I have opted for a dedicated computer based system with a few items of hardware for specific jobs. As well as this being specifically optimised for audio use with a more suitable processor & motherboard than the one I currently have, it also avoids all my other software slowing down what is already a power-hungry process. Should anything go wrong with either PC, the other could serve as an emergency backup for the duties of the other; another example of multiple supply. The computer interfaces with the other equipment in two ways. The main interface is through the computer's sound card which is connected via a dedicated cable to a hardware unit made by Edirol.

 This unit has the option to connect up to 8 inputs & 8 outputs at any one time & is expandable as other units can be connected to the system using extra sound cards. Up to four can be chained together giving 32 inputs & 32 outputs, though one unit is fine for my current purposes. I have chosen to use a keyboard to produce sounds & input MIDI data to control other modules, such as the DarkStar synth & some virtual software instruments. This gives me a multiple supply of sounds to use, controlled from the one instrument, minimising my learning needs. I can sample CDs if I wish using my hardware CD burner connected up to the keyboard, the digital multitrack recorder or the computer. A hardware unit provides reverbs without overwhelming my computer processor & also some other built in delay-based effects.

 I can record audio via a pair of microphones & a microphone preamp, which boosts the small signal to a level which can be used in the computer. The microphone preamp has a Compressor, Equaliser & a Gate which can also be used as hardware processing for signals from the computer. I can use this to record vocals, acoustic instruments & to sample fixed tone instruments such as my crystal singing bowls. I can take these recorded signals, pitch shift them to pure tones & use these to create chords or harmonies to go on my recordings. I am using different types of microphones as their different characteristics are suited for different purposes.

 I can control the mixing process in a much more hands-on manner using a mixing console, which is connected to the computer via MIDI through a USB connection. The screen provides me with a visual guide to what I am doing & the active studio monitors allow me to hear what I am creating accurately. The studio monitors have two inputs, so they can accommodate a signal from the computer or directly from another module, such as one of the synths. A DAT recorder & the CD burner allow me to produce different formats of red book compatible masters & the CD burner on the computer could burn CDs to sell directly if I need them. A signal splitter (CO2) allows me to use both the CD burner & the DAT recorder at the same time.

Capturing ideas

 In order to capture the ideas that I have I need to be prepared. Having a notebook &/or dictaphone to hand helps to avoid situations where I have to hum a tune for an hour or so until I get to somewhere I can record it. I have most of my ideas first thing in the morning in the hypnogogic state just after waking, so the dictaphone needs to be kept by my bed whenever I am at home.

 If I go out it can go with me in my bumbag & a little notebook that I have is ideal for noting down ideas for lyrics that I might think of. As I learn to read music & recognise different pitches, then I may be able to record much more information just in the notebook, though the lo-fi dictaphone can still capture a surprising amount, including rhythms & timbres. A further, but much less portable option that I have is the digital multitrack recorder. This will make CD quality recordings 'on location' with a pair of good microphones, but still needs a mains supply. Once I have my recordings I can go into the studio, which for ease of use is in my own home.

 My first job in the studio is to transfer my ideas onto the computer. The recording software is called Cubase SX & it enables me to do a great number of things, the basics being the recording of both audio & MIDI data. I plug the microphone directly into the Edirol as I'm only going to use this recording as a guide & then vocalise my melodies, words & rhythms into Cubase using individual tracks for each 'instrument'. I then have this as a guide from which I can later make proper recordings of different instruments.

Into the studio

 Each song I place into a separate file & I keep all my own ideas in one folder, while the recordings I do for others go in their own folders. Everything that I record goes onto the hard drive storage of the computer, but this alone is not sufficient for safe keeping. Once I have recorded my ideas into Cubase I will start re-using the dictaphone tape again, so I need to make back-up copies of my recordings. This I can do using rewritable CDROMs & extra copies of these can also be stored away from the studio for extra data safety.

