Kahenraz wrote on 2022-02-13, 20:37:

Look for broken traces or missing SMD capacitors. This is the most common failure of vintage computer parts if it's not a busted capacitor.

There are no SMD capacitors on that motherboard. Though it's possible that I messed up the traces when I was soldering some of the capacitors, I was having a difficult time opening up one of the capacitor pads to the point where I felt like I might have broken its trace. I was doing the recap in two different times on different parts of the board. First I've replaced the capacitors that were known to be bad (most of them were bulged) a few days ago, didn't fix it. Then I've replaced every capacitor of the same brand last night (this was when I was having trouble opening up one of the pads), still didn't make a difference. I think this was a dead motherboard to begin with.

In the last 2 weeks, I finished up a RGB2HDMI unit, which is a pi zero hat that can take in oldschool rgb (you can also solder dedicate it for analog video signals) and output HDMI. I followed Adrian's digital basement footsteps to set this all up with a Macintosh Classic 1.

This week, I got a cheapo USB hdmi capture device (that can reach 1920x1080p @ 30 Hz) which is more than sufficient for capturing gameplay, but not playing off of, so I got...

a powered HDMI 1in2out splitter so that the feed also goes in real time on a 2nd monitor.

I thus have a direct video + audio feed STRAIGHT from the BARE METAL of a Mac Classic, which I could be using for streaming.

I revamped a youtube channel which I call:

'1Bit Fever Dreams'

heh.

I took a few recording snippets:

Dark Castle part 1 (I have the shield, getting the fireballs, attempting the last section):

https://youtu.be/XoniJ9ftmoA

Shufflepuck Café (Biff destroys me)

https://youtu.be/3yZB6Woe1kU

More ShufflePuck Café (Eneg destroys me)
https://www.youtube.com/watch?v=1IiDuwjxtT8

Déjà Vu from Rubicon Publishing (not to be confused with the adventure game from ICOM simulations)
https://www.youtube.com/watch?v=zGeQf3Ilzu0

Crystal Quest (reached a new high score!)
https://www.youtube.com/watch?v=bR1sfqhJMdE

Tested the 4 outlets: 2 in my bedroom (both work) and 2 in the bathroom (open ground) since the hallway, bathroom, basement (stairs), and my room lights are flickering constantly.

Going to look in the switch areas and the outlets to look for any breaks in the connections (I turned the fuse off for safety reasons) and get that issue fixed.

First 3D-printed spare part I've ever made 😀 Repair part for my IBM monitor control door hinge. I only replaced left side, but right side might need some attention too. It is cracked, but still hanging in there, so I left it alone this time.

Glued in place.

Overview. I think the door lines up quite nicely. Gaps are even.

Door open. I considered painting the part, but decided not to. It would just get scratched and look worse.

Looks really good, I would be very happy with that outcome.

CrFr wrote on 2022-02-14, 08:57:

First 3D-printed spare part I've ever made 😀 Repair part for my IBM monitor control door hinge. I only replaced left side, but right side might need some attention too. It is cracked, but still hanging in there, so I left it alone this time.

looks good, this is a great way of replacing small parts like this, 3d printing forms part of the future of vintage

CrFr wrote on 2022-02-14, 08:57:

First 3D-printed spare part I've ever made 😀 Repair part for my IBM monitor control door hinge.

Great job, dude! 😃👍
3D printing is just so neat!

I had this ES688 card that I had marked as faulty a few years ago - apparently there was no audio on right channel. So I took it today, reflowed the solder joints on every component from the ES688 to the amp to the output, and now I have clean sound on both channels. Two thumbs up.

appiah4 wrote on 2022-02-16, 06:41:

I had this ES688 card that I had marked as faulty a few years ago - apparently there was no audio on right channel. So I took it today, reflowed the solder joints on every component from the ES688 to the amp to the output, and now I have clean sound on both channels. Two thumbs up.

Always had a soft spot for ESS sound cards, my first real sound card back in the day was a ISA ESS 1868 and then a little later a nice ESS 1938s Solo-1, always found they were amazingly compatible and the Solo-1 just works as a PCI Soundblaster in DOS with little fuss.

Nice to see this 688 resurrected and hopefully off to serve a few more years in another retro system !

Did some work on my Tualatin build: swapped the 4400Ti and the TNT2 M64 cards around and disabled a couple of AGP items since they might've been the root cause of the boot-up issues where it'll lock up after the Windows 98 logo.

