Re: virtual hand jive and human factors

On 9 Nov, CAD_CAM_EDM_DRO@egroups.com wrote:

> The 16C63's are good, I have a small cache of 16C73's. I don't have a
> development system, so I might even test on the new 16F877's (when I get
> them). 'F84 enough? Are you willing to share some code for the PIC and
> the TSR? Hate to reinvent the wheel. I think Ballendo is interested
> also. Are you using 'C' or .ASM?

Few things here.

Yes, an F84 is more than enough. ( Love that flash ) Look for a
keypress and bitbang a serial byte. Screw all handshaking .. none
required at these speeds.

And yes, if I can find the code, I'll certainly post it. I'll try to
check over the weekend.

I'm an ASM guy ! I got started flipping switches and pressing Load.
My first hex keypad had me thinking I'd died and gone to heaven.

When the programmer I work with scorn my use of ASM and tlak to me
about "high level languages", I remind them that with my past,
assembly IS a high level language !

The TSR, strangely, is C code !

A LONG time ago, I was asked to write a similar TSR for a fellow who was
blind. He was an SFU student AND a programmer. He walked around with
two devices.

One was a notebook, with software that would "read" his screen and
"speak" it back, through a thing plugged into his parallel port.

The other was a small "daytimer-like" device, geared to the blind, with
a six button keypad used in place of a conventional keyboard.
Combinations of buttons produced letters.

What he wanted was to be able to use his 6 button keypad with his
notebook and could I please do it in C as he had no knowledge of ASM.
Using fragments of code borrowed from here and there and bit of reading
of the Borland TurboC manual, I actually got something working.

However, even an ASM guy can figure out how to patch the C code to
"stuff" what he wants. I know, because I've long since forgotten all I
ever knew about C and I did it .. it's a pretty simple bit of code.

> I agree on using BIOS or DOS calls as much as possible. Are you
> revectoring int 9 (? keyboard int) also?

Nope, no need. Simple as

patching the serial int,
reading the key in the serial buffer when interrupted,
converting it to a "arrow key" scancode
outputting the scancode via an Int16h
return from interrupt.

Seems to work just fine.

> Yeah, CPM BIOS work taught me a little. Wrote one for the MAD 300 'élan
> processor ('386 imbedded) recently. And Socket Services, Memory
> Manager, Data Acquisition...

Cool !

Alan

--

Alan Rothenbush | The Spartans do not ask the number of the
Academic Computing Services | enemy, only where they are.
Simon Fraser University |
Burnaby, B.C., Canada | Agix of Sparta