PC Hoarder Patrol wrote on 2021-12-02, 12:22:

Another old timer saved from the bad fire - finally got round to recapping my old Abit BE6 (Caps Replacements). Only replaced the main Tayeh caps FTM with Panasonic FR series, but did get some others in just in case the rest start to show. Board now working & stable to 1GHz 😀

Abit BE6 Recap.jpg

I had a BE6-II, it was the most unstable POS Slot-1 board I ever used to be honest. When I got it all caps were shot, and even after a recap (to be fair I did use cheapo caps for the job..) it was flaky as hell. It frustrated me so much that I simply junked it.

Cleared out some of the junk from my Versa parts stash after attempted fixes.

- Ultralite Motherboard (too much leaky cap and seafood damage)
- screen parts (frames, glass, etc)
- a few other bits

Ordered another "V/50" (actually a 40EC) for parts. If I cant fix the real V/50 ill build a 2nd 40EC so me and the wife are on the same hardware when LAN gaming (40EC, 20MB RAM, Cisco 802.11b wifi, dos 7, wfw311). Ill put the marble screen on this one and give it to her (nicer) and Ill put the other on mine, then sell the V50 for parts (lots of good parts). Then its DOOM/Duke3d/Diablo time.

Fabricated this 60mm Noctua fan with 60-50mm adaptor on a Pentium 60.

I'd just figure out a socket 4 mount for an AX7 if I figured out a P5 clockchip mod to 100Mhz 🤣

PD2JK wrote on 2021-12-02, 21:09:

Fabricated this 60mm Noctua fan with 60-50mm adaptor on a Pentium 60.

What are you planning on doing to that poor p60?

Thermalwrong wrote on 2021-12-02, 00:57:

BetaC wrote on 2021-12-01, 08:05:

Today I had my first "I regret this hobby" day. It started with attempting the circumvent the 504MB limit imposed by my BIOS by using EZ-DRIVE. Sure, everything worked up until the program decided that it should just boot to the HDD despite being told to boot to the Floppy drive, so I couldn't install DOS 6.22. After that, I went and picked up a completely different brand of mouse, and again managed to somehow get one that can be detected by CTmouse and windows, but can't move. At all. Even when directly rolling the rollers, there's zero movement. I am at a total loss as to how I can fix the issue, as both my VLB and ISA IO cards have the exact same problem with both mice. I also can't google the issue, as I only get completely unrelated questions.

Hmm, is that serial or PS/2? check the voltages on the serial port / isa pins. I seem to recall on this forum that people have had similar problems and it turned out to be corrosion that had damaged the -12v line which disrupted serial comms.
If it's getting as far as detecting, the pinout should all be good at least.

Yeah, at this point I think it’s just the motherboard. The mice work on other computers without issue. The computer was also having trouble detecting the IDE CD drive that had zero issues using the CD driver from windows 95, so I said screw it. The system is once again a pentium 100 on a case that’s Badges Pentium 75

BetaC wrote on 2021-12-03, 02:27:

Thermalwrong wrote on 2021-12-02, 00:57:

BetaC wrote on 2021-12-01, 08:05:

Today I had my first "I regret this hobby" day. It started with attempting the circumvent the 504MB limit imposed by my BIOS by using EZ-DRIVE. Sure, everything worked up until the program decided that it should just boot to the HDD despite being told to boot to the Floppy drive, so I couldn't install DOS 6.22. After that, I went and picked up a completely different brand of mouse, and again managed to somehow get one that can be detected by CTmouse and windows, but can't move. At all. Even when directly rolling the rollers, there's zero movement. I am at a total loss as to how I can fix the issue, as both my VLB and ISA IO cards have the exact same problem with both mice. I also can't google the issue, as I only get completely unrelated questions.

Hmm, is that serial or PS/2? check the voltages on the serial port / isa pins. I seem to recall on this forum that people have had similar problems and it turned out to be corrosion that had damaged the -12v line which disrupted serial comms.
If it's getting as far as detecting, the pinout should all be good at least.

Yeah, at this point I think it’s just the motherboard. The mice work on other computers without issue. The computer was also having trouble detecting the IDE CD drive that had zero issues using the CD driver from windows 95, so I said screw it. The system is once again a pentium 100 on a case that’s Badges Pentium 75

Is this a propitiatory OEM motherboard? Iv had more issues with those then iv had anywhere else.

