A89: Re: MIDI-TI

[Prev][Next][Index][Thread]

A89: Re: MIDI-TI


Hi!

| As a technical sheet from TICALC says, the port works with 45-50 kBits/s
| so it is fast enough, but how do i put data to the port without having
| responding ACK?

Take a look at the sound library at ticalc.org.  I don't know much ASM, so I
can't do it myself, but it can't be too hard.  Of course, headphones don't
have ACK, and it seems that you can turn at least one side on and off
whenever you want.

| Does somebody know how to read and write to the port without queing the
| bytes like TIGCC does?

Same thing.

| is it possible to do an interrupt that reads or writes to the port every
32
| microseconds (thats the time a single bit would need for MIDI)

I guess no.  Ask Zeljko Juric.

| But there would exist plenty of apps, that donīt need to be 100% stable,
| like my patch-bay for example.

I'm talking about something in the range of 95%.  That would make 3 seconds
off per minute.  Imagine you are playing something, and the program is
changed three seconds before or after it should have been.

| And of course, if everything else runs, who cares if your on-calc
sequencer
| would be timing-drifting about 10% during a day? No one would realize.

I guess I would, really.  And what does "during a day" mean?
Mathematically, you can say "10%", or "2 hours and 24 minutes a day", but
"10% during a day" doesn't really make sense to me.

| one could also enter a beats-per-minute-factor that would correct this
while
| playing one track.

OK, sounds possible, but that would be rather annoying.

| i donīt think the calc drifts a lot during letīs say 10 minutes.

Depends on how you define "a lot".  There was a clock program; I think it
was released by DBA, and it was really inaccurate depending on the battery
power.

Bye,
Sebastian Reichelt
References: