Nice, we discussed about this with Tiido and others 3-4 years ago
PCI sound cards and Chipsets from various manufacturers...
and finally someone made it 😀

It's not clear to me, or I overlooked it, where Rasteri connect LDRQ# signal from to his adapter. All MBs that I met has not routed LDRQ# to TPM header - of course, why would TPM chip needs a DMA transfers for a tiny amount of data. So it need to be hacked out of the MB, mostly possible way would be to locate SuperIO chip and tap a pin there. SuperIO may use DMA for LPT port in ECP mode and floppy so it could't be used at same time.
I would be happy to blow off the dust from my bought Fintek bridge and build a copy of the adapter.
I bought mine from Ali but no longer avail from this seller https://www.aliexpress.com/item/32839231409.html

LSS10999 wrote on 2023-03-26, 12:43:

By the way, it seems the executable I built with DJGPP (actually cross-compiled from Linux) doesn't run when JEMM386 is active. It outright pagefaults. Running in real mode (without JEMM386)

Hm, this must be some specific linux crosscompiler issue as I normally use DJGPP for my tools (native under DOS) and all runs fine under EMM/JEMM/QEMM386 as well as only himem and bare RM...

For the initilaization there would be another possible way. If some one is hardcore in ASM enough, he can make init code as PCI option ROM - feel free to use my ROMOS as a starting point: http://rayer.g6.cz/romos/romose.htm
I already have a simple hack that enables VGA16 decoding in my chipset to fix bad Gigabyte BIOS bug in PCI option ROM together with ROMOS....

LSS10999 wrote on 2023-03-28, 00:48:

For Intel chipset, its LPC controller only allows up to four I/O ranges to be forwarded to the LPC-ISA bridge so you can't reall […]
Show full quote

Kreshna Aryaguna Nurzaman wrote on 2023-03-27, 14:20:

I read about it from my Android phone's Google news feed instead of here in Vogons. Great job nonetheless. I wonder, is it possible to have more than one ISA slots with this solution? Perhaps using split connector or such? For example, using a Sound Blaster AWE 64 Gold and a Yamaha SW60XG MIDI card --each requires its own ISA slot-- on modern PC.

For Intel chipset, its LPC controller only allows up to four I/O ranges to be forwarded to the LPC-ISA bridge so you can't really use every feature on an AWE64 Gold. I've tried that before on my RUBY-9719VG2AR.

An AWE32/64 requires four I/O ranges: 220, 620, A20, E20 (EMU8000 requires the latter three). As a result, to get it properly configured I need to configure the LPC controller and the bridge this way, giving up MIDI (330) and FM (388) in the process. Should note that FM does work in DIAGNOSE, which I suspect it might be accessing it from port 220. I haven't tested AWEUTIL, but if MIDI (330) provided by it is driven via EMU ports (620/A20/E20) rather than real port 330 then it could probably work.

If using the default forwarding option, the AWE64 Gold would be detected as a Sound Blaster 16, with no usable FM, as without accesses to all the required ports, EMU8000 (which contains the FM core) cannot be initialized, so only the DSP on the card is functional.

A pity. And it's not dISAppointment's fault, it's Intel's limitation.

But alright. Let say I use an ordinary ISA sound card like Sound Blaster 16 and Sound Blaster Pro, which only consumes one I/O address (220). Is it possible to use more than one ISA sound card with dISAppointment? Perhaps using split connector or the likes?

Kreshna Aryaguna Nurzaman wrote on 2023-03-28, 07:35:

A pity. And it's not dISAppointment's fault, it's Intel's limitation.

But alright. Let say I use an ordinary ISA sound card like Sound Blaster 16 and Sound Blaster Pro, which only consumes one I/O address (220). Is it possible to use more than one ISA sound card with dISAppointment? Perhaps using split connector or the likes?

I'm not sure, but these bridges themselves should in theory support more than one ISA slot... just normally only one is being provided considering a motherboard's form factor.

RayeR wrote on 2023-03-28, 02:14:

It's not clear to me, or I overlooked it, where Rasteri connect LDRQ# signal from to his adapter. All MBs that I met has not routed LDRQ# to TPM header

So there is kind of a standard for TPM headers, I've seen the same pinout across several manufacturers. However, the catch is LDRQ is normally connected via a zero-ohm link that is omitted from final production motherboards. Finding this link is pretty easy though, and can be replaced with a blob of solder. I'll write a general installation guide at some point.

