1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
|
USEFUL INFO
-----------
<Kain> 1000ms = 60 bpm
<Kain> well 1000ms = 1s of which 60 occur in one minute
dsp:
write (dsp,data,4000):
real 0m43.922s
write (dsp,data,8000):
real 1m27.853s
a rate of 8000 (8Hz) seems to give 60BPM
16000 (16Hz?) appaers to give 30BPM
i can only guess 4000(4Hz?) is 120BPM..
2000 must be 240BPM
NOOO!!! All that is wrong...
8000 is 60BPM.. yes
16000 is in fact 120BPM!
4000 is 30BPM (4000 is half that of 8000.. 30 is half that of 60!)
I've been so stupid ;]
so.. the linking number between all of this is __133__
60 multiplied by 133 is approxomatily 8000..
so Rate (in Hz) = BPM * 133
NO! THis is all crap too..
gtk_timeout info:
The timeout function requires the Interval for the timeout to be called, so this
can be used for the BPM setting.. as the first argument in the gtk_timeout_add is
the interval in ms, we can use 1000 as 60bpm.. and so on...
The next problem is working out the best settings for the DSP rates..
ok, thanks to Kain (#LinuxOS/DALnet), we have a nice formular:
1000/(BPM/60.0) (BPM is a float).. this can be easily used as the Interval arg
required for the gtk_timeout_set..
speaker:
<Kain> ok you write to /dev/console
<Kain> you have to turn it on, usleep() then turn it off
<Kain> is usleep() blocking?
If this is the case for speaker output, I need to find a better way..
Timing:
from what i know...
2/4: Tik Tok Tik Tok
3/4: Tik Tok Tok Tik Tok Tok
4/4: Tik Tok Tok Tok Tik Tok Tok Tok
gint count;
time = 1 (if time = 0)
if (time = 1)
write(output, to, DSP);
time ++;
if (time = count)
time = 1;
That code (when written in corect C) should allow for any timing..
Which can allow for future upgrades.
|
