First post, by Demo
- Rank
- Newbie
Hey everyone, I’ve been tinkering with my IBM ValuePoint 466DX2/SP (6472-L5G), a 486DX2/66 I picked up recently. My goal was to breathe new life into it with some upgrades and get it running like it’s 1993 again—maybe even better. I’ve hit some snags along the way, so I’d love to share my progress and get your thoughts!
Base Specs:
- CPU: 486DX2/66 MHz
- RAM: 16 MB (originally, I think it shipped with less)
- Video: Onboard S3 86C864 (Vision864)
- Storage: Originally a small IDE HDD (long gone)
- BIOS: IBM ValuePoint BIOS (updated to 1998 version, ldjt80a.exe)
Upgrades:
Storage - IDE to CF Swap
- Replaced the dead HDD with an 2 GB Compact Flash card via an IDE-to-CF adapter.
- Why? Lower power, no moving parts, and easier to swap configs and OS
- Process: Set the adapter to master, formatted the CF, installed Win98lite with Win95 shell swap.
- Performance: It runs surprisingly fast, but spend most of my time in MS-DOS.
RAM Boost
- Upped the RAM from 16 MB to 64 MB (max supported, is 128mb).
- Why? More headroom for DOS games, Windows games, and because I can. (overkill I know)
- Found some compatible 72-pin SIMMs—smooth sailing here.
CPU Upgrade
- Installed an Intel PODP5V Pentium Overdrive 63Mhz
- Had to turn SL Enhanced off via J6 to get a post
- BIOS detected at 66Mhz setting FSB to 25mhz
- Windows see's it correctly as 63mhz Pentium
- Can get post and playing Doom with 33mhz FSB @ 83mhz
- I get random memory errors after long periods of time, set back to 63mhz
Sound Upgrade
- Added a Sound Blaster 16 VIBRA (CT4180) I had lying around.
- Why? That sweet sound for DOOM and Duke.
- Plugged it into an ISA slot, no conflicts so far.
CD-ROM - TEAC 52x
- Installed a TEAC 52x CD-ROM drive (model CD-552E, I think—need to double-check).
- Why? Wanted to load games and utils straight from disc—plus, it’s peak ‘90s vibes.
- Process: Connected it as master on the secondary IDE channel, paired it with MSCDEX for DOS support. Added “DEVICE=C:\DOS\OAKCDROM.SYS /D:MSCD001” to CONFIG.SYS and “MSCDEX.EXE /D:MSCD001” to AUTOEXEC.BAT. Spins up fine, reads discs like a champ—though a 52x on a 486 feels like strapping a jet engine to a bicycle!
Problems:
Hard Drive
- BIOS Drive Detection (Error 1780)The BIOS refuses to see my 8 GB CF card properly—shows as "0MB IDE."
- Tried: Updated BIOS to ldjt80a.exe (1998), checked jumpers, swapped 3 IDE cables, used Ontrack DM and EZ-Drive (both format fine but no boot).
- Result: Still get error 1780, and it won’t boot. Works fine on a newer Athlon XP rig, so the adapter’s good. What’s up with this picky ValuePoint BIOS?
- Moved on and replaced 8gb CF with a 2gb problem resolved.
CPU
- Tried a CPU Blaster with a AMD DX4 120Mhz, computer refused to boot
- Underclocked at lower speeds didn't help.
- Possibly a bad CPU, or the cpu blaster. (noted the blaster was missing it's 4th pin for x4 for 5x86 133mhz, I can easily bridge it with a bit of solder)
- The CPU blaster I used is used old stock from the mid 90's (image below)
- Resolved with Pentium Overdrive 63Mhz
VIDEO
- Attempted to add a Cirrus Logic CL-GD5424 VLB
- Disabled the onboard video via J32
- Card at first refused to post, sprayed the vlb slot with a bit of deoxit, I got a post.
- I then started getting random glitches on screen, sometimes at post in the blue in the IBM logo.
- Windows would sometimes have random glitches but seemed to go away after idling on desktop.
- Updated drivers, cleaned the card thoroughly didn't help, went back to onboard S3.
Mistakes were made.
Here’s where I made a significant mistake. I decided I wanted to upgrade the onboard S3 memory from 1MB to 2MB. After searching eBay, I found what I believed to be the correct memory: "UPD424260LE-60 NEC DRAM Chip FPM 4Mbit 256Kx16 5V 40-Pin." Both my local LLM and Grok reassured me that this was the right memory to upgrade my onboard S3. The chips arrived quickly in the mail, and upon close inspection, I noticed that my onboard video memory was labeled "42S460-6OR." To be certain, I double-checked with my highly intelligent AI, which confirmed that the new chips would indeed work.
I carefully installed the chips, ensuring that I aligned pin 1 correctly and paid close attention to their proper orientation. To my surprise, the computer successfully powered on and booted into Windows! However, my excitement was short-lived. When I moved the mouse toward the Start menu to run DxDiag, the mouse refused to respond. I thought I could use the keyboard instead, suspecting that I might have bumped the system and dislodged the mouse from its port. But to my astonishment, the keyboard didn’t work either. What was going on?
I pressed the power button to shut down the computer and decided my best option was to remove the chips. After doing so, I restarted the system and noticed something strange: as long as I didn’t move the mouse or click any buttons, the keyboard functioned perfectly. However, the moment I interacted with the mouse, the keyboard stopped responding. This issue persisted whether I was in DOS or Windows.
So here is what I've tried:
- CMOS reset F5 to defaults and removed battery(didn't work)
- Tried a different Mouse (didn't work)
- Tried a different Keyboard (didn't work)
I’m assuming this was either an unfortunate coincidence or the memory I purchased was incorrect and somehow damaged my PS/2 port for the mouse. I’ve carefully inspected the motherboard for any obvious signs of damage that might have caused this, but I couldn’t find anything. At this point, my last resort seems to be accepting that I’ve ruined my PS/2 port and trying a serial mouse instead.
If anyone has encountered this issue before and has insight into what’s happening, I’d greatly appreciate it. Feel free to also point out how reckless I was for impulsively installing the memory chips without more caution.
The Retro Web Specs for The IBM PS/VALUEPOINT 6472
https://theretroweb.com/motherboards/s/ibm-ps … 472,-6482,-6484