The P55TVP4 is getting worked on again today. I'm going to try putting it in a Lian-Li. I haven't built an AT system in a ATX case before, but the one I have is perfect for the job, as it has room for AT power switch.
I initially started with trying to design a modification to the face, to add a stock-looking turbo button and light. I then realized the board doesn't have a turbo button, 🤣 The case I took it out of DID, so I got fooled.
I have an I/O panel printing for it, then I can start figuring what else I want to put in it
Ended up deciding to leave the PCIs for now. I don't have enough to replace them ALL, so I'll save that job for when I do. Those things are just so damn ugly...
I was putting a shorter cable on the PC speaker and realized something: I don't have the HDD tray for this case...
Oh well. I decided to remove the bracket instead. Using a 9/64" drill bit and my powered screwdriver, I was able to easily remove the 4 rivets holding it in. I still have a ton leftover from another project if I want/need to reinstall it.
At least, this is what I'm thinking. I have external optical drives, since I'm going to run a SCSI setup, it'll be no problem. I have a few Adaptec AHA-1542 laying in a drawer I can use. I was considering NO HDD at all! Use external storage ONLY, that'd be pretty fun. I REALLY want to use this big-honkin' drive!
I got some time again and continued working on the ASUS P55TVP4. It was having troubles booting from the onboard IDE. Not sure what IS (still) going on, but the problem is sudo-gone. It wouldn't boot my boot-all drive, it gives me "I/O error, replace disk". I used a Promise PCI ATA(133) controller and it booted right away.
I tried to then load the SCSI drive/CDROM to do a windows install, but THAT was also giving me "issues". Operator error - the AHA-1542CF I was trying to use can only handle less than 1G drives, so the max partition I was able to make was 1074MB (I was using a 4.3G drive). I decided to throw another blank IDE drive (QuantumCR) at it and I couldn't load the SCSI CDROM driver now... This was a lot of messing around... I removed everything and started from scratch again. Pri-Master HDD, Sec-Master CDROM, booted the w98 floppy and everything was working absolutely NORMAL now... Ugh.
Got windows 98 installed and FINALLY got to the desktop. NOW I can start adding more goodies (hopefully). I put the 1542 back in first and it loaded just fine. Enabled USB (irq11), booted fine. Installed CT2770. Still fine, but with sound. I added next AHA-2940U, then I got crash during driver/initialization. I remember this being IRQ conflict, so I removed it, restarted and checked all available IRQs. AHA1542 was on 9 (where 2940 wanted to be), so I moved it to 12. I got to say, there isn't much room for anything else now.
My configuration was starting to look pretty good, so now I move to modding the case, which is what I'm still doing now.
I started with the PSU. I replaced the fan earlier, but that b*tch was LOUD. What I did, was knock a hole in the top and install a 120MM I had laying around. This thing is going to be QUIET.
Now, I usually NEVER cut or otherwise damage any "retro" case, unless it needs fixing. This is the first time I've ever hacked up a NICE (not really) Lian-Li (PC60 variant). I'm adding ALL the ports to the empty space on the back; they were consuming most of my PCI lanes. The AHA2940 is going to be connected to external ports. I want to use this computer for testing SCSI external devices, so I want MORE options for connectors. The thing is really starting to have "AT" vibe about it.
I made a ton of progress on the P55TVP4(I'll name this computer "Boom-Box" now)
I was stuck on needing 3d printer fuel. I used the last of the white to print a bracket (and two prototypes) for the amplifier. I was able to feed the printer a bunch "scrap" white and black bits and had just enough for the I/O panel.
I slid the motherboard tray in and ran into a major clearance issue with the amplifier and the PSU fan. Pretty lame, but I had to get one of those thin Noctua fans to put on the PSU. Took a while for that, but it's installed and the thing fits now. Next, I needed to print adapters for the speakers to fit in the front, so I was stuck again, until...
I happened to find some "Wood PLA" for free; hell yeah. Printed the adapters for the 70mm speakers to 80mm fan in the wood.
Note on the W0od PLA:
WOW, it's like I actually printed wood! Very wood feeling and texture - all of it. Like pine or balsa. It was hella stringy and the strings shaved-off just like wood. The strings felt like real wood fibers. This was Inland brand "Wood PLA" PLA. I'm going to have my brother dry-out the rest of it, see if it helps the string situation.
