[Coco] OS-9 loading NEW WAY

[Gene Heskett]

On Tuesday 21 April 2015 03:18:53 K. Pruitt wrote:
> ----- Original Message -----
> From: "Stephen H. Fischer" <SFischer1 at Mindspring.com>
> To: "CoCoList for Color Computer Enthusiasts" <coco at maltedmedia.com>
> Sent: Monday, April 20, 2015 11:51 PM
> Subject: Re: [Coco] OS-9 loading NEW WAY
>
> > No,
> >
> > It was a replacement for the DECB "DOS" command, track 34 and
> > OS9boot files.
> >
> > It likely fell on deaf ears and no kind words were said about it. I
> > hear a lot of that.
> >
> > SHF
>
> Yeah I saw Tormod's message.  I was unaware of the CocoBoot project. 
> I've been looking over the site and I am really impressed with this
> project.  It is now bookmarked.
>
> Good work, Brett!

Some good ideas there I believe.  The partition table thing intrigues me 
a bit, but its placement in LSN0, displacing the info there that Nitros9 
uses pretty well wrecks Nitros9.  There is however, over half of that 
sector that presently has no use defined, and I see no reason not to 
make use of that area of LSN0 for a partition table.  Doing so could be 
useful to all levels of Nitros9.  We have a chd/cd, and a chx/dx 
commands, why not a cht/ct (chp/cp clashes with one of our copy 
utilities) command.

I can also see a Q&D basic program to read that area, and poke the 3 
values into HDBDOS to "switch disk banks/partitions" making this space 
100% available to HDBDOS users as is shown for Nitros9 below.

One statement on Bretts cocoboot site is erronious however.  When he 
states that nitros9 cannot use more that 131 megs of a disk, that is 
assuming that the cluster size=sector size, or 1.

This can be expanded in powers of 2, to address up to a maximum of about 
4Gigabytes, and I am currently running 2 seagate 1Gb scsi drives on my 
machine, the first of which uses a cluster size of 4 to address the 
first, nearly 500 megabytes (498,831,360 bytes TBE) as an 
os9 "partition" with the remainder of the disk available as a series of 
256 disk hdbdos virtual disks, each of which occupies an additional 
82,575,360 bytes for each of those 256 virtual disks. It would be half 
that but HDBDOS does not "deblock" on a 512 byte sector drive. So with 
that 500+ megs of space, about 6 more banks of 256 vdisks each, or 1,792 
of these could be fitted in.  I dare say that would be sufficient space 
to store every 35 track SS distribution disk each of us has ever had!

There is however the potential for a drive loss and losing it all too.  I 
don't relish that thought personally.  The all eggs in one basket leads 
to a bit of paranoia, I suspect with all of us ...

That 2nd 1Gb seagate drive?  Whole drive is for os9/nitros9, so 
the "cluster" size there is $10.

Point is, its doable.  It is up to US to do it.  But getting everyone of 
us on the same page to do it is like herding cats.  FIRST we need an 
agreed on set of defines, well thought out, and which includes the 
cluster size on a per partition basis.  And for nitros9's vdisk as we 
can read/write them using the /sh descriptor functions, a method to 
define 2 more vdisk sizes, a 40 trk DS, and an 80 trk DS, giving us the 
ability to backup those two floppy disk formats.

Without that foundation set of LSN0 defines, we are a committee meeting 
with no rules of order and Nitros9 at least would be destroyed from 
within.  I do not relish that thought either.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>