[colorforth] Umbilical colorForth
- Subject: [colorforth] Umbilical colorForth
- From: "Howerd Oakford" <howerd.oakford@xxxxxxxxxxxx>
- Date: Thu, 12 Feb 2004 22:29:26 -0000
Hi,
I have found a 350MHz Compaq PC that does not work with the colorForth
keyboard, driver which looks at I/O port 0x64.
Compaq seem to have a different type of incompatibility with every machine!
The 350MHz PC has a display that only works with Chuck's video driver - a
450MHz Compaq only partially worked with Chuck's code - the image was there
but masked by horizontal lines ( could it be the VESA linear array bit ? ).
At least the floppies read OK...
Debugging this could be tricky - one idea I have is to use another PC
running colorForth as a "terminal", and connect through the serial port.
This would also make a good testbed for network block access.
In the early '80s I was using a simple protocol that used ASCII control
characters to initiate block read/write, and a QUIT loop.
As we have loads of memory, I would simple allocate 256 K for the remote
machine, with a table of updated flags.
We will also need to read/write I/O address space. I don't see the need for
QUIT as the host PC's stack will be fine for entering text.
But we will need to select local/remote execution. chipForth uses
vocabularies for this ( HOST and REMOTE ) . Could we use the macro
vocabulary for this? Or a different colour token? Any ideas???
I will probably go for a PPP style HDLC subset for packet transfer.
[ BTW for ANS speakers : vocabulary = wordlist ]
Regards
Howerd
PS I like Mark's Matrox driver. I will try to find one to try it out on...
---------------------------------------------------------------------
To unsubscribe, e-mail: colorforth-unsubscribe@xxxxxxxxxxxxxxxxxx
For additional commands, e-mail: colorforth-help@xxxxxxxxxxxxxxxxxx
Main web page - http://www.colorforth.com