wow : ) are you working with elektron ?? lets start a STARTUP company !!! i have a program that says "learn c c++ in 5 days " : ) will make synthesizers (soft/hard) with waves with totally different values !!! LETS CHANGE THE MARKET !!!! even the SIDSTATION became a "best-seller" and uesed on several monster hits !! (log to sidstation.com /sounds hear >> zombie nation , sandman psychotoons etc... to understand !!! : ) or i'll just fligh to moscow (again)and look for old KGB synths hehehe cheers izik sandman -------------------- its a dog eat dog world out there , but from where i stand there aint enough DAMN DOGS >From: shifty@... >Reply-To: elektron-users@yahoogroups.com >To: elektron-users@yahoogroups.com >Subject: Re: [elektron] phase accumulated... >Date: Thu, 24 May 2001 16:13:45 -0400 (EDT) > > > > what does the term "phase accumulated" mean? > >Well, phase accumulated generally refers to the fact that an >index into a wavetable is used to play back a sample. For a >given frequency, there will be a delta which is added at some >regular time interval (e.g. 1/44100 s) to the current index into >the wavetable, and that sample is played through a D/A conv. > >In most typical wavetable players (e.g. mod players, many non- analog >emulation synths and more), some extra computation/dsp work is performed >on the sample data. This is to account for many things which are >just barely on the edge of my understanding. But, basically, it has >to do with playing back wavetables at too slow of a rate. For example, a >waveform could have overtones in it that go above the Nyquist >rate when hte sample is played at a sufficient pitch. Therefore, you've >go to filter them out, or suffer aliasing. > >However, people often describe the sid as "phase-accumulated" instead of >wavetable in order to refer to the SID's superior synthesis method. What >is so special about it? Nothing, except two things: 1) the waveforms >are not stored in memory somewhere. They are generated based on the >index (I believe it's a 12-bit accumulator?) and sent to D/A converters. >That alone is not such a big deal, it just means the SID doesn't require >external waveform memory, however, when you combine it with the other >superiority, it's important. > >2) The waveform generators in the SID are >generated (I believe) at 1MHz. (Someone please correct me if I goof >on the exact value here ) Therefore, even the highest frequency sounds >that you're going to generate will not generate neigh imperceptible >amounts of aliasinhg distortion. > >e.g. a 20KHz Sawtooth wave, which has a frequency spectrum roughly >equal to 1/h, which h is the harmonic number. The Nyquist rate for a >1MHz plyaback rate is 500KHz. 500KHz/20Khz=25. The magnitude of that >frequency, and those beyond it, is between 3 and 4 % of the fundamental's >amplitude. > >Hope I got that right, >-N > > > > i have seen it in reference to the Sidstation's oscillators but > > have no idea what it refers to. anyone? > > cheers, > > wendell. > > > > > > > > > > > > Your use of Yahoo! Groups is subject to >http://docs.yahoo.com/info/terms/ > > > > > > > _________________________________________________________________________ Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.
Message
Re: [elektron] phase accumulated...
2001-05-26 by izik levy
Attachments
- No local attachments were found for this message.