LSS10999 wrote on 2023-03-26, 12:43:

By the way, it seems the executable I built with DJGPP (actually cross-compiled from Linux) doesn't run when JEMM386 is active. It outright pagefaults. Running in real mode (without JEMM386)

Hm, this must be some specific linux crosscompiler issue as I normally use DJGPP for my tools (native under DOS) and all runs fine under EMM/JEMM/QEMM386 as well as only himem and bare RM...

I have the version built directly from DOS here : https://github.com/rasteri/dISAppointment/blo … re/SAPPHISA.EXE - maybe that'll work better for @LSS10999 ?

For the initilaization there would be another possible way. If some one is hardcore in ASM enough, he can make init code as PCI option ROM - feel free to use my ROMOS as a starting point: http://rayer.g6.cz/romos/romose.htm
I already have a simple hack that enables VGA16 decoding in my chipset to fix bad Gigabyte BIOS bug in PCI option ROM together with ROMOS....

Yeah this would be the ideal way, especially if we want to experiment with ISA graphics adapters. The Fintek bridge actually has the neccesary decoding logic for option ROMs built-in so it would be trivial to add one. The next revision of the hardware will probably have a footprint for a ROM.

Thanks for the project to base it on - 8086 real-mode assembly is something I've stubbornly refused to learn (segmented memory grumble grumble) so this is a big help to get started.

LSS10999 wrote on 2023-03-28, 09:11:

Kreshna Aryaguna Nurzaman wrote on 2023-03-28, 07:35:

A pity. And it's not dISAppointment's fault, it's Intel's limitation.

But alright. Let say I use an ordinary ISA sound card like Sound Blaster 16 and Sound Blaster Pro, which only consumes one I/O address (220). Is it possible to use more than one ISA sound card with dISAppointment? Perhaps using split connector or the likes?

I'm not sure, but these bridges themselves should in theory support more than one ISA slot... just normally only one is being provided considering a motherboard's form factor.

I have seen many 845 and 865 motherboards that only provide four expansion slots, one AGP and three PCIs. As far as form factor is concerned, could we use dISAppointment to add three ISA slots to such motherboards?

An Asus P5PE-VM Intel 865 I've seen on Ebay.

Kreshna Aryaguna Nurzaman wrote on 2023-03-28, 13:23:

I have seen many 845 and 865 motherboards that only provide four expansion slots, one AGP and three PCIs. As far as form factor […]
Show full quote

LSS10999 wrote on 2023-03-28, 09:11:

Kreshna Aryaguna Nurzaman wrote on 2023-03-28, 07:35:

A pity. And it's not dISAppointment's fault, it's Intel's limitation.

But alright. Let say I use an ordinary ISA sound card like Sound Blaster 16 and Sound Blaster Pro, which only consumes one I/O address (220). Is it possible to use more than one ISA sound card with dISAppointment? Perhaps using split connector or the likes?

I'm not sure, but these bridges themselves should in theory support more than one ISA slot... just normally only one is being provided considering a motherboard's form factor.

I have seen many 845 and 865 motherboards that only provide four expansion slots, one AGP and three PCIs. As far as form factor is concerned, could we use dISAppointment to add three ISA slots to such motherboards?

An Asus P5PE-VM Intel 865 I've seen on Ebay.

Maybe I've missed something, but doesn't this require a TPM header?

Ozzuneoj wrote on 2023-03-28, 14:02:

Kreshna Aryaguna Nurzaman wrote on 2023-03-28, 13:23:

I have seen many 845 and 865 motherboards that only provide four expansion slots, one AGP and three PCIs. As far as form factor […]
Show full quote

LSS10999 wrote on 2023-03-28, 09:11:

I'm not sure, but these bridges themselves should in theory support more than one ISA slot... just normally only one is being provided considering a motherboard's form factor.

I have seen many 845 and 865 motherboards that only provide four expansion slots, one AGP and three PCIs. As far as form factor is concerned, could we use dISAppointment to add three ISA slots to such motherboards?

An Asus P5PE-VM Intel 865 I've seen on Ebay.

Maybe I've missed something, but doesn't this require a TPM header?

Ah you're right. TPM started in 2009. Intel 865 was launched in 2003.

How about more modern motherboards with mATX form factor then. When placed in a full ATX case, there would be space available for three more expansion cards. I wonder if this solution can provide more than one ISA slots.

Kreshna Aryaguna Nurzaman wrote on 2023-03-28, 14:19:

I wonder if this solution can provide more than one ISA slots.

It could but limitations present in the chip that does the LPC to ISA would severely hinder usage of multiple cards at the same time.