Shagittarius wrote on 2021-12-02, 23:10:

PD2JK wrote on 2021-12-02, 21:09:

Fabricated this 60mm Noctua fan with 60-50mm adaptor on a Pentium 60.

What are you planning on doing to that poor p60?

Nothing special, just running it at 66 MHz, because why not. And I like Noctua fans. 😉

bearking wrote on 2021-12-02, 12:37:

Nice! […]
Show full quote

PC Hoarder Patrol wrote on 2021-12-02, 12:22:

Another old timer saved from the bad fire - finally got round to recapping my old Abit BE6 (Caps Replacements). Only replaced the main Tayeh caps FTM with Panasonic FR series, but did get some others in just in case the rest start to show. Board now working & stable to 1GHz 😀

Abit BE6 Recap.jpg

Nice!

I also have a BE6, it was my first slot 1 board, so it is a bit special for me. Unfortunately it is very unstable! Do you think a full recap would help?

Thanks

It's certainly possible - the instability could easily be a sign of failing / failed caps. I only changed the 6 main Tayeh 1500uF as they were visibly failed (blown + either leaking or popped base plug) and the board would no longer even POST, but many caps regularly fail with no visible sign at all.

appiah4 wrote on 2021-12-02, 12:44:

I had a BE6-II, it was the most unstable POS Slot-1 board I ever used to be honest. When I got it all caps were shot, and even after a recap (to be fair I did use cheapo caps for the job..) it was flaky as hell. It frustrated me so much that I simply junked it.

BE-II was more of an attempt by Abit to produce a performance / OC BX board than the basic BE6 so maybe that hand a hand in it (never had a BE-II but the similar MSI MS-6163Pro v2 is still one of my best BX boards)

Claris wrote on 2021-12-03, 02:44:

BetaC wrote on 2021-12-03, 02:27:

Thermalwrong wrote on 2021-12-02, 00:57:

Hmm, is that serial or PS/2? check the voltages on the serial port / isa pins. I seem to recall on this forum that people have had similar problems and it turned out to be corrosion that had damaged the -12v line which disrupted serial comms.
If it's getting as far as detecting, the pinout should all be good at least.

Yeah, at this point I think it’s just the motherboard. The mice work on other computers without issue. The computer was also having trouble detecting the IDE CD drive that had zero issues using the CD driver from windows 95, so I said screw it. The system is once again a pentium 100 on a case that’s Badges Pentium 75

Is this a propitiatory OEM motherboard? Iv had more issues with those then iv had anywhere else.

It’s a GA-486VT with an absolutely basic BIOS from 1993.

I've been getting into contributions to vintage tech wikis lately. Started adding to preterhuman, thinking about adding to the Vogons wiki. Mostly hardware info.

Shot a very candid video on the Crystal CS-4231-KQ last night (NEC Versa M/75). I think my next sort of "when I can" project will be figuring out ways to get that crazy monster working in DOS. Showed off some games that work, some that should but don't (7th Guest and AD Cop puking onto the DOS prompt from setup) and explaining what I know. Seems theres not enough WSS codec stuff out there. Not sure if other laptops have the same settings but maybe it'll help someone with an M/75 at least.

Today I got everything back up and running with my P5 system, and I am wondering if I really even need L2 cache on a completely DOS Pentium system.

BetaC wrote on 2021-12-03, 23:32:

Today I got everything back up and running with my P5 system, and I am wondering if I really even need L2 cache on a completely DOS Pentium system.

That's heresy!

*EVERY* system needs to be the absolute cutting edge it possible can be! It has to have the FASTEST P5 processor, all the memory banks need to be filled with as much ram as humanly possible! Overclock it! replace the processor with an Overdrive (tm) and overclock THAT!

*inarticulate rambling continues*

luckybob wrote on 2021-12-03, 23:35:

That's heresy! […]
Show full quote

BetaC wrote on 2021-12-03, 23:32:

Today I got everything back up and running with my P5 system, and I am wondering if I really even need L2 cache on a completely DOS Pentium system.

That's heresy!

*EVERY* system needs to be the absolute cutting edge it possible can be! It has to have the FASTEST P5 processor, all the memory banks need to be filled with as much ram as humanly possible! Overclock it! replace the processor with an Overdrive (tm) and overclock THAT!

*inarticulate rambling continues*

If only I could afford to actually do that with some of my systems, like the TI-99/4A and VIC-20. Especially the TI. Those expansion boxes are so expensive.

