DJs who choose digital tools enjoy many benefits compared to their analog counterparts, one of which is time-stretching. Time-stretching is the act of changing the speed, or tempo, of a song without affecting its pitch or key. This extremely powerful tool enables the digital DJ to push the limits of his or her mixes by matching songs of different tempos without causing bizarre changes in the tone or timbre of the voices and instruments in a song. (Without time-stretching, the pitch of a song will change in direct relation to the speed of a song—so when you slow it down, female vocals become deeper and male vocals sound like the Cookie Monster, and when you speed it up, male vocals sound female, and female vocals sound like "The Chipmunks.")

While traditional turntablists can only mix songs within a few BPM (beats per minute, a measure of tempo) of each other in order to keep each song sounding natural, Torq® software is equipped with a time-stretching engine that enables you to mix songs that differ in tempo by many BPM. But time-stretching is a tricky, resource-intensive task for a computer to perform. In certain circumstances, time-stretching can result in audible changes, or artifacts, that did not exist in the original song.

Torq lets you choose between the Standard and Elastique time-stretching algorithms.

Torq 1.5 contains two different time-stretching mechanisms, or algorithms, that have different quality results and processing requirements. The first algorithm is the "Standard" algorithm, which is a time-domain algorithm. The second algorithm, called "élastique™," is a special algorithm created by zplane development in Germany.

The Standard algorithm has been in Torq since its first release, and is the only algorithm available in Torq 1.0.7 and earlier versions. This algorithm works by breaking up the audio into extremely tiny "slices" that can be stretched apart from one another in order to slow the playback tempo. Stretching them apart introduces a small amount of silence between each of the slices. To counteract this, the algorithm loops the end of each slice to fill in this silence, resulting in non-stop audio playback at the new tempo. When speeding up the music, the slices will instead be squished together causing them to overlap with one another. In this case, each slice will prematurely cut off the one before it at the faster tempo so there is no need to repeat any audio.

While the Standard time-stretching algorithm is good when you need to make small changes in playback tempo, you will run into certain challenges when making large changes in tempo. If you continue to slow a song down, there will come a point where you start to hear the repeating audio slices described above—which can sound robotic, metallic, or grainy. You may also experience repeated or truncated transients, which are sharp sounds, like the very first instant of a drum sound. If the time-stretching algorithm attempts to repeat one of these transient sounds in order to fill the gap between slices, it will sound like drums are being doubled, or flammed, instead of being sharp and tight.

But don't worry, Torq provides a solution to this problem. By following a few simple steps, you can maximize the way Torq performs time-stretching to achieve a more transparent result. Learn how to set the Beat Grid to optimize sound quality.