Mouse interface stuff - K.O.?...

From: Hársfalvi Levente (levente_at_terrasoft.hu)
Date: 2000-09-17 17:02:13

Hi all!


As I'm back from holidays (and finished with some other time consuming
tasks) I continued the 'mouse interface saga'. ...I think I have some
troubles :-(. 

After fiddling with a diodes+capacitors type voltage quadrupler, I see
it proven that I can't provide the required supply voltage with such a
circuit. The test circuit generated an about 18v voltage swing, but the
output quickly dropped to 8-9v when it had to source some 2-3
milliampers :-(. Since I become quite fed up with this matter, I decided
to use a MAX232 (or compatible) RS232 line driver chip. ...Would assume
that most of you know this well. Maybe this solution is a bit over the
requirements, but as I've seen, similar but simpler (8-pin DIP) chips
like the MAX680 and the LT1026 are even more expensive and harder to
find. So I think I'm better off with a '232 that becomes more popular
and cheaper month to month.

(But if any of you knows an appropriate chip that generates either dual
+-10v or a single +20v from a single 5v supply @ about 15mA without
serious voltage drops, I'd check and maybe use that, if it results in a
smaller/simpler/ or cheaper design).

As speaking of the PS/2 mouse plan, after reading the specs (thanks to
Richard) I see it'll reveal much more problems in the code than the
RS-232 (async) communication ever had :-(. The problem is, the bitrate
is quite high. According to the docs, the bit time is about 30-50
microseconds. I can delay the byte transmit even for several
milliseconds, but unfortunately not during transmitting a byte
(transmitting a particular bit in a byte can't be delayed). During
generating the duty cycle for the POTX/POTY inputs for the SID in 1351
mode, the microcontroller can be off even for 140-150 cycles (because of
the needed exact timing). (One cycle corresponds to one microsecond @ 4
Mhz osc clock). 'Of course', the PIC16c84 has no SPI interface or
similar toy, one has to do everything by software. Bit banging rulez
like hell... :-/

Maybe I should change the microcontroller I use. Two possible options
should be either another PIC micro (...they're EPROM based, so dropped)
or an Atmel AVR. This latter is more promising, but I'm not that
familiar with their ASM programming at the momemnt that I could
quick-change to a 90s2313 in the design, without serious delays in
finishing this cr*p. I suspect it would cheat me several times until I
discover all critical features.

After re-thinking everything I'll probably find some solution (last year
I also doubted that true 1351 emulation would be possible with this
hardware, and see what happened :-) ). But it will be complicated,
that's for sure.

Until that, I go on for completing the serial mouse interface. ...And
who knows? After cleaning the source, I'll see if I can modify some
parts to cope with the fast sync data stream of the PS/2 mouse.

Regards,


L.

-
This message was sent through the cbm-hackers mailing list.
To unsubscribe: echo unsubscribe | mail cbm-hackers-request@dot.tml.hut.fi.

Archive generated by hypermail 2.1.1.