BetaC wrote on 2021-12-03, 23:39:

luckybob wrote on 2021-12-03, 23:35:

That's heresy! […]
Show full quote

BetaC wrote on 2021-12-03, 23:32:

Today I got everything back up and running with my P5 system, and I am wondering if I really even need L2 cache on a completely DOS Pentium system.

That's heresy!

*EVERY* system needs to be the absolute cutting edge it possible can be! It has to have the FASTEST P5 processor, all the memory banks need to be filled with as much ram as humanly possible! Overclock it! replace the processor with an Overdrive (tm) and overclock THAT!

*inarticulate rambling continues*

If only I could afford to actually do that with some of my systems, like the TI-99/4A and VIC-20. Especially the TI. Those expansion boxes are so expensive.

I use to have a TI for a little bit before i realized I didn't have much money to dedicate to another type of retro computer. Its a weird little machine but fun for sure. You really need the ram expansion and atari joystick adapter if you want the best out of it though. Last i remember there was a community made PI based replacement for the peripheral box. But i might be wrong.

Speaking of weird things. I do have a PC9821 i got stuck with. Not the biggest Touhou fan on earth, so it just kinda ....sits in the cupboard and judges me....

Cool talking piece though! Love Japanese oddities.

Claris wrote on 2021-12-03, 23:54:

I use to have a TI for a little bit before i realized I didn't have much money to dedicate to another type of retro computer. It […]
Show full quote

BetaC wrote on 2021-12-03, 23:39:

luckybob wrote on 2021-12-03, 23:35:

That's heresy!

*EVERY* system needs to be the absolute cutting edge it possible can be! It has to have the FASTEST P5 processor, all the memory banks need to be filled with as much ram as humanly possible! Overclock it! replace the processor with an Overdrive (tm) and overclock THAT!

*inarticulate rambling continues*

If only I could afford to actually do that with some of my systems, like the TI-99/4A and VIC-20. Especially the TI. Those expansion boxes are so expensive.

I use to have a TI for a little bit before i realized I didn't have much money to dedicate to another type of retro computer. Its a weird little machine but fun for sure. You really need the ram expansion and atari joystick adapter if you want the best out of it though. Last i remember there was a community made PI based replacement for the peripheral box. But i might be wrong.

Speaking of weird things. I do have a PC9821 i got stuck with. Not the biggest Touhou fan on earth, so it just kinda ....sits in the cupboard and judges me....

Cool talking piece though! Love Japanese oddities.

Awesome. The TI I got was in box, “non working” and even had an Atari adapter. Of course, the only real problem was keeping it from booting was the fuse glued to the end of the transformer. Problem for me is, though, that I have the absolute worst keyboard membrane ever produced in there, making it hard to even use the system.

Had a wild hair tonight when I woke up to grab some water....ive been on a roll with card services for awhile and learning PCMCIA stuff. What I thought was a Panasonic KXL-D20 comes up as a KXL-C101. Got the bare driver going in DOS and the Win95 driver working. Port 330 OPL3 FM is working and so is SB compatible audio on port 240h...just cant hear them....no breakout box.

Looks like next will be setting up some kind of probing setup to listen to I/O for the audio signals off each pin. Once pins are identified, it appears pin 62 (SPKR) on the card is unused. Maybe Ill patch into that. That should send audio out thru the internal speaker. Then Ill have OPL, Soundblaster, and WSS all in one laptop...still some major R&D is needed first...may ho extetnal flat speaker instead. All depends how the 62/spkr line works.

On a desktop board speaker would be a 2 pin INPUT to play the PC onboard speaker through the soundcards filter and amp and out through headphone or line out.

BitWrangler wrote on 2021-12-04, 13:54:

On a desktop board speaker would be a 2 pin INPUT to play the PC onboard speaker through the soundcards filter and amp and out through headphone or line out.

I know. I've wired up a few crazy things in my time on my vintage PC's to do things like direct all audio through the internal speaker or direct audio through other things. It's a bit of an extention of my guitar building/pedal building capabilities. How I think the speaker pin works is that audio goes to ground (as usual) on the "-" side, and the "+" side goes to the speaker. Almost all audio, for guitars at least, are wired in this way.

Actually making some headway on this this morning from the connector.....not much but I did figure this out using an external speaker...

