Guru wrote on 2026-01-25, 12:40:

You already have 98% of the pinout. Trace the board. The PCI bus pinout is known. The IDE connector is known. The jumpers are pr […]
Show full quote

Comos wrote on 2023-02-16, 11:50:

jakethompson1 wrote on 2023-02-15, 23:52:

Are you looking for pinouts or other hardware stuff, or information for programming those? It's unclear whether the UM8673 ever existed as a discrete chip or only within the UM8886A by the way.

Pinout, electrical specs, programming...as usual what is written in regular datasheet. The UM8673F existed as a discrete chip, I have it as a normal PCI HDD controller.

You already have 98% of the pinout. Trace the board.
The PCI bus pinout is known. The IDE connector is known. The jumpers are probably labelled. You may actually have 100% of the pinout of that chip.

When I am repairing something and I don't have the schematic and I think it is worth fixing it then I make the schematic.
I'm a retired engineer so I have plenty of time.
Here is an example... this took about 12 hours each day full time for roughly 10 weeks. Now I can fix these slightly more easily than before.

This is not related to pinout which is the easiest part to trace down, but for the hardware configurations, IO Registers, PCI Configuration Registers, etc. which differ a bit within the series and this normally comes with the datasheet.

Comos wrote on 2026-01-30, 18:59:

Guru wrote on 2026-01-25, 12:40:

You already have 98% of the pinout. Trace the board. The PCI bus pinout is known. The IDE connector is known. The jumpers are pr […]
Show full quote

Comos wrote on 2023-02-16, 11:50:

Pinout, electrical specs, programming...as usual what is written in regular datasheet. The UM8673F existed as a discrete chip, I have it as a normal PCI HDD controller.

You already have 98% of the pinout. Trace the board.
The PCI bus pinout is known. The IDE connector is known. The jumpers are probably labelled. You may actually have 100% of the pinout of that chip.

When I am repairing something and I don't have the schematic and I think it is worth fixing it then I make the schematic.
I'm a retired engineer so I have plenty of time.
Here is an example... this took about 12 hours each day full time for roughly 10 weeks. Now I can fix these slightly more easily than before.

This is not related to pinout which is the easiest part to trace down, but for the hardware configurations, IO Registers, PCI Configuration Registers, etc. which differ a bit within the series and this normally comes with the datasheet.

If you're still interested in the UM8672 VLB IDE, mkarcher did a deep dive into the a BIOS for such a card a while back: UM8672 VLB IDE interface chip - kind of documentation
Because only the UM8886B supports bus mastering, and it is poor man's bus mastering (IRQ14 happens and only then do you bus master, so the CPU must spin around waiting for it to happen, and no scatter-gather support) I'm afraid the 8672 and 8673 aka 8886A aren't that interesting even if you had the datasheet.

There is an open issue with 8886B. Feipoa has one of those with his 8433UUD (as do I). The issue is, when the PCI bus is bumped from 33 MHz to 40 MHz (maintaining 1:1 divider even though this is out of PCI spec) the 8886B IDE speed decreases rather than increases, and I just haven't had time to dig into it.