On your 486 if you go under MSD and then (D)isk Drives, and give us the information under C: (C/H/S values according to FDPT as well as CMOS) it might help in explaining exactly how your bios is doing it.

serialShinobi wrote on 2023-06-01, 17:52:

Hello. I found out just now that the programming for accessing the disk using INT13h looks like assembly code. I don't understan […]
Show full quote

Hello. I found out just now that the programming for accessing the disk using INT13h looks like assembly code. I don't understand the topic and I wonder if anyone knows of a way to help me understand something about addressing.

Given a VM with MSDOS 6 instantiated; Does the virtual BIOS & PATA adapter address with LBA and int 13h ? Isn't this locked into a standard if using the same (int 13h) LBA routines?

Do all INT13 motherboards really have their own unique versions of INT13h? So bytes from one machine are addressed differently by another?

Can a disk image map original byte and address structure to a unpartitioned media?

I had the idea that DOS diskettes are a reliable way to deal with incompatibility between computers. The industry and developers release software in archival (read only) form. To cross beyond dissimilar disk addressing and hardware designs.

INT 13 provides BIOS level access to disk drives.
DOS uses INT 13 to access disks.
A BIOS needs a first sector on the disk containing the 55AA signature word as last 2 bytes so that it will try to start from it. Basically this sector contains boot code and includes the partition table.
BIOS' INT 13 routines basically are matched to a chipset.

Basic BIOS INT 13 supports CHS addressing scheme.
INT 13 Extensions support LBA addressing scheme and are a must for disks > 8,4 GB.*
MS-DOS up to 6.22 do not use LBA scheme and therefore are limited to 8,4 GB.

* Yes there is a LARGE addressing scheme; you can deal with it when you like headaches.

jakethompson1 wrote on 2023-06-01, 18:21:

On your 486 if you go under MSD and then (D)isk Drives, and give us the information under C: (C/H/S values according to FDPT as well as CMOS) it might help in explaining exactly how your bios is doing it.

I think about bytes and hexadecimal numbers when reading your message. Is there some viewer utility I need? DOS Debug?

A BIOS needs a first sector on the disk containing the 55AA signature word as last 2 bytes so that it will try to start from it...BIOS' INT 13 routines basically are matched to a chipset.

...MS-DOS up to 6.22 do not use LBA scheme...

Considering your advise, virtualbox has a virtual PIIX3 chipset. I will try to get the data sheet for following:

The 82371FB (PIIX) and 82371SB (PIIX3) PCI ISA IDE Xcelerators.

I'm hoping I can see if this chipset supports CHS addressing and the 55AA byte for the last two bytes of sector 1.

Thanks for offering the details about INT13h I will likely make use of them soon.

serialShinobi wrote on 2023-06-01, 17:52:

Can a disk image map original byte and address structure to a unpartitioned media?

So will a disk image prevent a drift in CHS addressing? For example I use a disk image of a machine with DOS installed. The target machine has completely zero filled media. The image from one disk is applied to the zeroed disk. Now the target machine has the same CHS values as the original machine.

Is this a good way to install DOS? That is if one thinks imaging is an option?