If it aint long enough, stretch it
Timestretching is one of the most used features on samplers. On the S3000XL it can be found in the Edit Sample menu under DSP section (see manual page s 142145). A screen shot of the section is shown under the next heading.
The ability to make a sample play back at half the speed without changing the pitch is very useful for two main applications. The first is the obvious one...
Speed  cruise control
If you have a loop (for example a repeating drum pattern) which you want to be at a different speed then timestretching is the thing you want. You need to know the current BPM (beats per minute) of the sample and the BPM you want it to be. Then the timestretch factor (which is a percentage) can be calculated using the formula:
Timestretch Factor= 
Destination BPM
Current BPM 
x 100% 
Once this is calculated you simply enter the value under the heading time factor as shown in the diagram below:
For more details on the other options see the S3000XL manual and read the note at the end of this article.
Higher, higher. Lower, lower
Simply playing a loop back at a different pitch will result in the sample being played at a different speed to the original. This is a basic principle of audio which you may have experienced when playing a 33 record at 45 or listening to a tape at dubbing speed. To change the pitch without this happening requires a dab of the old timestretching. To transpose up one octave (12 semitones) you simply set the timestretch factor to 200%. This is effectively making the sample twice as long. If you now play the sample one octave higher then you will here the loop at the same speed as the original but one octave higher. If you want the sample transposed down an octave then you use a timestretch factor of 50%, halving the speed, then playing one octave lower.
For other values you need formula. The basis of this is due to the nature of sound which I won't go into now. Here it is:
Timestretch Factor = 
2 
( 
number of semitones / 12 
) 




x 100% 
So you need to divide the number of semitones (using negative numbers to indicate transposition down and positive for transposition up) by 12. Then calculate 2 raised to the power of this number. Finally multiply by 100. This might seem a bit complicated and so I've calculated from 24 semitones (2 octaves down) to +23 (just short of 2 octaves up) in the table below:
Semitones 
Stretch Factor 
24 
25 
23 
26 
22 
28 
21 
30 
20 
31 
19 
33 
18 
35 
17 
37 
16 
40 
15 
42 
14 
45 
13 
47 

Semitones 
Stretch Factor 
12 
50 
11 
53 
10 
56 
9 
59 
8 
63 
7 
67 
6 
71 
5 
75 
4 
79 
3 
84 
2 
89 
1 
94 

Semitones 
Stretch Factor 
0 
100 
+1 
106 
+2 
112 
+3 
119 
+4 
126 
+5 
133 
+6 
141 
+7 
150 
+8 
159 
+9 
168 
+10 
178 
+11 
189 

Semitones 
Stretch Factor 
+12 
200 
+13 
212 
+14 
224 
+15 
238 
+16 
252 
+17 
267 
+18 
283 
+19 
300 
+20 
317 
+21 
336 
+22 
356 
+23 
378 

A zipped Microsoft Excel version of this table can be downloaded here: transpose.zip
A word for the wicked
One thing you need to beware of when using timestretching features is that no sampler can perfectly alter the speed of a sample and so some degradation of the audio is inevitable. Unless it is the effect you're looking for, I would not use very large timestretch factors. If you do, then most samplers have a quality option which you should put up high (remembering this increases the processing time).
S3000XL Tips By Stephen Tallamy 