Which makes me wonder...

Can you attach multiple LPC to ISA chips to the LPC bus ?

By the way, I wonder if there is any room of improvement for the adapter's power input. Right now it's using a full ATX 24-pin connector which requires a splitter of enough length to connect both the motherboard and the adapter to the PSU.

Considering modern PSUs no longer have the -5V (the pin is empty), the ISA slot's -5V line will not be powered this way.

I'm thinking if it's possible to drive it using a 5-wire SATA connector (that comes directly from the PSU which usually includes the 3.3V line). That still leaves -5V and -12V needed to be provided through other means, however.

Kreshna Aryaguna Nurzaman wrote on 2023-03-28, 14:19:

Ozzuneoj wrote on 2023-03-28, 14:02:

Kreshna Aryaguna Nurzaman wrote on 2023-03-28, 13:23:

I have seen many 845 and 865 motherboards that only provide four expansion slots, one AGP and three PCIs. As far as form factor […]
Show full quote

I have seen many 845 and 865 motherboards that only provide four expansion slots, one AGP and three PCIs. As far as form factor is concerned, could we use dISAppointment to add three ISA slots to such motherboards?

An Asus P5PE-VM Intel 865 I've seen on Ebay.

Maybe I've missed something, but doesn't this require a TPM header?

Ah you're right. TPM started in 2009. Intel 865 was launched in 2003.

How about more modern motherboards with mATX form factor then. When placed in a full ATX case, there would be space available for three more expansion cards. I wonder if this solution can provide more than one ISA slots.

I also have an ITX board with TPM. The smaller boards can make it so no expansion box is necessary, and all can be kept in a single ATX or mini ATX case.

rasteri wrote on 2023-03-28, 10:57:

So there is kind of a standard for TPM headers, I've seen the same pinout across several manufacturers. However, the catch is LDRQ is normally connected via a zero-ohm link that is omitted from final production motherboards. Finding this link is pretty easy though, and can be replaced with a blob of solder. I'll write a general installation guide at some point.

I think there's no such rigid standard. There's only defined some minimal subset of LPC signals that needs to be present on TPM header but the rest is on manufacturer will. Also there are at leas 2 different header pitches. I luckily have a leaked schematics of my Gigabyte GA-P67-DS3-B3 mainboard so I can tell you that pin 20 is connected to SUSCLK via resistor somewhere to PCH, not the LDRQ. The PCH has LDRQ0 - wired to SuperIO and LDRQ1 - wired to pull-up only, nowhere else (which is good because it makes it easily accessible). So I guess that many users will not have easy way to hook LDRQ...

I didn't studied Fintek datasheet much detailed - what option ROM interface it supports? Some parallel flash (sharing Axx, Dxx ISA bus signals) or SPI flashrom? If the 1st case then it would probably require that the bridge will be configured to decode specific memory range from/to LPC bus that I doubt BIOS will do by default. Then it would lead to chicken-egg loop problem 😀 So I think it would be easier to insert PCI option ROM image into main BIOS.

LSS10999 wrote on 2023-03-28, 16:24:

I'm thinking if it's possible to drive it using a 5-wire SATA connector (that comes directly from the PSU which usually includes the 3.3V line). That still leaves -5V and -12V needed to be provided through other means, however.

Yes, it's not hard to make an inverter switched power supply or charge pump from 5V to -5V and 12V to -12V. Usually ISA devices will not sink much current from this rails. RS232 drivers or OpAmps at soundcard would just take some 10-20mA or so,...

RayeR wrote on 2023-03-28, 18:21:

I think there's no such rigid standard. There's only defined some minimal subset of LPC signals that needs to be present on TPM header but the rest is on manufacturer will. Also there are at leas 2 different header pitches. I luckily have a leaked schematics of my Gigabyte GA-P67-DS3-B3 mainboard so I can tell you that pin 20 is connected to SUSCLK via resistor somewhere to PCH, not the LDRQ. The PCH has LDRQ0 - wired to SuperIO and LDRQ1 - wired to pull-up only, nowhere else (which is good because it makes it easily accessible). So I guess that many users will not have easy way to hook LDRQ...

I have half a dozen motherboards from different manufacturers that all use the same pinout as the disappointment, but you're correct that there are a lot more pinouts out there. And yeah LDRQ is missing from a lot of them. I also have a gigabyte motherboard with the same pinout as yours - I'm going to try recovering LDRQ so I'll let you know how I get on.

