Tried to disable the harddisks in the system BIOS and set them in the controller's BIOS?

h-a-l-9000 wrote:

Tried to disable the harddisks in the system BIOS and set them in the controller's BIOS?

Yes, I set the mobo's IDE HDDs inside the mobo's "standard BIOS section" to "None".

Incidentally, does it matter that the controller card featured in the photo has 2 slots cut in to its PCI slot section? Does it mean anything? Is it important?

When I get access to the Raid controller's BIOS section, sometimes it behaves strangely. The Raid controller's BIOS appears to "see" the HDD, most of the time anyway, but the mobo's BIOS usually locks up after I try to boot-up the machine.

Thanks for the help! 😀

retro games 100 wrote:

h-a-l-9000 wrote:

Tried to disable the harddisks in the system BIOS and set them in the controller's BIOS?

Yes, I set the mobo's IDE HDDs inside the mobo's "standard BIOS section" to "None".

Tried to disable the onboard IDE controllers as well? The setting for this should be in one of the "advanced" sections.

retro games 100 wrote:

Incidentally, does it matter that the controller card featured in the photo has 2 slots cut in to its PCI slot section? Does it mean anything? Is it important?

Nope. these notches are there to prevent inserting PCI cards into slots which don't provide the correct voltage. Your card is an universal 32-bit PCI card and should work in any PCI slot. See here for more details.

retro games 100 wrote:

When I get access to the Raid controller's BIOS section, sometimes it behaves strangely. The Raid controller's BIOS appears to "see" the HDD, most of the time anyway, but the mobo's BIOS usually locks up after I try to boot-up the machine.

Looks like a resource conflict with the onboard IDE controller. It's also possible that the controller requires a newer PCI revision than available on the mainboard. 486 chipsets are often buggy in that respect. Try to find out what the jumpers on the controller are for. Maybe you can disable some of the more advanced features on the card (busmastering, etc...).

what if you use a software solution like DISK MANAGER ? I succesfully use it on my IBM 386sx and 486/100 pc's ... did not had any troubles so far, it's like it's not there...

Thanks very much for your helpful advice! 😀

I searched the mobo's BIOS, looking for a way to disable the onboard IDE controllers, but I couldn't see a setting that seemed to be an obvious candidate. (I thought this was strange.)

The controller card I am using is an "ultra 100" card, I think. The big chip on the card has a year date of 2001 - I am thinking that this date is quite "late", in terms of using it with an ancient 486 mobo, circa '95.

Would a better option be for me to try an "ultra 66" controller card? This type of card might be older, and a bit more suited to an ancient 486 mobo. Question please - can I still connect a modern HDD to an "ultra 66" card, so I can have access to approximately 128gb of storage?

I recently tried two more booting up attempts. Attempt #1 succeeded, with a DOS 6.2 floppy boot disk in the floppy drive. When I was at the A:\ prompt, I typed fdisk, and noted that the HDD had -132 megabytes of storage space, which was a "messed up" value. After quitting fdisk, I could not go to the C:\ prompt, as this gave an error message of "invalid drive" (or words to that effect.)

Attempt #2 gave a strange display, coming from the controller card's BIOS. It is shown in the attached photo below (the red blob is the camera's flash). The information seems all "messed up". Perhaps I need to do one of the following 2 ideas -

#1 Try an "ultra 66" card instead.
#2 Try this current "ultra 100" card on a more modern mobo.

Thanks a lot for any additional comments! 😀

keropi wrote:

what if you use a software solution like DISK MANAGER ? I succesfully use it on my IBM 386sx and 486/100 pc's ... did not had any troubles so far, it's like it's not there...

That's really interesting. Are you able to connect a very large capacity modern IDE HDD (say 300gb) to your old mobo's onboard IDE controller? Sure, you're not going to get access to all of the data (I think) - perhaps somewhere between 2gb and 128gb ??, but will your mobo's BIOS accept the HDD plugged in to it's onboard IDE controller without locking up?

Thanks for any extra info about this software. 😀

The 386sx is limited to 512MB and I use there a 6GB drive (3x2GB fat16 DOS partitions) and on the 486 I have connected a 20GB HDD (had a limit at 8GB IIRC)
Since DOS only allows 2GB partitions, a 300GB HDD is insane! 🤣

keropi wrote:

The 386sx is limited to 512MB and I use there a 6GB drive (3x2GB fat16 DOS partitions) and on the 486 I have connected a 20GB HDD (had a limit at 8GB IIRC)
Since DOS only allows 2GB partitions, a 300GB HDD is insane! 🤣

A 300gb HDD certainly is insane in terms of capacity, but modern HDDs are cheap, reliable, very quiet, and very fast too. That's why I like 'em so much.

true... maybe you can cap them at 32GB and use them... it might worth installing win98se on the 486 (and make it boot to dos by deafault) only to have FAT32 so you don't get a zillion 2GB partitions

keropi wrote:

true... maybe you can cap them at 32GB and use them... it might worth installing win98se on the 486 (and make it boot to dos by deafault) only to have FAT32 so you don't get a zillion 2GB partitions

Yes, I tried capping two large drives to 32gb, but unfortunately I still had major problems with them when they were connected to the controller card.