 At this point I separate my song ideas with words & titles into one sub-folder & the purely melodic or rhythmic ideas into another sub-folder. The former are what I use to build up ideas upon & I begin by ascertaining what the tempo of the song is going to be. I do this by using a facility within Cubase that allows me to tap a key on the computer keyboard in time with my musical idea & it tells me what tempo it is in beats per minute. I can then enter this into the software & use it as a basis for recording the MIDI data.

 My next job is to decide which of the sounds that I have access to best suits what I hear in my head. Inevitably, this is unlikely to be exactly what I am looking for, but it provides a basis from where to start & even if the end result sounds different from my first idea, it may still be good! Those ideas that amount to simply melodies & rhythms I can revisit from time to time to see if there are any other ideas that mesh in with them. These could be new lyrics, other melodies or they may fit in perfectly with a new song that I am working upon.

Building it up

 I can replace each of my rough audio tracks in turn, either with MIDI tracks or with properly recorded audio. At this point, if I am intending to sing a vocal on it I need to make sure that it is in the right keys for me to be able to sing. If it isn't, this is the most sensible point to transpose it to a more suitable one; before I add anything else. I can then try out some chord progressions to see what works, first making sure that they are appropriate for the keys that I have chosen!

 Once this is done, I can start duplicating & cutting up MIDI data & moving it around to create the arrangement that I want. This will need to fit around a verse structure, so I need to decide this which will also relate to what lyrics I have & how they fit together. These things then will come together fairly simultaneously & any further instruments that I want to add to the mix can also be a part of this process. Once I have decided upon the approximate lyrics & melody for the vocal & can record a guide vocal which will help me in arranging the other instruments around it.

 I could use a basic microphone for this as it isn't intended to be used in the final mix, but then again if it is a really good take I might want to keep it. It could then be worth the effort of setting up the microphone properly to record this & I have read stories of famous artists who have ended up using these takes on final recordings as it was a relaxed & natural take. With all of this done I can adjust the different volume levels on the mixing console & do some rough panning too to get an approximate mix.

Adding it all together

 This is the point at which the song is really starting to come together. Any new ideas that I might have as a result of working on the mix get added here, as do any of those 'homeless' melodies & rhythms that fit, that I filed away previously. I can also add any samples to the mix that would suit it. These days there are a lot of artists who sample sections of old recordings to use as part of their own & there are also lots of different 'samples' CDs available to buy, which unlike the former don't have any potential copyright issues.

 A third option though is to sample my own instruments & use the pitch-shifting facility within Cubase to create notes that the original instrument is not capable of producing. I can file these away for use in any of my recordings & if they work in this particular context I can use them to create chords or harmonies within the song. With all this in place I have hopefully reached a point where I can add the final vocal track, but this of course requires me to come to a 'final' decision about my lyrics. Of course nothing is actually final until it is mastered, but as the vocals affects the whole song structure, it is good to get this as good as clarified here.

Fine tuning & Mastering

 I can now experiment with some harmonies for the main vocal & record those that seem to work in the current context. I then move into the fine tuning cycle of tweaking & listening back. This involves the different stages of continually adjusting the mix levels to get them right as each new element is added in, overlaying effects & bringing elements in & out of the mix. These can all vary throughout the song, effects levels can be altered, track volumes faded up & down & elements panned from side to side in the mix. This cycle continues around until I am happy with the final mix. At this point the song is ready for mastering & this involves several processes. Usually effects are applied to the overall mix, such as reverb & it is mixed down to stereo. A further process occurs at this point called dithering. This is where the track is reduced to a 16 bit resolution suitable for a CD.

 This is necessary as Cubase can use 24 bit or 32 bit processes to make the best signal to noise ratio in the final recording. Every time the signal is edited, a little resolution is lost, so only starting out with a 16 bit signal, may produce a final recording of only 13 or 14 bits. By starting with 24 bits for instance, the final recording may still be 21 or 22 bits, which can then be reduced to 16 bits & will still be able to make the most of the CD medium. The downside to the higher bit rate recording processes is that they produce files of much greater sizes, requiring bigger hard drives to store them upon & more RAM to process them.

