Subject:
|
DLM for the modern age
|
Newsgroups:
|
lugnet.robotics.handyboard
|
Date:
|
Wed, 21 Feb 1996 05:39:05 GMT
|
Original-From:
|
Fred G. Martin <fredm@[spamless]media.mit.edu>
|
Viewed:
|
1762 times
|
| |
|
|
Hi all,
Given Kazuyoshi and other's problems with DLM, if there are any
Windows programmers out there who would want to take a stab at a
Windows or W'95 implemention, it would be a Good Thing. I'm afraid
I'm not qualified to take this on myself.
The one trickiness which should be a nice challenge is that there must
not be more than a four character delay in the bootstrap sequence when
sending the initial zero page RAM program to the e1 chip. the a1 chip
is friendlier in that it will wait indefinitely until it gets the 256
bytes of program, but the e1 will say "ok you're done" as soon as
there's a 4 char pause in the boot download sequence.
Getting this kind of real-time performance is trivial in DOS but
perhaps not in Windows. The strategy would probably be to load a
buffer with the whole bootstrap sequence and hand it off to the serial
xmit call. Interestingly, this isn't where DLM is failing. DLM uses
BIOS calls to set the serial port and I think this might be the
problem.
if anyone is interested, i can provide further details. source code
for dlm is online cherupakha.
-Fred
|
|
1 Message in This Thread:
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