I started by finding out the grounds of the card using a VOM - I found actually, when the card is inserted into the computer, the card becomes a part of the computer's ground plane, at least the ring around the connector does, and the casing of the PCMCIA card itself (if conductive). Basically, put one connector on the metal bracket on the back outside of the Versa (which goes to the motherboard's main ground/- side), and then put the positive onto the case on the continunity setting and got full continuity from the backplane to the card.

So I setup an alligator clip to the M/75's rear backplane, carefully wired up a test harness to a 4ohm internal speaker I had laying around, and I think I got some audio signal off the pins, but it looks like the card uses a different method. The KMI KXL-C101 has a 34-pin header on the back with the pins all on the bottom of the connector - makes this relatively easy actually.

When I posted on this earlier, someone gave me a link to gut-shots of the card on this forum. I was somewhat able to trace signals from the ES1688S to the pins on the back, the first half of the pins (facing back of card, label up up, left to right) are audio pins based on where they are going, they go through a via, to a series of SMD components, then through another via somewhere, and following a lot of those traces often lead at least toward the ESS chip for the left side of the connector. It looks like (theoretical) Pin 1 goes to ground based on the solder pad it's attached to. Also, when I looked at pin 62 on the I/O header to the PCMCIA bus, there was just a bare pad, no via, no trace, so that pin is not even used.

I know, from building my own audio stuff, and a blurry picture of the breakout box, assuming the breakout box is not doing some of the work...
- There should be roughly about 3-5 audio signals coming off the card - 2 for the left/right output channels (+), one for Line-In/Mic in, or maybe those are separate.
- If the audio is using a separate ground plane, it could be 2 pins per each I/O, one for -, one for +, that makes things a bit harder

Anyway, fired up Versa, booted up Windows 95, opened up "shop.mid" in Media Player....went to town finding signals

Pin 1 = Ground
Pin 2 = unknown - maybe 2nd ground or no sound?
Pin 3 = Sound? - Small tic when connected to - maybe mic/line in?
Pin 4 = Sound? - Small tic when connected to - maybe mic/line in?
pin 5 = Squealing noise - possible + or - audio out?, got louder when touched with speaker + wire
Pin 6 = Squealing noise - possible + or - audio out?, got louder when touched with speaker + wire
pin 7 = Squealing noise - possible + or - audio out?, got louder when touched with speaker + wire
Pin 8 = Squealing noise - possible + or - audio out?, got louder when touched with speaker + wire

Going off of this, it seems one of two things apply.....or both...

1.) The "+" and "-" for the audio signals are all hosted off the card-edge connector on the back, so would need to build my own breakout connector to figure it out
2.) I think I MAY have heard some faint sound coming through the card-edge on some pins on louder sound pins (1/2 signal), which may suggest this
3.) If the sound comes out of the quieter pins, it could be the breakout box uses +5VDC? off the PCMCIA card to drive an amplifier
4.) Worst case scenario, I'm getting raw DIGITAL data and the audio has to hit an AD/DA converter in the breakout box (then it's time to find a box then I guess, or sell the card)

Either way, neat little experiment. I think to figure out what data signals come off the card, if not needing AD/DA would mean I can find that kind of card-edge connector or something I can modify to fit (right pin spacing, I really need a micrometer) and then wire up the first eight wires to it and see if I can pull full audio signals off. either way, I'm just glad I got the card setup and detecting. I really should ad the drivers to the Driver Library that I got.

BTW, to get the card working in DOS wihtout card services, it requires ASPIS365.SYS from Disk 1 of the driver disks with the following entries, I used DEVLOAD in FreeDOS to get that one up...but should work in CONFIG.SYS.

DEVICE=C:\XXXXXXXX\ASPIS365.SYS /PORT=240 /ADLIB /INT=5

If you have EMM386.EXE loaded you also have to (as with most PCMCIA cards) exclude the memory address of the card, in this case CE00-CFFF and /MEM=XXXX to set the memory address. CE00 is the default, but you can also re-assign to a different address using EMM386 and the /MEM= variable when loading the driver.

Anyway, even if It comes to nothing for me, at least this info is out there for someone to pick up where I left off if they feel like.

Found a Q6700 so started fiddling around with this Abit board Iv been procrastinating on fixing.

Does anybody have any idea where to find a Bios for this Abit IN9 32x Max? It ALMOST works, but it locks up in the bios no matter what CPU, ram, PSU or GPU I use with it. A fresh Bios battery seemed to add about 5 seconds of time spent in the bios before

crashing.

Next order of business is a deep clean and thermal paste refresh on the chipsets.