[net.micro.pc] converting from 2.X to 3.X

markz@microsoft.UUCP (Mark Zbikowski) (08/05/85)

>/* Written  6:47 pm  Jul 30, 1985 by slerner@sesame in ima:net.micro.pc */
>>> This worked fine for me on many machines.  Takes about 5 minutes.  Note
>>> that you cannot upgrade from 2.x this way because the space reserved for
>>> the operating system on the hard disk isn't enough.
>>
>>Sorry to disagree, but I just upgraded my XT from 2.0 to 3.1.  I suspect
>>that your problem had to do with trying to SYS 3.1 onto a 2.X floppy.
>>There is plenty of spare room alocated on the hard disk to sys 3.1 on,
>>(at least there was on my system, I haven't gone peeking yet to see
>>what really went on, and am not is a position to do so now.
>>Simcha-Yitzchak Lerner
>
>You were lucky - in general you can't count on there being enough room
>free at the start of a 2.x hard disk.  It's probably worth a try, though.
>
>John Levine, ima!johnl


John is correct.  The requirement for correctly "SYS"ing a new operating
system is the following:

o   IBMBIO.COM and IBMDOS.COM are the first two directory entries on the disk.

o   IBMBIO.COM is allocated contiguously beginning with the first cluster on
	disk.

These two restrictions occur because there is not enough room in the 512-byte
boot sector for the logic to walk the FAT.

John is referring to the fact that is it possible to have IBMBIO for 2.X
at the beginning of the disk and IBMDOS scattered across the disk.  When
3.X is SYS'ed, the larger BIOS might not fit within the previously allocated
space.

Note that this is an unlikely condition.  Most people will SYS to their hard
disk when they first get it;  well in advance of putting any files on the disk.
This means that IBMDOS and IBMBIO are continuous.  It is unlikely that IBMBIO
will grow beyond their combined size.

bc@cyb-eng.UUCP (Bill Crews) (08/10/85)

> Note that this is an unlikely condition.  Most people will SYS to their hard
> disk when they first get it;  well in advance of putting any files on the
> disk.
> This means that IBMDOS and IBMBIO are continuous.  It is unlikely that IBMBIO
> will grow beyond their combined size.

What about those of us who are used to using the /b switch on format to
distribute our software without also distributing DOS?  DOS 3.x users will
not be able to SYS their DOS onto the diskettes to make them bootable.  Am I
missing something?
-- 

  /  \    Bill Crews
 ( bc )   Cyb Systems, Inc
  \__/    Austin, Texas

[ gatech | ihnp4 | nbires | seismo | ucb-vax ] ! ut-sally ! cyb-eng ! bc