I didn't studied Fintek datasheet much detailed - what option ROM interface it supports? Some parallel flash (sharing Axx, Dxx ISA bus signals) or SPI flashrom? If the 1st case then it would probably require that the bridge will be configured to decode specific memory range from/to LPC bus that I doubt BIOS will do by default. Then it would lead to chicken-egg loop problem 😀 So I think it would be easier to insert PCI option ROM image into main BIOS.

It's a parallel flash it supports, and by default it's enabled for 0x000F0000-0x000FFFFF. So yes not terribly useful. Maybe I won't add a rom after all.

Managed to mod a Gigabyte GA-Z77-D3H, it works really well. LDRQ was pretty easy to recover, there was an unpopulated pullup resistor under the PCH heatsink. I just brought it out to a pin header.

Good work. what LDRQ did you steal? 0 or 1?
BTW need a schematic? This web is quite rich of GB leaked docs 😀
http://smd.db-x7.ru/load/boardview_mat_plat/g … df/299-1-0-7448
I surely want to build my adapter, thinking about combine it with 2nd superIO from my older LPCIO project. But first need a time (busy with other real-life stuffs) to finish superIO init SW and test the floppy. I would also include DC-DC conversters to avoid need of ATX huge connector...
I have also one spare very similar MB to one I currently use Gigabyte GA-P67A-D3-B3 (also schematic available) http://rayer.g6.cz/hardware/gap67ad3.htm
So I can freely solder wires over it without disturbing the main PC 😀

Also it would be good idea to test how long the LPC wires from TPM could be. I guess it's not designed to go out far from MB so it wouldn't like expose to noisy environment, crsosstalk and excessive capacitive load on the LPC bus (weak PCH line drivers). No idea how to detect errors on LPC and how system will behave when it happens, I would expect some random hangs...

The decoded range 0x000F0000-0x000FFFFF would conflict with main BIOS, so IMHO not usefull, don't bother adding flash...

A question regarding LPC-ISA bridge in general:

Recently I've been experimenting with the F85226 bridge on my RUBY-9719VG2AR and I noticed something. It appears I don't really have to program the generic address decoders (ADDR1-ADDR4) on the Fintek bridge side. I tweaked the program and purposedly omitted the code for programming the address decoders so that only the decoders on the Intel LPC controller gets programmed (all the address decoders on the Fintek bridge remain default, which is 0000H for the generic ones), yet my Sound Blaster 16 (with a MIDI daughterboard) can still be detected and initialized by UNISOUND and I get proper audio in games. I wonder what exactly the generic address decoders on the Fintek bridge were meant for.

On the other hand, I checked the datasheets about Winbond's W83626F, and I noted its pinout is almost identical to that of F85226, except it needs both 5V and 3.3V (whereas Fintek only needs 3.3V). However, Winbond's datasheet has a rather vague documentation about its configuration registers. While W83626F uses the same routine (26H to 4E twice) to activate these registers, the registers don't appear to be 100% identical to those of F85226, and I'm not sure how those CRxx in the document translate into actual addresses (that actually needs to be written to 4E). Winbond also only mentioned two sets of address decoders compared to Fintek's four, though given my previous tests, it's probably not required, either.

rasteri wrote on 2023-03-29, 15:46:

Managed to mod a Gigabyte GA-Z77-D3H, it works really well. LDRQ was pretty easy to recover, there was an unpopulated pullup resistor under the PCH heatsink. I just brought it out to a pin header.

Awesome to see this. I was thinking of doing similar to a Z68A-D3H-B3 (Not sure if it's possible). Is there a difference between the Fintek F85226FG and the F85226A or can they be used interchangeably?

this is great, i can finally use my old foxxconn board that i have lying around for a long time equipped with the TPM pin headers.

Stevogamer wrote on 2023-03-30, 03:59:

rasteri wrote on 2023-03-29, 15:46:

Managed to mod a Gigabyte GA-Z77-D3H, it works really well. LDRQ was pretty easy to recover, there was an unpopulated pullup resistor under the PCH heatsink. I just brought it out to a pin header.

Awesome to see this. I was thinking of doing similar to a Z68A-D3H-B3 (Not sure if it's possible). Is there a difference between the Fintek F85226FG and the F85226A or can they be used interchangeably?

I couldn't find any important difference between F85226FG and F85226A(F) from their respective datasheets.

I think these should be interchangeable, though I wonder which one is newer, and whether such ICs are still being produced...

I remember asking this in another thread...

Has anyone who is playing around with these investigated whether any of the other address decoders ( RTC MC KBC ) can be re purposed to allow a few more port ranges ?