabled only, if I go to control panel, add hardware, add a standard IDE controller , and set the hardware resources to the exact settings as the secondary channel, it will show up the cd/dvd drive in windows 98, but if you then reboot windows 98SE will refuse to boot.

  • edited February 2020

    I still can't get it to work , but I did figure out the resource conflict and the primary and secondary channel will show up, but windows 98 will refuse to boot if they both are enabled at the same time.

    It seems windows 98 will only boot if one channel is enable only. If I disable the 2nd channel through device manager windows 98 will boot fine , and the odd thing , if I than re-enable the 2nd IDE channel after the restart , it will tell me to reboot, I click cancel so I can reboot later , so I can check to see if the cd/dvd drive shows up in explorer. The answer is it does , and I can read a cd or dvd in the drive fine, but than if I reboot , windows 98 freezes at the boot screen and all drive activity stops, even the Led light for the hard drive stops blinking.

    I am sure the issue involves both IDE channels being enabled at the same time because I can disable the primary channel as well, which just makes the C drive run in ms-dos mode , and the secondary channel will work just fine, with the cd/dvd drive working fine.

    So the issue seems to be a bug in the IDE driver for windows 98, on this particular computer. It could also involve the bios as well, as if resources aren't being allocated correctly, its a issue with acpi / plug n play.

Sign In or Register to comment.