[Coco] Help with Nitros9 on SDC?
don at dgb3.net
Fri Apr 9 22:24:43 EDT 2021
Hi folks, hoping someone here can help me out. I first taught myself
basic on a coco3 circa 1989 at 9 years old, but lost it later after I
got out on my own and my parents moved a few times. Since then I've had
a few decades more of experience, including a CS degree and 25+ years of
IT and development work, and I've picked up another coco3...and its even
more fun now that I have a clue whats going on under the covers!
However, I'm trying to get NitrOS9 going with a new cocosdc, but I'm
having issues, and hope someone can help point me in the right direction.
I've grabbed the latest 6809 coco3 zip (from 2014) from
I've extracted nos96809l2v030300coco3_cocosdc.dsk, renamed to nos9.dsk
and placed it on the sd card.
I start up sdc-dos, and enter:
and nos9 boots, seemingly fine. However, before long if I try to copy
anything new to the drive, or create any sizable files, I start getting
error 247 (seek error) and then following that, error 244 (read error).
The trigger can be as simple as trying to 'dcheck /d1' on another floppy
image and says:
dcheck: cannot open workfile '/DD/dcheck030'
However, when I boot to nos96809l2v030300coco3_40d_1.dsk and
nos96809l2v030300coco3_40d_2.dsk instead of the !_cocosdc.dsk image,
everything behaves as expected. When I boot to the !_40d_1.dsk, with the
!_cocosdc.dsk image mounted on drive 1, I also get error 247 (seek
error) when accessing /d1, so it seems to be a problem with accessing
the larger disk image, but not a problem with the 360k floppy images.
This makes me conclude one of the abstraction layers (nos9, the sdc rom,
something) isn't handling the larger images correctly....but I'm at a
bit of a loss on how to debug this further.
I think I covered the basic debugging: 1/ its a good fat32 filesystem;
ran "mkfs.fat -F 32 /dev/sdb1" on a linux box myself 2/confirmed cocosdc
is running the latest version (SDC-DOS 1.6) 3/the memory in my coco3 is
good (used the 512K memtest from Performance Peripherals found on
colorcomputerarchive.com and let it run for several hours) 4/Confirmed
the same result using the nos9 nightly build images 5/Tried a different
In case it helps, dcheck on the 4.5mb nos9 image shows this:
$0900 bytes in allocation map
1 sector per cluster
$004800 total sectors on media
Sector $00000A is start of root directory FD
$0013 sectors used for id, allocation map and root directory
Building allocation map work file...
dcheck: fatal read error
Nothing on the sector math jumps out to me as being wrong, so I think
nos9 is detecting it fine....does this point to a problem internal to my
cocoSDC? On Linux I'd be completely at home diving into dmesg and /proc
to figure this out, but I'm at a bit of a loss here.
Does anyone have any ideas on what I'm missing? Should I be mounting the
larger drive images differently on the cocosdc? Is there a nos9 driver
I'm missing for accessing larger images on the sdc?
More information about the Coco