Ramón Pajares Box
|
Ramón Pajares Box |
|---|
Personal and professional background
My name is Ramón Pajares Box. Please note that Pajares it is my main name, as it's the norm in all Spanish-speaking countries (this is different to the Anglo-Saxon usage). I was born in 1948 in Madrid, Spain. I still live there with my wife Puri and my son Miguel.
All my public life has turned around education, first as a student, then in teaching posts and now in managerial ones. I obtained a PhD (whatever that means) in Philosophy and I've taught this subject matter at the Spanish Open University and at a high school for an overall span of fifteen years.
In the '80s, in parallel to the microcomputer revolution, I shifted my interests gradually to computing matters. As a result I have served ever since as a programmer, system administrator, webmaster and computer do-it-all at several departments in the Spanish Ministry of Education. I became also introduced in statistical analysis and involved in a couple of IEA and OECD international projects on educational evaluation and statistical indicators. I've written some papers on education and philosophy and a book on Ventura, a sadly forgotten computer layout software.
Music and MIDI background
My education didn't include any formal training in music. The little musical knowledge I have was acquired through self-learning. But I've always been fascinated by the magic of classical music without understanding its inner structures.
My lack of musical training prevented me to play any musical instrument when my neurons were young enough to expect some positive results. But when all hope was lost I discovered sound cards and their ability to do the hard part of performing music for me. Through them, the possibility of playing music was now open without the overload of mastering an instrument. So what I really had to master was the art of commanding a sound card to produce the right sounds at the right moment.
This meant to master the writing of MIDI sequences. I began by transcribing (or 'sequencing' in the jargon) some scores of classical pieces to MIDI files. But the results were disappointing. When I listened to the magical sequences of "Scherezade" by Jun Nishio I realized that a lot more than copying notes was needed to get a good MIDI performance. By that time I established a strong friendship with Jean-François Lucarelli who was also trying to learn the secrets of MIDI transcriptions. And since then I have made my progression into the mastering of MIDI with his valuable help and skilful directions.
Equipment
My computer is a no-name Pentium II at 350 Mhz and 128 MB RAM (soon to be upgraded). My sound card is a Sound Blaster Live! always loaded with the 8 MB General MIDI sound font provided with it. I have still to found a better alternative. The sound card output goes through an AIWA XR-EM71 stereo system to feed non-spectacular AIWA bass-reflex loudspeakers or much better sounding AKG K 240 Monitor headphones.
My main sequencing software is NoteWorthy Composer (NWC), a fantastic notation program for its price that allows me not only to copy notes in a superior score interface, but to include almost the full range of MIDI controllers. The final touches are given with Cakewalk. When I need to produce a MP3 file from one of my transcriptions I make a WAV file, thanks to the full-duplex abilities of the Live! sound card, by recording the MIDI performance with CoolEdit Pro. I then convert the WAV file to MP3 with dbPowerAmp Music Converter, a nice free utility.
Sequencing procedures
My way of making a MIDI sequence is as follows:
The huge effort needed to achieve a good sequence requires that I fall truly in love with the musical piece to be transcribed. It is a must to listen several times to a good CD recording of the piece in order to become familiar with the twists and nuances required by a good performance.
I begin to copy into NWC the printed score, note by note with mouse and computer keyboard. I work with passages of 10 to 20 bars at a time throughout the full range of instrument staves. Musical ornaments like trills, grace notes, mordents, grupetti and the like must be expanded in full. Trills, in particular, are always tricky because it is very easy to specify a note expansion that is too quick or too regular.
Next I work on dynamics (piano, forte, sforzando, hairpins) and loudness balance by means of note velocities, stressing the melody lines and blending together the harmony accompaniment, trying always to remain within the dynamics required by the composer. But as classical composers are famous for not giving enough dynamics directions I have always to include a lot more dynamic variances than specified in the score, trying to be expressive without excess.
Together with dynamics I use to adjust phrasing and articulation. Musical phrases tend to start softly, increase their loudness and end softly again but this rule has many exceptions. Articulations are very important in strings, and for every passage I must decide if it should be played legato, marcato or staccato, usually without any composer indication. I also pay special attention to long notes, which almost never can be played with the same loudness throughout their duration. It is necessary to insert gradients of Expression MIDI controllers to modulate their perceived volume while they sound.
Once all this is in fairly good shape it is time to direct my attention to tempo. A MIDI sequence cannot have a uniform tempo marking throughout the whole piece. Tempo, at least, should be slightly modulated all the time to follow phrasing. It is important also to insert stop points to separate musical passages or to enhance important attacks. I owe to Lucarelli a more thoughtful use of tempo that stresses measure beats. For want of a better word he calls it 'swing'. This swing is built by decreasing the tempo value on the first beat of every measure or by increasing it on the last one, or even both. This turns out to be very effective to insert a rhythmic or dance spirit in some pieces, like minuets.
The final stage is to have many auditions of my sequence, making scores of minor edits here and there. Some auditions should be made with full attention, looking at the NWC score, and others only with floating attention with no score at sight. I then export the NWC file to MIDI, load it in Cakewalk and add it some sysex for initialization and some tiny time delays between staves, which contributes to humanize the performance.
Once I think my sequence is ready I make it pass the test of time. I forget it for several weeks and hear it again after this detoxification period. If it still seems valuable I send it to Lucarelli for his analysis and comments. After including the corrections he suggests, this is the time to post the final sequence in my web page and other places like the Classical MIDI Connection.
Only occasionally, as disk space and bandwidth use in web servers are always limited, I also make an MP3 file from my sequence and post it in my own site, in case somebody wish to check what I hear when I prepare my MIDI transcriptions, since the main drawback of MIDI files is that they are totally dependent on the quality of the sound card they are played in, and so some unexpected results may be obtained when my sequences are played in a sound card which does not resemble enough to mine. MP3 files are much more resistant to changes in the MIDI gear.
It is not widely acknowledged how long it takes to finish an acceptable MIDI sequence. Some of my sequences have taken more than a year in the making, with two to four months as an average value. Why so long? Because my free time is scarce but also because sequencing a MIDI file is an art in itself.
A MIDI sound card can produce sound with exact pitch, exact timing and approximate loudness and instrument color. But the music must be added to the tune. And the music in this case means the full set of small alterations of tempo, loudness, note length and the like that should be added to enhance musical expression.
At the end, a careful crafted MIDI sequence becomes a personal interpretation of the musical piece, a true performance where the creator has given solutions to the same problems that a performer or a conductor is faced to. So, even if a machine produces the sound, all the value of a good MIDI sequence lies in having succeeded in removing the mechanical character of the sound produced. And this is neither automatic nor easy.
Ramón Pajares Box
rpajares@rpajares.net
www.rpajares.net
|