I meant flash chip. Doh.

In reply to a post by Banger:

I meant flash chip. Doh.

Yep, the UEFI is likely on a 2Mbit or larger flash, whereas BIOS back to the original IBM PC (1981) was I think 64 kilobytes. UEFI is an operating system. BIOS was a tiny program. World has moved on.

On the 5150 the BIOS was an 8 KB EEPROM. You can do a lot in 8KB.

And the 5150 manuals contained a full listing of the BIOS.

Edited by TinyMongomery (Fri 27-Sep-24 21:51:07)

8K ! Thank you.

I remember learning that Compaq was the first company to recreate the BIOS without copying the IBM code. Hard to prove given the IBM code was published!

Publishing the code made it harder for others to clone it due to copyright. There’s a discussion about it here: https://retrocomputing.stackexchange.com/questions/1...

I won’t provide a link as I’m not sure of it’s legality, but if you Google “IBM 5150 Technical Reference Manual” you can find a scan of it online.

Edited by TinyMongomery (Sat 28-Sep-24 06:38:38)

In reply to a post by TinyMongomery:

Publishing the code made it harder for others to clone it due to copyright.

Makes sense.

In reply to a post by jchamier:

In reply to a post by Banger:

I meant flash chip. Doh.

Yep, the UEFI is likely on a 2Mbit or larger flash, whereas BIOS back to the original IBM PC (1981) was I think 64 kilobytes. UEFI is an operating system. BIOS was a tiny program. World has moved on.

I remember Compaq had the bios on the hard drive, but they must have had something to recognise the hard drive in the first place, so I presume a tiny bit of code in ROM and then the main bios on the drive. I remember it was a pain in the neck and I never wanted to work on a Compaq again,

And now we have gone past that stage of the BIOS, I suppose there was no choice as hard drives got larger.

Well, I tried my PC again today and it turned on, and I played a quick game, so it is still working. I will wait until next year and decide what I am going to do unless my POC thinks otherwise

The BIOS always has to be in ROM of some description. Apart from anything else you need basic routines to provide console I/O to configure the BIOS. And, of course, you need the code to read from external storage. The original IBM 5150 could operate without any disks at all as it had a ROM containing BASIC and cassette storage routines in addition to the BIOS. So it could run without any disk-based operating systems.

I wonder if you are thinking about the DOS low-level routines (“interrupts”) that are, indeed, loaded from disk? All PCs running DOS worked that way. The BIOS provided a very limited set of commands; most useful low-level routines were from IO.SYS.

In reply to a post by zyborg47:

I remember Compaq had the bios on the hard drive, but they must have had something to recognise the hard drive in the first place, so I presume a tiny bit of code in ROM and then the main bios on the drive. I remember it was a pain in the neck and I never wanted to work on a Compaq again,

Not sure what that was you were working on, but it wasn't a PC. The BIOS has always been in EPROM or Flash on the system board/motherboard, otherwise you can't boot the operating system on the disk.

The original 5150 IBM PC also had BASIC in ROM, so if you didn't insert a floppy, it would enter ROM BASIC, also a Microsoft product. The hard disk versions of the PC (PC/XT) and the 80286 chip version (PC/AT) also worked in the same way.

Compaq was first with the 386 chip, and the industry dislike the IBM PS/2 models (which in the UK were sold in Dixons in high streets) as they used the incompatible MCA expansion card; so the "clones" profited, and the famous Amstrad models appeared (PC1512, PC1640).