Recent content by Makefile

Unicomp still sells a "Classic 101" with no Windows keys, which is as close to a new production model M as you can get. I have been using them for years. They aren't just an imitation; I believe they are even made on the same machinery/molds (IBM Lexington -> Lexmark -> Unicomp). I agree that...

Possibly. It could be that, like Unix, nothing echoes as you type a password. And further, if it is reading random junk since the Dallas has 50 bytes, the password checking algorithm could be going awry. Some Intel motherboards of that era are also known for having BIOS Setup programs that lock...

It makes more sense to put your hard drives on the PCI connector and the ATAPI CD-ROM on the ISA. The ISA bus can sustain about 16X CD or so before it's saturated, and any faster is probably louder than you are willing to tolerate anyway. Also if you run Windows 3.x, the 32-bit disk access...

And of course int86() would be like that for the 8086. But I guess you can always construct a stub of code on the stack to avoid generating an inconsistency between in-memory code versus the on-disk binary, if that's what one considers "self-modifying code."

https://devblogs.microsoft.com/oldnewthing/20180209-00/?p=97995

I used to have one of those, but I don't any more. It is an LPX form factor system with PCI and ISA slots on the riser card, right? And I believe an Intel 420 chipset of one variant or another. The onboard video is the Cirrus 543x PCI chip they mentioned in the screenshot. Those have drivers in...

Should be about $1 apiece on digikey or mouser

Indeed it is soldered in. I used Littelfuse 0251005.NRT1L (FUSE BRD MNT 5A 125VAC/VDC AXIAL) which I believe is an exact match for the original.

Not sure about the keyboard problem or Norton Commander, but I believe 5150 cassette is in the release build now: https://glabios.org/

You might also try looking in diskmgmt.msc. If they aren't partitioned, or none of the partitions are Windows-compatible, they might not get assigned any drive letters.

I have never had any compatibility problem that was helped by adding USB to the mix. It's probably because the converter is hardcoded to use LBA addressing when making ATA commands to the drives, while the drives only accept CHS addressing. Also, it likely relies on ATA IDENTIFY to gather...

Your write-back external cache allows cache and RAM to go out-of-sync, since writing to cache is faster than writing to RAM. In contrast, a write-through cache (including your 486DX internal cache) always updates RAM immediately and doesn't allow them to get out of sync. When the cache line is...

It's probably for a x1 or x4 "Dirty RAM" for the external cache, with a pull up resistor present so it operates as "Always Dirty" if it's missing.

That's what I do, but it looks like the motherboard above is super wide--even wider than a 486 VLB board--and some ATX cases have a little indentation in the motherboard tray that would block you from installing a motherboard this wide. If it's any consolation, many baby AT cases you might find...

Also one of the other "basic configurations" for LPT1 may include one that doesn't use an IRQ. Because parallel port polling mode should only poll when the port is being actively used (e.g., during a print job) it may not be as bad as it sounds.