Preface
Recording the piano is one of the most tedious processes one can face in sound engineering. Having an electric piano simplifies the process at large by removing the need to proper acoustic environment, placement of microphones, etc. Nevertheless, it is still a challenge to get it right. The gear I use and the reasoning behind it is described in-depth in a separate section.
Problems
Dynamic range
The main hurdle with recording the piano is its absurdly wide dynamic range. This feature is at the same time one of the greatest features of this musical instrument as it enables the widest range of possible musical expression. But when recording performances for later playback on consumer audio equipment, wide dynamic range causes the overall track to be too quiet. Of course, I’m far from taking part in the Loudness Wars, but still - having your tune play significantly lower than other music on the same hardware - is a problem from the listener’s perspective.On the other hand, tuning the recording hardware too high, especially when the improvisation starts quietly, can often cause distortion when it comes to louder and more heavy parts.
Noise
Every analog signal processing device is prone to noise. Recording at very low levels on one hand minimises the noise produced by the instrument and the recording equipment but on the other hand causes the overall recording to be too quiet. So this is a matter of finding the sweet spot.Not every improvisation is successful
Recording audio signal always needs some adjustment and attention which can “steal the moment” in terms of artistic expression and inspiration. When to record an improvisation you need to set up all this software and hardware, it gets pretty cumbersome when the consecutive takes don’t seem right from the musical perspective. Like Jan Garbarek said: “Sometimes it works, sometimes it fails”. Ideally, to start improvising a pianist should only need to turn on his electric piano (or open his analog one ;-) ) and start performing. This gets even more important as one tries to maintain the recording process in a daily manner, like this website does.My solution
CP300 Sequencer
The current way of recording and releasing the material here started off when I didn’t own a Mac, let alone Logic Studio. At the time, I was recording all my improvisations in the built-in sequencer on the electric piano. This was not thought of as a way to conveniently release music, but just a way of ensuring that my elusive ideas get stored anywhere for later retrieval.Using the Sequencer as a MIDI source
Yes, this is where things get really interesting. The CP300 is actually so well thought out that Yamaha enables the sequencer to act as the source of MIDI signals. So, I open a new Logic Pro project, add an “External MIDI” track and enable recording. The piano MIDI source is set to “Song” which means the sequencer. I start recording in Logic and press play on the piano. The piece is transferred from the device to the project. To prevent unwanted sound ghosting I can turn off “Local controls” on the CP300 which means that the built-in sequencer does not produce sound on the piano, only the MIDI signals. Then the signals are routed to the computer and get back, and only they invoke actual sounds from the synthesizer in the CP300.Cutting out unsuccessful parts
I tend to improvise for extended periods of time. During 2007, when I recorded Anomalies, I usually recorded the whole thing which produced pieces often longer than 10 minutes. This could be sensible for a couple of tunes produced in the period of a year. Now things are pretty different and uploading a new piece every day and actually having something interesting to publish each time seemed contradictive. So, now I still record a whole, usually quite long performance, using the sequencer. Later on, I cut out parts which didn’t go very well and leave only the stuff that sucks the least. But, have in mind that the unedited scores are also published as Logic Pro files, so if you’re curious, you can always take a peek at the whole thing.Replaying the MIDI part on the piano
Now the process gets reversed: I select a stereo audio track in Logic, choose the input source for this track to the pair of inputs on the Firepod where the piano is connected to, arm the track for recording, and simply start recording in Logic. The software plays the MIDI part back, and records the audio on another track. During recording, I can adjust any knobs and sliders to maintain the correct volume level.Finishing touches
Having the final audio recording in place, I can apply a reverb on the track, compress the sound a bit, draw a volume envelope, etc. Finally, I mute the MIDI track to prevent surprises, and “bounce” the audio to an MP3 file. Voila.Advantages
I guess the biggest advantage of this approach is the ability to start and stop recording while sitting by the keyboard. Also while the laptop is away or I just play at night with my headphones on. Just record, save in the sequencer memory and leave it there until you have time to transfer it to Logic.
The other naturally useful thing is the ability to edit the notes without any restrictions. This is one of the places where I actually limit myself and not use this ability, to maintain a certain level of aunthenticity in the improvisations. Still, this feature is useful when I cut improvisations because the actual piano sound is not affected and when I drop some unsuccessful passages out of the improvisation, the thing still flows and sounds as if there wasn’t any omissions.
Lastly, the re-recording process is done after actual performance. This enables me, and myself only, to be performer, sound technician “at the console” and the producer, at the same time. Re-recording MIDI as audio enables me to control the volume and gain knobs on the piano and the interface, to delicately decrease the dynamic range for longer passages of certain amplitude. And, what’s most important for a mastering amateur like me, I can re-run this process if I choose poorly and the recording is too hot or too low.
And at the end of the day, the produced files sound exactly like my performance would if you were here.
Closing thoughts
There is always room for improvement. In particular, I still fail to find the sweetest spot for the gain knobs on the Firepod and the volume knobs and sliders on the piano. I am sure that there’s a better way to do this.
Another thing is the MIDI part of the recording which basically makes every performance to be recorded at least thrice (firstly on the piano, secondly MIDI from piano to Logic Pro, and lastly from Logic Pro to Logic Pro as audio). This takes time.
So, if you have any suggestions on how to improve my method, drop me a line.