PipeChat Digest #3605 - Sunday, April 13, 2003
 
Re: Digitally Created Electronic Sounds
  by "Bruce Miles" <bruce@gbmuk.fsnet.co.uk>
 

(back) Subject: Re: Digitally Created Electronic Sounds From: "Bruce Miles" <bruce@gbmuk.fsnet.co.uk> Date: Sun, 13 Apr 2003 09:50:54 +0100   A few thoughts on this.   Samples (which are just short recordings) may be of real organs or of synthesised sounds. The output of a sample-using instrument depend on the quality of the samples used. The best modern-day recorders, amplifiers and loudspeakers can handle a range of frequencies and = amplitudes in excess of the response of the human ear, so assuming perfect samples, = the human ear with all its imperfections will set the limit what we actually hear.   Limitations to the length of the samples (ie how long each can play before repeating), and how many notes one sample is used for are likely factors, but using very long samples (several seconds)and one sample for each pipe can overcome this - although at some (not impossible) cost. The multiple sound sources of real pipe organ are a more difficult problem, = and this limitation cannot be entirely overcome by the use of separate amplifier/louspeaker chain for each chamber or pipe rank. The more subtle effects of one speaking pipe on another pipe are almost impossible to quantify let alone reproduce electronically.   'Physical modelling' is a quite different way of generating sounds electronically in which the real life characteristics of a pipe, its = length, diameter and other characteristics are modelled in software as accurately = as possible. The 'excitation' (ie the wind supply) is also modelled and = applied to the pipe model, and the result is a purely electronic creation of the sound. How accurate this is depends of course on how accurate and complete the modelling is. The limitations referred to above will still apply. This is an interesting, indeed exciting method , but I suspect the jury is still out on whether it can produce more realistic sounds than the 'recorded samples' method. If I read Musicom's website correctly this is = the method they have used. On the face there does not seem to any good reason why it should give more realism than sampling.   There are also hybrid methods. The sound can be synthesised in software either directly or using physical modelling and then recorded (ie sampled) and used in the same way as real recorded samples. I have used the direct synthesis/samples method in my 'English Organ' and 'Cinema Organ' sounds. = As an amateur hobbyist in this field, I don't claim they are up to top professional standards but they have fooled quite a few people. I think = they are quite good (well I would say that wouldn't I. :-) And they are free, see my website - = http://www.gbmuk.fsnet.co.uk/index.html   Bruce Miles   ----- Original Message ----- From: "Nichols, Joseph A." <JAN03@mhc.edu> To: "PipeChat" <pipechat@pipechat.org> Sent: Friday, April 11, 2003 3:21 PM Subject: Digitally Created Electronic Sounds     >I have noticed the discussion of samples the last few days on the message board. I do not understand why so many people prefer sampling in = electronic instruments. On a side note, this is my opinion and I do not intend to offend anyone by what I am about to say. Samples, because they are = recorded sound, do not reproduce an accurate "pipe" sound. It is not possible for = the recording equipment, no matter how good it is, to capture all of the harmonics of the pipe in the recording. Once it has been recorded it may = be altered even more before it is used in organs, which would randomly omit more harmonics. And depending on what amplification system is used, that also effects what harmonics might be accented, which could make the sample sound even less like the original pipe rank. The room in which the organ = is installed also plays a roll in accenting certain harmonics, which could be to the advantage or disadvantage of the sound of the organ. Samples do not provide the voicer with near as much details as a pipe voicer would have, simply because it is a recording and not much can be done to alter it = except for equalizer aspects. Samples also allow the organ builders to = mass-produce instrument models. A pipe organ company would have a signature sound, such as Aeolian-Skinner, but every instrument would have its OWN character. Not like a sampled mass-produced model that might be installed at First = Baptist Church, and First United Methodist, and the Presbyterian Church, maybe = even the Lutheran Church. There is a system that exists that is capable of producing a more realistic tone and creating individual instrument character. That system is Musicom and is used by organ builders such as Saville, Copeman-Hart, and Schantz. The Musicom system is a pipe control system with capabilities of producing electronic sounds. I am the son of = the organist at First Baptist Church Murphy, North Carolina, where we had an organ installed in May 2002 that utilizes the Musicom system. It is comprised of 4 ranks of pipes and 79 digital ranks provided by the Musicom system. The possibilities of voicing with the Musicom electronics are endless. This system provides 250 harmonic adjustments per note per stop. Not only can you create a true sound from voicing the harmonics, but you also have the elements that give life to the speech of the pipe such as articulation, wind sound, and pitch sag as you descend to lower notes of lower pitched ranks or when you have a full registration and the = reservoirs would sag a bit to compensate for many stops being on. The organ companies that use Musicom can create a truly unique instrument that is capable of playing any musical style. If you would like more information on the = Musicom system go to www.musicom-ltd.com or if you would like a stop list or a recording of the 82 rank instrument at First Baptist Church in Murphy, = North >Carolina, e-mail me at jan03@mhc.edu   Thank You,   Anthony Nichols