Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!uunet!zephyr.ens.tek.com!orca.wv.tek.com!pogo!bobt From: bobt@pogo.WV.TEK.COM (Bob Tidrick) Newsgroups: sci.electronics Subject: Re: Need circuit to drop music an octave Message-ID: <9592@pogo.WV.TEK.COM> Date: 9 Aug 90 19:05:43 GMT References: <1990Jul20.223615.4305@portia.Stanford.EDU> <1332@fs1.ee.ubc.ca> <58975@lanl.gov> <32424@cup.portal.com> <1990Aug06.150222.23167@pmafire.UUCP> <15547@thorin.cs.unc.edu> <6101@tekred.CNA.TEK.COM> Reply-To: bobt@pogo.WV.TEK.COM (Bob Tidrick) Organization: Tektronix, Inc., Wilsonville, OR. Lines: 27 In article <6101@tekred.CNA.TEK.COM> vekurpan@tekred.CNA.TEK.COM (Vincent E Kurpan) writes: >Unfortunately, the simplest scheme may be to digitize one cycle of the >waveform and then play back with a clock that is synthesized for each >note. This is not totally trivial and works only one note at a time. >Theres no free lunch. This is not exactly true. When one waveform is digitized it can be played back at half speed. If you are sampling at say 50 Khz send it out again at 25 Khz and you will have the wave form at half speed. A reasonably fast micro processor should be able to keep up with the playing. All you will lose is half of the waveforms. but you will need to lose these anyway. What needs to be done is sense the highest peak of the wave form and look for it again. This should give you one period. send this to a circuit that will clock it out again at half of the sample rate. Another waveform can be caught while this one is going out. Higher notes will take less samples and lower ones more. You would want to be sure to allow enough memory in your playback circuit for the lowest notes. Something similar to this has been done to compress speech. The wave forms are put out on a tape running at half speed. When it is played back the speech is at the proper frequency but twice as fast. Just my $.04 worth (inflation you know) -- Bob Tidrick GPID Engineering Tektronix Inc. Wilsonville OR.