Re: Question about modifying DOS 3.3 or ProDOS

["Jayson Smith" <ihatespamratguynospamplease@insightbb.spamsucks.com>]

On a somewhat related topic, I've always hated the way DOS 3.3 goes 
"BZZZZZZZZZZZ!" when recalibrating the drive after a read/write error. Is 
there any way to shorten the time DOS moves the head toward track 0, so as 
to possibly eliminate the buzzing in some circumstances? I much prefer the 
way ProDos does this, going back a few tracks instead of making absolutely 
sure the head is at track 0, and probably going way too far overboard in 
most situations.
Jayson

"schmidtd" <schmidtd@my-deja.com> wrote in message 
d5505532-6024-4588-9396-723cb5b09e00@22g2000yqr.googlegroups.com">news:d5505532-6024-4588-9396-723cb5b09e00@22g2000yqr.googlegroups.com...
On Jan 8, 12:27 am, BLuRry <brendan.rob...@gmail.com> wrote:
> On Jan 7, 11:16 pm, bluebird <mildst...@gmail.com> wrote:
>
> > To use serial I/O instead of disk II (I know which is already made.) I
> > want to know how to make modifed dos image.
> [...]
> The trick is to replace the Boot 0 and Boot 1 loaders with a custom
> serial approach and look at each sector being transferred to see if it
> is a RWTS routine. If so, replace with your own with the same entry
> points. You might get more traction building off of the AGS codebase
> rather than starting from scratch.

That way, you could do DavidDOS or DiversiDOS or DOS 3.3 or whatever
over the serial line, kinda automatically: replace the RWTS as it gets
loaded.  But is that what bluebird is really after?  Would it be
sufficient to just serve one already-patched DOS?  I.e. is the
flexibility of patching any arbitrary DOS really a requirement?