Playing MIDI on iOS with CoreMIDI and CoreAudio API
Written by Ben Smiley
Developing games for mobile platforms has changed a lot since it first became possible to develop games for mobile platforms. MIDI was the only globally supported music format for game-developers to use on mobile phones back in the day. Nowadays you have a lot more options. But MIDI can still be used today on the modern platforms, such as iOS.
The main reason for going with MIDI rather than MP3/OGG, is probably filesize. Some iPhone games are up to 100mb in size, just because of the music. This won't happen if you decide to use MIDI music in your game.
In this tutorial I will demonstrate how you can play MIDI with the native CoreMidi and CoreAudio API's.
The code below sets up an audio graph which consists of a sampler connected to a mixer which connects to an output unit. The sampler is used to convert midi note commands into sounds from a SoundFont. The mixer unit allows a number of channels to be output through the speakers and the output unit sends the sounds to the speakers.
Setup the Audio Graph and play the MIDI file
First, set up variables for the audio graph
OSStatus result = noErr;
AUNode ioNode, mixerNode, samplerNode;
Then, specify the common portion of an audio unit's identify, used for all audio units in the graph.
AudioComponentDescription cd = {};
cd.componentManufacturer = kAudioUnitManufacturer_Apple;
Instantiate an audio processing graph.
result = NewAUGraph (&_processingGraph);
NSCAssert (result == noErr, @"Unable to create an AUGraph object. Error code: %d '%.4s'", (int) result, (const char *)&result);
Next, you need to specify the sampler unit to be used as the first node of the graph.
cd.componentType = kAudioUnitType_MusicDevice;
cd.componentSubType = kAudioUnitSubType_Sampler;
Now, create a new sampler node, and check for errors.
result = AUGraphAddNode (_processingGraph, &cd, &samplerNode);
NSCAssert (result == noErr, @"Unable to add the Sampler unit to the audio processing graph. Error code: %d '%.4s'", (int) result, (const char *)&result);
Specify the Output unit, to be used as the second and final node of the graph - and add the Output unit node to the graph.
cd.componentType = kAudioUnitType_Output;
cd.componentSubType = kAudioUnitSubType_RemoteIO;
result = AUGraphAddNode (_processingGraph, &cd, &ioNode);
NSCAssert (result == noErr, @"Unable to add the Output unit to the audio processing graph. Error code: %d '%.4s'", (int) result, (const char *)&result);
Add the mixer unit to the graph.
cd.componentType = kAudioUnitType_Mixer;
cd.componentSubType = kAudioUnitSubType_MultiChannelMixer;
result = AUGraphAddNode (_processingGraph, &cd, &mixerNode);
NSCAssert (result == noErr, @"Unable to add the Output unit to the audio processing graph. Error code: %d '%.4s'", (int) result, (const char *)&result);
Finally, open the graph.
result = AUGraphOpen (_processingGraph);
NSCAssert (result == noErr, @"Unable to open the audio processing graph. Error code: %d '%.4s'", (int) result, (const char *)&result);
Now that the graph is open, you can get references to all the nodes and store them as audio units.
First, get a reference to the sampler node and store it in the samplerUnit variable.
result = AUGraphNodeInfo (_processingGraph, samplerNode, 0, &_samplerUnit);
NSCAssert (result == noErr, @"Unable to obtain a reference to the Sampler unit. Error code: %d '%.4s'", (int) result, (const char *)&result);
Now load a soundfont into the mixer unit.
[self loadSoundFont:@"gorts_filters" withPatch:1 withBank:kAUSampler_DefaultMelodicBankMSB withSampler:_samplerUnit];
Now create a new mixer unit. This is necessary because if we want to have more than one.