So... I booted it and it did nothing. Just dashes on the POST card, too. Whelp, I totally forgot the ZIP drive was faulty and don't remember it being THAT faulty to the point it completely stops the show. Not until finally unplugging everything and it booting like nothing happened, did I remember, then re-assembled it (minus the ZIP). I don't have another ZIP, so I'll install a blank in the face, for now.
I have the ZIP drive open right now and don't physically see any damage. If I knew HOW it was stopping the computer, I bet it would lead me to the failed part(s). Dead-Short in the 40-pin connector, but which one? It has 5/12V still when it stops the boot. It also stops the HDD from spinning. If it's the RESET line, that goes to the main IC, which is bad news. Any ideas or should I consider it toast?
The Boom-box is working good, so I'll set it to the side for a while to bring up this: Gemlight GMB-P55IPS-V0. I've been working on it for the last couple of days and made good progress. It needed the Dallas RTC replaced, but it was soldiered it the board. I got the chip out and replaced it with a socket months ago, but have done nothing with it. The board itself was "untested" until a few days ago. I was suddenly inspired by watching a video by Necroware, going over what options are available to replace RTCs. This time, I had the parts, tools, and skills to do "the thing" so I modified a RTC for the first time.
Modifying the RTC
I tried boiling water at first, to heat it up, but it only helped take the casing off. I set the chip on a heatsink with appropriate spacing, to hold the thing and not bend the pins. I started chiseling away at it, but it was hard as rock still. I punctured the crystal at some point then changed heating methods.
Butane torch. Light touch of torch made the resin material become soft and breakable. It took no effort to remove all of it. I pulled a replacement crystal from a dead board, then used scrap capacitor legs to reposition the battery holder terminals. I glued the holder onto the chip first, then added the last touch of solder and the crystal.
I tested the board with the RTC installed, but no battery, because I wanted to know if it would remember anything after a RESET. It did not. Most of my computers can remember until power is removed, so this may be a quirk of the RTC.
The board, of course, ran POST and was looking for boot media. Very good news. I booted a drive with a fresh install of DOS/games and it ran everything just fine. I was thinking this would be the perfect time to try this CF-IDE adapter I've had laying around again.
The adapter was inside of a "project" system I bought a while back and I was excited to test it right away. The 4GB card it came with worked, I could read/write to it with a card reader. I could see the drive in the BIOS and it would get a drive letter, but it would NOT boot. Even if I started from scratch on the computer I was testing with, win98 installer would tell me my format is invalid. The DOS installer, too. It would pretend to format the drive, then tell me it couldn't.
I was going through the same process on this computer, but this time, it WORKED. What? Why?! My process is identical, the boot disk/drive is identical. I set the drive mode to LBA, on all of the tests. Windows98 successfully installed and it boots from it just fine.
It's strange seeing the old OS and no wrrrRRRrrr going on. Just like when I first got a solid state, I can still hear the drive reading in my head (it's a lot faster, though).
Now that the 4GB card works and is setup for the Gemlight, I want to install DOS on this other CF I have, a 30MB. I put it in, and the results were promising. The BIOS showed the different card and the boot disk took me to a usable space. I removed/installed a new partition then rebooted into 6.22 installer. All good, until it's completely done. It reboots and now gets stuck on boot device (after POST, system summary). It will do nothing, but still responsive (ctrl+alt+del works). This is where I'm stuck now. I might try repartitioning/reinstalling DOS. If I load into it with the boot disk, the drive is there and DOS is the only thing on it.
I'll be pausing until I can get a case for now, because I have a few more dirty to clean. A Compaq Multi-media, and a Asus P5KPL-CM in some "Ultra" case with unknown CPU.
It's some kind of Micron (with the Micronics Board) with a socket 5. It has two sticks of memory. (and a huge-A## modem and the controller for the PC Card adapter) There is a socketed Dallas clock onboard, so yay! I get to try that again! The RetroWeb Entry:
Pulled it all apart and cleaned the case last night. I'm starting my testing right now. I have done nothing to the board yet (I usually clean first). I grabbed my known-good AT power supply and a PCI video card. Here we go! *click*
Sweet! It works! Of course, it complains about the RTC. This board also forgets everything after a reboot, so I can't go further yet. I'll clean it up and fix the RTC.
