Chris Tribo wrote:
> 
> on 7/26/01 2:29 AM, Dave Burgess at burgess@cynjut.neonramp.com wrote:
> 
> > If only I had a "PC" capable disk image that I could drop onto a floppy, I'd
> > be in pretty good shape.
> 
>     Find an FTP program that understand MacBinary and you could do it.
> 

I don't have a Mac, which is part of the problem.  The other part of 
the problem is that none of the Macs I do have came with any OS.  
What I *have* is 29 (25 servers at work, 4 at home) NetBSD PC and DEC 
systems and two Win95 systems (one each for wife and kid).

Even with an FTP program that understands MacBinary, the question is
still "... and then what?"  I have the macutils and xbin packages
both installed on my primary home server, so I can deal with the
.bin stuff.  The .hqx and .sea extensions still have me perplexed.

So far, the assumption seems to be that, once I've un-binhexxed 
the files, they will just magically fly through the air to exactly 
where they need to go.  This isn't a one-button kind of operation.

I did manage to get the "Disk Tools for the PPC" onto a floppy image
which I then DDed to a floppy to boot with.  I can now bring up
MacOS (which seems a huge step backwards).  

Things I have learned so far:

1)  Hooking a 'terminal' up to the serial port is wholly unsatisfying.
My VT102 and VT220 both are limited to 19200 bps, but the serial port
on the 7600s both default to 38400.

2)  With enough cable, you can hook anything to anything.  As in, the 
wife's PC to the 7600.  I used HyperTerm to set the output-device and
input-device setenv's appropriately so now I can get into OF.

3)  Both 7600's are in a more 'advanced' video mode than the monitor
I want to use.  I found a reference to the "Apple" icon, followed by
"Audio and Video" settings.  Right now, the screen appears to sit more
or less 'side-by-side'. 

4)  It is not possible to 'start' NetBSD without at least one working
computer already available.  Apparently, this working computer will
need to be a Mac as well.

Point 2 needs some more explanation.  I have gotten the OF interface
working, as well as the MacOS fixit disk.  How do I use the .smi 
image to patch 1.0.5?

My choices are:
	
	a)  Copy it onto an MS-DOS floppy as a program.
	b)  Use one of the macutils to put it onto a floppy.
	c)  'dd' it onto a floppy.
	d)  Some other 'if you had a Mac you'd understand' thing.

One of the other "interesting" things that I've run across is that
the "Disk Tools" disk won't eject once I've booted from it.  This
makes it really hard to put additional software onto the server.  I'm
sure that part of the problem is that I can't actually read anything
on the screen clearly.

I've been working with Mike Wolfson, who has been a real help on this, 
but I'm still unclear on at least part of the basic concept.  As we 
get closer and closer, I'll be making some suggestions about how to
actually make the PPC part better.

> > I haven't even started on that one yet.  I'm a little perplexed as to the
> > usability of a lap top computer where you can't use the screen that comes
> > with it....

Rephrase that to "where you can't use the screen ...." to "where you 
need another computer to run a program on it."   Right now, it looks 
like I'm going to need about $1000 worth of computer to run programs 
on a old, slow Mac with 8 Meg of memory.

> 
>     Feel free to write a video driver. AFAIK it's unknown, unsupported, and
> undocumented. 

Thanks for your permission.  I was wondering who'd approve that.

I've been curious about this for a couple of days now.  Is the video 
completely useless (as in, no text-only display mode) or is it just
that there is no extended (graphics) mode?

The reason I ask is because at least one of the documents I've been 
through in the past couple of days talked about using the driver
entry points for several of these devices as a basis for the 
interface.  It's only been a thought exercise so far....

> Probably with the only source of information coming from
> decompiling the Apple ROM and searching through ~2MB of Assembly language
> for the init bits. Possibly using a logic probe on chip itself, which is
> probably built into a custom motherboard ASIC. As is the case with the ADB
> controller on my 950.

There are certainly other sources of information.  Besides, searching 
through 2M of Assembly code is a SNAP!  I've done tougher things in my
short career (core dumps by IBM, DEC, and Honeywell; as well as writing
my own BIOS for the CP/M I built in the early 80's).  I'm a trained
assembler programmer - it doesn't scare me at all.

>     Or you could send it to a willing developer to see if they can figure it
> out.

For the $100 I'm out so far, I probably could, or I could try it myself.

I've started a book on the "Design and Implementation of NetBSD", so 
writing a Mac driver to go with the Mitsumi driver I wrote would be
good practice.

Dave Burgess