Tutorial - Build the system

Build the system

Design it first

It helps to have a goal in mind when setting up a gAlan mesh. Choose a layout that's appropriate for the job you want to do - effects processing, drum sequencing, synthesis, mixing-down etc. This tutorial will cover a fairly general-purpose layout - a three-voice drum machine with a delay effect.

Select and lay out the components

Laying out components on the mesh is simple: click the black background with the right mouse button and a pop-up menu appears. At the top is a submenu named "New", containing categories of component. The component types themselves are found inside each category (with the exception of the Control component, which isn't in a category). When you click on a component type, the system creates a new instance of that class and puts it on the backdrop at the spot where you brought up the menu.

You can move components around the mesh by dragging them within their main body (the rectangular part). If you pop up the menu while over a component, you will find appearing just below the "New" submenu a menu of options relating to the component you popped the menu up over - the per-component menu. On the component menu you will find options such as Rename, Delete, and sometimes a Properties selection and a New Control submenu. Rename and Delete do the obvious things.

The Properties menu-option is used to allow components to pop up a dialog box with settings other than their tweakable parameters. For instance, the Control component uses its properties option to pop up a dialog allowing you to select the lower and upper bounds the signals sent out from the control are limited by, as well as setting the single-step and page increments. The Sampled Voice component uses the properties option to pop up a dialog box allowing you to choose a sample file (WAV, AU, RAW etc) that the Sampled Voice is to play to its listeners.

The New Control submenu allows you to create a control which is linked to one of the parameters of the component. The new control, once created, is placed on the control panel. The control panel is initially not visible - to make it so, you must choose the "Show Control Panel" option from the "Window" menu. Section 3 of the tutorial has more on controls.

Starting with a blank sheet, create the following components (the text in parentheses is the path within the New menu where you will find the relevant component class):

• 1 audio output (OSS Output or Win32 Output, depending on platform) (Outputs/OSS Output)
• 1 delay (Delay/Delay with feedback)
• 1 gain (Levels/Gain)
• 3 trigger sequencers (Pattern/Trigger Sequence)
• 3 sampled voices (Sources/Sampled Voice)
• 3 randomaccess-to-realtime converters (Misc/Randomaccess to Realtime Converter)
• 1 pattern selector (Pattern/Pattern Selector)
• 1 clock (Misc/Clock)
• 1 pattern sequencer (Pattern/Pattern Sequencer)

Make the connections

The next step is to connect all the components together. The most complicated part of this is probably connecting up the trigger-sequencers: do those first (just like the diagram).

You connect components together either vertically or horizontally. The handles on the tops and bottoms of components are event connectors and the handles on the sides are audio connectors. Click the left mouse button on a connector, and move the mouse over the connector you wish to link to. Release the mouse button, and a line will be drawn between the two connectors, indicating an active link. (Note that there's no feedback while you're dragging out a connection: this will be addressed in a future release.)

Next, link the sampled-voices together with their respective randomaccess-to-realtime converters. Note that red connectors are for random-access signals, and blue connectors are for real-time signals. You can't connect the one kind to the other.

Finally, link up the delay effect and connect the triggers of the RA/RT converters to the outputs of the trigger sequencers.

Load the samples

Once all the connections have been made, you need to load a set of samples into the sampled-voice components. On each sampled-voice component, bring up the component-menu and select Properties. In the dialog box that comes up, click "Choose File...". This brings up a file-selection dialog. Choose the sample file you wish to have this sampled-voice component reproduce.

(You might like to assign a kick drum sample, a snare drum sample, and a closed hi-hat sample respectively to each of the three sampled-voices you created.)

Next Section