Interestingly, in the past I've noticed different boards/BIOS do the floppy check differently. This board is really loud about it, with my already loud drive. Pretty satisfying boot. It got late too quickly, so I'll have to do the RTC later.
Major Jackylwrote on 2025-04-11, 17:47:I've powered on the ASUS P55TVP4 and it works!! It came with a 133 Pentium ,but I wanted to run MMX in this thing, so I installe […] Show full quote
I've powered on the ASUS P55TVP4 and it works!! It came with a 133 Pentium ,but I wanted to run MMX in this thing, so I installed one and it's also good.
This board has some kind of modified RTC, could that possibly cause this? 75x3 would leave me at 225, but the bus *should* be 66 and the multiplier 1.5? (the board doesn't go to 3.5...) based on the manual. BUT... The screen printing on the board defines the jumpers differently and actually HAS a 75Mhz bus setting...
Major Jackylwrote on 2025-04-11, 17:47:I've powered on the ASUS P55TVP4 and it works!! It came with a 133 Pentium ,but I wanted to run MMX in this thing, so I installe […] Show full quote
I've powered on the ASUS P55TVP4 and it works!! It came with a 133 Pentium ,but I wanted to run MMX in this thing, so I installed one and it's also good.
This board has some kind of modified RTC, could that possibly cause this? 75x3 would leave me at 225, but the bus *should* be 66 and the multiplier 1.5? (the board doesn't go to 3.5...) based on the manual. BUT... The screen printing on the board defines the jumpers differently and actually HAS a 75Mhz bus setting...
CPU Core is 2.78V and I/O is 3.32V, so those are correct. Multiplier jumpers are both 1-2, 1-2.
Don't know if I'm too late with this, but;
There is no 3.5x setting on jumpers - the 233 MMX (and IIRC also some 200 and 166 MMX) interprets the 1.5x setting as 3.5
I wonder if the FSB jumpers are being misinterpreted, as it seems 66 and 75 FSB are the excact opposite setting?
I was starting to assume, since the manual wanted it at 1.5x. SpeedSys reports the correct information (so does everest). I wonder what the actual clock was when the BIOS was reporting 16Mhz? I will get back to find out.
I only recently (finally) extracted speedsys and put it on a floppy, so I don't have a lot of results to compare yet.
Finishing up re-assembling the Micron (since it works fine). I painted the cover for the floppy drive, because it looked like hell and was rusty. Haven't tested it yet, but it is a Teac FD-235HF and those seem to be not only common, but bullet-proof.
Looks good, so I'll clean it up and look closer before reassembling. I think the whole computer can go back together at this point, leaving only the RTC modification.
I have all the parts cleaned (but untested). The thing had MSI MS-6712 (aka. KT4V) with an AMD Athlon AXDA2400DKV3C and three sticks of 512MB PC3200. The optical drives are both bad, but the floppy works (Teac). The case is, I'm guessing, original to the build! It has MSI sticker/AXP sticker and the MSI USB/diag panel. I don't want to use it, though, as it is chunky AND has rust all over. I can fix the rust, but I don't feel like doing that type of work anytime soon.
I'll probably be trying to put it in this AOpen case:
I'll hopefully be able to test the board today, so can even know if I need to prep a case for it. All the cases I have piled up need service: ugly and dirty.
I replaced the crystal on the RTC I needed for the M54Hi, but it still did nothing... Gah. After sticking in the diag card and seeing no signs of life, I was starting to think this is why it was "broken", but no. I tested the RTC in my Gemlight and it worked, so I was able to rule that out. Then, I unplugged everything, and it came back to life. I came to a diagnosis of bad IDE cable on secondary channel. Crazy how that stops the whole show. I bet the crystal I replaced is fine still. Testable? I'll find out soon.
It works again and is now fully assembled and ready to add to the collection. Going to play with it a bit more first, add more goodies. It only has a 1.6GB HDD and DOS 6.22. I want to leave it as a fairly speedy DOS system.
I "accidently" completely built the MSI board into the AOpen case without testing it first. Looks great, though... Just need a few more parts and I'll see how do.