result = AUGraphNodeInfo (_processingGraph, mixerNode, 0, &_mixerUnit);
NSCAssert (result == noErr, @"Unable to obtain a reference to the Sampler unit. Error code: %d '%.4s'", (int) result, (const char *)&result);
Rest needs
// Obtain a reference to the I/O unit from its node
result = AUGraphNodeInfo (_processingGraph, ioNode, 0, &_ioUnit);
NSCAssert (result == noErr, @"Unable to obtain a reference to the I/O unit. Error code: %d '%.4s'", (int) result, (const char *)&result);
// Define the number of input busses on the mixer unit
UInt32 busCount = 1;
// Set the input channels property on the mixer unit
result = AudioUnitSetProperty (
_mixerUnit,
kAudioUnitProperty_ElementCount,
kAudioUnitScope_Input,
0,
&busCount,
sizeof (busCount)
);
NSCAssert (result == noErr, @"AudioUnitSetProperty Set mixer bus count. Error code: %d '%.4s'", (int) result, (const char *)&result);
// Connect the sampler unit to the mixer unit
result = AUGraphConnectNodeInput(_processingGraph, samplerNode, 0, mixerNode, 0);
// Set the volume of the channel
AudioUnitSetParameter(_mixerUnit, kMultiChannelMixerParam_Volume, kAudioUnitScope_Input, 0, 1, 0);
NSCAssert (result == noErr, @"Couldn't connect speech synth unit output (0) to mixer input (1). Error code: %d '%.4s'", (int) result, (const char *)&result);
// Connect the output of the mixer node to the input of he io node
result = AUGraphConnectNodeInput (_processingGraph, mixerNode, 0, ioNode, 0);
NSCAssert (result == noErr, @"Unable to interconnect the nodes in the audio processing graph. Error code: %d '%.4s'", (int) result, (const char *)&result);
// Print a graphic version of the graph
CAShow(_processingGraph);
// Start the graph
result = AUGraphInitialize (_processingGraph);
NSAssert (result == noErr, @"Unable to initialze AUGraph object. Error code: %d '%.4s'", (int) result, (const char *)&result);
// Start the graph
result = AUGraphStart (_processingGraph);
NSAssert (result == noErr, @"Unable to start audio processing graph. Error code: %d '%.4s'", (int) result, (const char *)&result);
// Play middle c on the sampler - sampler unit to send the command to, midi command i.e. note on, note number, velocity
MusicDeviceMIDIEvent(_samplerUnit, 0x90, 60, 127, 0);
Load a SoundFont into a Sampler Unit
If you wish, you can also use a specific soundfont to play the MIDI. You would typically do this, when you wish to get a certain sound. MIDI files at IndieGameMusic.com may be supplied with a soundfound that you can use, in order to make sure the MIDI sounds the way the author intended.
The following code demonstrates how to load a soundfont.
-(void) loadSoundFont: (NSString*) path withPatch: (int) patch withBank: (UInt8) bank withSampler: (AudioUnit) sampler {
NSLog(@"Sound font: %@", path);
NSURL *presetURL = [[NSURL alloc] initFileURLWithPath:[[NSBundle mainBundle] pathForResource:path ofType:@"sf2"]];
[self loadFromDLSOrSoundFont: (NSURL *)presetURL withBank: bank withPatch: patch withSampler:sampler];
[presetURL relativePath];
[presetURL release];
}
// Load a SoundFont into a sampler
-(OSStatus) loadFromDLSOrSoundFont: (NSURL *)bankURL withBank: (UInt8) bank withPatch: (int)presetNumber withSampler: (AudioUnit) sampler {
OSStatus result = noErr;
// fill out a bank preset data structure
AUSamplerBankPresetData bpdata;
bpdata.bankURL = (CFURLRef) bankURL;
bpdata.bankMSB = bank;
bpdata.bankLSB = kAUSampler_DefaultBankLSB;
bpdata.presetID = (UInt8) presetNumber;
// set the kAUSamplerProperty_LoadPresetFromBank property
result = AudioUnitSetProperty(sampler,
kAUSamplerProperty_LoadPresetFromBank,
kAudioUnitScope_Global,
0,
&bpdata,
sizeof(bpdata));
// check for errors
NSCAssert (result == noErr,
@"Unable to set the preset property on the Sampler. Error code:%d '%.4s'",
(int) result,
(const char *)&result);
return result;
}
For more tutorials on MIDI playback on iSO, go visit Ben Smiley's website.
Back to table of contents
