About Streaming Mode
As opposed to loading audio into static buffers, you can also set up RTGS3 to stream audio from either a soundfile or an audio input device (such as a microphone connected to an audio interface).

Activate Streaming Mode by clicking its toggle box in the Buffer section of the main interface window:
You'll notice that the Buffer Position slider and the "View Buffers..." button have been deactivated, and the Buffer automation controls have been replaced with different sliders. (The Buffers window closes as well, if it was open.) This is because Streaming Mode takes control over the Audio Buffers, filling them and erasing them automatically as audio streams into RTGS3.

Let's start by streaming an audio file into RTGS3. Locate the Input section of the main interface window and select "Soundfile" from the Mode menu:

Click the "Open..." button and navigate to an audio file on your hard disk. Any uncompressed audio format can be used. If you have Apple's GarageBand installed on your computer, load the file /Library/Audio/Apple Loops/Apple/Apple Loops for GarageBand/Acoustic Picking 04.aif . The settings we'll be trying out in this tutorial will work well with that file.

Turn Loop on by clicking its toggle, and then start the soundfile playing by clicking the play button on the soundfile playback controller (which is just below the "Open..." button).
You can mix the soundfile playback into the main output by moving the Thru slider to the right. We'll look at routing options in a later tutorial.

Make the following settings in the main window:
Grain Density = 10 milliseconds
Grain Length = 150 milliseconds
Transposition Amount = Full Random (-24. to 24. semitones)
Transposition Mode: Pitched | Set to: octaves

Turn the Grains On to hear the result. This kind of effect works well with instrumental input, although I wouldn't recommend it for processing spoken or sung voice.

Now let's try applying this effect to live audio coming in to your Mac. You'll need to have some kind of input device connected to your computer, such as a microphone or instrument plugged in to a USB interface, although you can also use the Mac's built-in microphone.
BE CAREFUL WITH YOUR EARS AND EQUIPMENT! Always make sure to turn down the volume on your amplifier or the RTGS3 main output faders before turning Grains On, especially when using live input.

Bring the Output Gain faders down a little bit before switching the Mode menu of the Input section to "Live Audio In". Check that sound is getting into RTGS3 by looking at the audio level meters:
If you're not getting any signal from your input device, click the gear icon in the Input section to see if your device and source are recognized by RTGS3. If the device and source you want to use are not available in these menus, then you will need to make sure that your Audio Driver is properly set up and selected in the System Setup window.

If you're using a microphone, try scratching your fingernail along its opening to hear how RTGS3 is processing the sound.
Bring the main output faders up slowly until you reach a good listening level without feedback.

Let's finish up by checking out the Latency and Stretch sliders in the Buffer section of the main interface window. First, bring the Transposition Amount Random Range down to zero (you can do this easily by clicking the small inward-pointing arrows located just above the slider). You'll notice that the effect on the live input is not as pronounced, and that there is a quarter-second delay (or Latency) between sound input and output.
Adjust the Latency slider in the Buffer section of the main interface window. You'll notice that moving the slider to the left reduces the delay time between input and output, but that at a lower setting there is a higher chance of undesirable artifacts in the sound. When using long grains or extreme transposition settings these artifacts will be produced more frequently. Setting the Latency value higher will help to prevent these artifacts from happening.
Now try moving the Stretch slider up to a value of 50% or higher, and speak some text into the microphone. What's going on here is that RTGS3 is recording chunks of sound from the audio input and playing them back at a slower speed. When the slider is set to 50%, 5 second long chunks of sound are being stretched out to 10 seconds. When it's set to 100% the input sound is virtually frozen at one point, because 100 millisecond chunks of sound are being stretched out to 10 seconds of output.