Will reinstall the 4400Ti and V2 card to see if that issue continues or if it's resolved completely.

Stuck some copper heatsinks on my SDRAM, just because! It isn't required but I felt they looked naked, since every other ram kit I own has heatsinks on

http://www.imgur.com/a/eFvYNfq

A succesful evening. Removed the leaking battery from a 386-board I unexpectedly got last week. Fortunately there were no signs of critical damage on the board and indeed it's still working 😀
My second PC was a 386/40 so I can rebuild my system from back then (or at least a similar one)

Ah good, you did actually get it, and it's not badly affected by it's battery vomiting. I want all the 386-40 mini boards.... All. Of. Them. 🤣

Good save, my friend. Got lucky on a couple of boards and the rest, not so much.

My Colt PC suffered from battery leakage and it worked for a couple of days until the system decided to crash half way through the stream.

Still need to fix it.

HanSolo wrote on 2022-02-17, 00:13:

A succesful evening. Removed the leaking battery from a 386-board I unexpectedly got last week. Fortunately there were no signs of critical damage on the board and indeed it's still working 😀

Nice. Maybe the board was stored upside down and none of the acid made it onto the board. Who knows? Thank you gravity. 😀

Troubleshooting weirdness with a PLE133% + Celeron 1300 build. Whenever I run a doom timedemo it crashes out immediately with an s_sound.c error. Whenever I try a quake timedemo, it completes, but when I try it again, it either crashes out or locks up. Also, when I play Doom, I get weird visual artifacts. Any ideas? I'll try downgrading to a Celeron 533 and see what that does. I tried memtest86+ on the 256MB PC-100 stick for half an hour and got no errors, so RAM is not the issue.. The system does reserve 8MB for onboard VGA (which I am not using, but can't disable because shitty BIOS) so I am wondering if RAM allocation is somehow weird?

Note: Using shareware versions of Doom and Quake from Phil's benchmark pack.

On other board/CPU combos, I might think that was a cache thing. Timing problems likely if nothing had gone bad... but all the timings are baked in here.. But, I remember the Celeron Tualatin has slightly different cache arrangements than the Tualatin PIII and released a little later... so could it be possible you have a BIOS with tualatin support, but not specifically celeron tualatin and it's handling the cache wrong???

Tested a health monitoring software on an NT Workstation and NT Server VM to see how they would perform on the OSes and the health monitoring looks really nice.

Sadly, the DMI software for Windows 95 wouldn't open as it would crash after the splash screen and I think that has something to do with the VM vs real hardware, so it'll get tested on my Tualatin build that has an Award BIOS.

BitWrangler wrote on 2022-02-17, 14:01:

On other board/CPU combos, I might think that was a cache thing. Timing problems likely if nothing had gone bad... but all the timings are baked in here.. But, I remember the Celeron Tualatin has slightly different cache arrangements than the Tualatin PIII and released a little later... so could it be possible you have a BIOS with tualatin support, but not specifically celeron tualatin and it's handling the cache wrong???

It's not that, for sure. The board refused to boot up with a Mendocino Celeron (known working), a VIA C3 (possibly faulty?) or a Coppermine Celeron (also possibly faulty?) but it did work with a Coppermine P3 800. However, the problems remained. It is also incredibly picky about RAM it accepts. Regardless, it's a very flaky, and frustrating board.

One thing I noticed is that almost all problems happen in games that run in protected mode or with their own memory managers of some kind, ie. Doom, Quake etc. I am guessing the BIOS on this board does something wonkny with memory management, probably due to reserving memory for shared VRAM and AGP Aperture, that messes with memory management for these titles. I've come to the conclusion that PLE133T is a way too advanced platform to build a DOS PC around, so I am going to try my luck with an MVP4 board instead.

Fingers crossed..

Spent the afternoon cleaning and then repasting a GeForce 4 TI4600 128mb with a new cooler, however the GPU has what looks like a bad memory chip(s) on it, large square patches of corruption on an otherwise clear and viewable picture. Perhaps I need to reflow the memory BGA chips with a hot air gun or replace them. Unless another member has some insight as to other possibilities for failed components.

Either way the GPU is going back into storage till I have the time and patience to reflow it.

Shame too as its a nice looking GPU.

Knowing me ill grab another Ti4600 and this one will stay in storage 🤣.