I tried the IDE controller in another mobo - in a TX II Pro socket 7 - which is my first ever socket 7 mobo (that I've had time to test) 😀 It's got a Pentium MMX 200 in it.

This mobo's BIOS, unlike the previous 486-based one, had an option to disable the onboard IDE controller(s), and also to boot up from "other", which in this case, is the 'SCSI-like' IDE Raid controller.

It seems to work, although the IDE controller exhibited strange behaviour similar to my previous tests, in as much as when the controller displayed information about the HDD attached to it, it did it in an odd way - the text on the screen had curious colouring about it. A few letters were inappropriately coloured red, for instance. It looked odd, like it didn't make sense.

But after the IDE controller displayed this HDD information, the mobo then successfully booted up using the HDD, and windows 98 appeared.

I am beginning to think that the other 486 board cannot use an IDE controller, either because you can't disable the onboard IDE controller [but I might be stupid and can't find it], or because you cannot boot from "other" device. I must admit, this does sound rather unlikely, but I guess it's possible.

I want to test an old ultra66 controller card in the 486 mobo, so I've put in an order for one...

That card looks really dirty. Clean th edge connector with a wet q tip.

retro games 100 wrote:

I think the controller is a CSA-649U Ultra, with : […]
Show full quote

I think the controller is a CSA-649U Ultra, with :

PCI Bus
• 32-bit, 33 MHz
• PCI 2.1 compliant (wonder whether my 486 mobo is OK with this?)

The UMC 8881/8886 chipset on your board is said to be PCI 2.1 compliant, but like with all 486 PCI chipsets, you have to take this with a grain of salt. It probably has enough hardware bugs to make many modern cards incompatible .

retro games 100 wrote:

Is the problem due to a potential mismatch of PCI specification, between the mobo and the controller card?

More like the failure of the 8881/8886 to 100% comply to the PCI 2.1 specifications, but yes, I think the root of the problem is somewhere in that direction.

Maybe you can configure the CSA-649U to work in the board.
My Google-Fu is weak today, did you find anything about the jumper settings?

Moogle! wrote:

That card looks really dirty. Clean th edge connector with a wet q tip.

🤣 Looks more like JPEG artifacts than dirt.

Hehe, 5u3 is correct, the "dirt" on the PCI slot connector area is much more to do with the compression used on the photo, than anything on the card itself - it's actually quite clean! 😀

I've been messing about with the controller card quite a bit. I abandoned both the 486 board, and the pentium mmx board, and tested it in a 440bx board instead. It seemed fine, but again when the controller displayed the information about the attached HDD, some of the letters on screen were red colour, which seemed odd. It's always the same little cluster of letters too which are red!

Anyway, I took the opportunity to run SeaTools for DOS on the Seagate 320gb HDD, and limit it's capacity to just 1.5gb. I have done this in order to hopefully not confuse any ancient BIOS inside either the aforementioned 486 or pentium mmx boards, as I now going to re-run my controller tests on those boards again.

Also, sorry I couldn't find anything about those jumpers.

Maybe have a look at flash drivers, they are small in capacity.

I've done some more tests on the PCI IDE controller card -

Test 1) Tried it in a 440BX board. Passed.

Test 2) Tried it in a socket 7 pentium board. Passed.

Test 3) Tried it in the 486 mobo. Failed.

I think the problem could be any one of the following -

1) my lack of knowledge regarding the controller's jumpers.

2) can't seem to explicitly disable the mobo's IDE controllers in the BIOS.

3) controller + mobo = not best friends. 😉

Just for some fun, I ran my first speedsys test on this mobo. I used SeaTools to limit the 320GB HDD's capacity to about 1.5GB, and plugged it directly in to the mobo's onboard IDE controller. I don't know why speedsys reports the capacity as only 503MB though. fdisk seems to think the drive is 1.5GB, so too does the mobo's BIOS.

Edit: I see the "problem". Speedsys seems to think the number of cylinders in the HDD is 1023, but I think it should be 3030. Hence, speedsys seems to think that the HDD is only about one third of its actual capacity.

h-a-l-9000 wrote:

Maybe have a look at flash drivers, they are small in capacity.

Do you mean those little flash drives, you plug in to USB ports? I would need a PCI USB card...do you think it's possible to get this old 486 mobo to boot in to either DOS or Windows 95 from a flash drive?

Maybe is referring to adapters that plug into IDE ports and let you use solid state devices [aka flash drives] as hard drives. Such things exist and there are posts about them on this forum iirc.

I mean those you plug into IDE ports, yes.