aboutsummaryrefslogtreecommitdiff
path: root/help/vd16~-help.pd
blob: 1f4a4c5c97ac0179f832559cb379a6ae641ef8e2 (plain)
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
#N canvas 88 40 717 480 12;
#X floatatom 50 254 0 0 0 0 - - -;
#X obj 50 347 outlet~;
#X text 130 346 signal output (delayed signal);
#X obj 50 282 sig~;
#X text 99 279 signal input (delay time in ms);
#X text 218 310 creation argument: name of delay line;
#X text 35 400 see also:;
#X text 77 10 reads a signal from a 16bit delay line at a variable
delay time (4-point-interpolation);
#X text 31 51 vd16~ implements a 4-point interpolating delay tap from
a corresponding delwrite~ object. The delay in milliseconds of the
tap is specified by the incoming signal.;
#X text 16 433 similar pd-objects:;
#X obj 201 432 delread~;
#X obj 285 432 delwrite~;
#X obj 378 432 vd~;
#X obj 123 403 del16write~;
#X obj 242 403 del16read~;
#X text 411 386 updated for iem16 version1.0;
#X text 28 116 This is very similar to the pd-object;
#X text 56 137 It uses only 16bit to store the samples \, which will
need only half of the memory of pd's floatingpoint-based object.;
#X text 57 175 However \, there are 2 drawbacks: there will be some
additional noise (because floats are more precise than 16bit) \, and
you cannot have values>1 stored in the delay-line;
#X obj 375 112 vd~;
#X obj 50 314 vd16~ del_example;
#X obj 24 16 vd16~;
#X connect 0 0 3 0;
#X connect 3 0 20 0;
#X connect 20 0 1 0;