 I can compile a collection of these mastered songs together, deciding upon a running order & then record a 'red book compatible' CD & DAT (plus safety backups) which will be suitable for duplication purposes at a mastering house if necessary. I can also then burn CDs to sell direct from my hard drive if I wish. All that is then left is to do the artwork for my CD cover & print it off (using the other PC) & put it all together.

Final Design

 So here it is; my final design in its entirety:

Implementation

 Putting all this into practice has been not quite so straightforward, but mainly because of the steep learning curve that I have had to follow with all the new equipment. The first part of the design is generally going well; waking up in the morning I often find myself with plenty of musical ideas & if I allow myself to float in & out of the hypnogogic state I can pull quite a lot of an idea together. I bought the dictaphone originally to help me with dream recall & this has also been ideal for this job too (multiple yield!). My notebook is already filling up with ideas, some of course better than others & a track list for a CD which is a long way from being finished (some of the 'tracks' are still merely titles!). I have discovered that the dictaphone is best kept by my bed, but I don't always remember to take it when I go out... usually when I get a good tune appear in my head. Going into the studio & putting all these ideas down is a very satisfying experience, but already I have a glut of ideas that I haven't got around to doing anything else with yet. Having them in this form works well though as I can focus on vocalising one instrument at a time & layering them on top of each other, making it much easier to recall the tune that was actually in my head at the time than from the lo-fi dictaphone.

 I have discovered that unless I am clear about vocalising rhythms & timbres one at a time, I can listen back to something & not get what it was about at all! I have found my self getting annoyed at my inability sometimes to find the right sound for the idea that I have. There is something inherently frustrating about not being able to just take the idea in my head & plonk it directly onto a CD!

 I have stumbled up against my lack of knowledge with the equipment & have sometimes taken a lot of time just to sort out why there is no sound coming out of my studio monitors for instance. There are so many interconnecting elements here, even, in fact especially within the software itself, that this kind of thing happened quite a lot to begin with. I have found that the easiest way to learn has been to get on & develop my ideas & look up what I need to know as I go along. I have found that in general the design works well, most of the challenges have been in working out the finer detail.

 I have found that I need to be aware of the acoustics of being in a small room & how that affects frequencies within the bass region. If I'm not careful this can give me a false idea of how my music will sound, so I need to keep playing commercial CDs through the same system, to give me a sound to aim for.

 I have done some sampling of my own & in the process come up against the limitations of using microphones. However quiet it may be outside, when I had a swarm of flies buzzing around indoors it made recording the quiet fades of the crystal singing bowls for instance, impossible. Choosing my times to do such things is also crucial it seems. As yet I haven't done any serious mastering, but I know that when I get to that stage I will learn that as I go along too. As for learning the instruments, I found that the guitar gave me problems holding down strings on the frets without touching the strings either side.

 I found the keyboard much easier in that respect & that is where I have made my most progress. I will be returning to the guitar to have another go soon though; when I can fit it into my busy life! In the meantime I continue to play with it all & learn as I go. As for opening up the studio to other like-minded musicians, I have put together a web page that was later adapted to be part of Paddox Farm's own website. This is my Sounds of Light web page.

 The easiest thing for me to record so far has been my singing voice. I have recorded a particular favourite song of mine that was written by one of the people I lived with in Eire. It is included on this CD to be played in a normal CD player or it can be listened to by clicking on the link below:

Maintenance

 With this design, I am really filling in the finer detail as I go along & to date don't really have enough experience with it to come to any firm conclusions about what tweaking to do. I am however remaining very aware of what does & doesn't work well so as to instigate any improvements in phases as I go along.

 An evaluation of this whole process can be found on the Design Review page.