#468241  11/06/03 01:23 PM
Varispeed

SlySi
Planeteer
Registered: 10/15/03
Posts: 38
Loc: Hertfordshire, UK

Can anyone please tell me what the KHz intervals are for varispeed adjustments i.e concert pitch is 440hz.. what is a semi tone up/down etc etc.
Does this make sense?
cheers Simon www.colvex.com

Top




#468242  11/06/03 04:58 PM
Re: Varispeed

fullcity
Planeteer
Registered: 05/28/03
Posts: 52
Loc: New Jersey

I'm going to take a WAG at this one. If you are recording at 44.1 kHz as your reference sampling frequency, then 88.2 kHz would be double the pitch (+ 12 semitones) and 22.05 kHz would be half the pitch ( 12 semitones).
Extrapolating from there, one semitone would be equivalent to (44100 / 12) = 3675 Hz, or 3.675 kHz. Make sense, or total hogwash?
_________________________
My newest CD , "Looking Past The Present" and other VSrecorded tracks

Top




#468243  11/06/03 05:09 PM
Re: Varispeed

mike buzz
Planeteer
Registered: 04/28/99
Posts: 1958
Loc: paso robles ca usa

Full I don't think semitones are evenly spaced in frequency , octaves are but not semi tones ( I think ) SLysi remember that the varispeed is just like a tape speed knob it will change tempo as well as freq. if you apply enough varispeed the tracks will start to sound unnatural Also you cannot cd burn at anything other than 44.1kz on the 2480
Later Buzz

Top




#468244  11/06/03 06:05 PM
Re: Varispeed

Alterboy
Planeteer
Registered: 11/28/01
Posts: 437
Loc: Vancouver, Canada

Here is an explanation of the relationship of frequency and pitch. I hope you enjoy it.
By definition an octave is a halving or doubling of frequnecy. A440 [in the octave above middle C] is the standard of tuning and you will discover that the A string on the guitar, which is two octaves lower than A440, would have a fundamental frequency of 110hz. The A string on the bass would have it's fundamental pitched at 55hz. The lowest note on the piano would be 27.5 hz.
If you take a moment and calculate, you will see that with each successively higher octave, there is an increase of twice the number of frequencies per octave. For example, take the range of human hearing, listed as [arguably] 20hz to 20,000hz. The first octave from 20 to 40 hertz contains 20 "frequencies". The next octave from 40 to 80 contains 40 "frequencies"; the third octave from 80 to 160 contains 80; etc. What we have here is an exponential increase in frequencies with a linear increase in octave.
The upshot of all of this is that there is no specific frequency measurement from semitone to semitone although it certainly can be calculated. Here's how: Start with the frequency in question and multiply it by 1.05946. Take middle C for example 261.624hz and multiply by 1.05946 to find C# = 277.18 To find D, multiply C# by 1.05946 = 293.661hz. You can use this multiplier to calculate semitone change starting with any known frequency.
I hope this helps your understanding. Two more things to keep in mind. Pitch changes can be measured in cents. [Some of the electronic guitar tuners have cent measurements.]There are 100 cents for each semitone, whether you step up or down, no matter what note you start with. In otherwords, each octave always has 1200 cents. This is a much easier method of calculating change in pitch. If you wish to increase any given note by a perfect fourth for example, [5 semitones] you would raise the pitch by 5 X 100 cents or 500 cents. And finally, the human ear is nonlinear with respect to pitch. A good demonstration of this is the way pianos are tuned. Stretch charts are used to tune the upper and lower ranges of pianos. These charts indicate that to sound in tune, very high notes half to be tuned sharper and very low notes have to tuned flatter than the mathematical relationship defined by "an octave is a doubling or halving of frequency"
Fun with math...

Top




#468245  11/06/03 07:52 PM
Re: Varispeed

SteveDWalker
Planeteer
Registered: 06/01/02
Posts: 3036
Loc: Clearwater

Alterboy, you didn't mention that the sampling rate has nothing to do with semitones. So you don't use 44.1khz to derive what the semitone frequencies are. Look to the piano instead as Alterboy explains.
Steve
_________________________
Steve
Roland VStudio 100, Sonar Platinum, VS2480HD, MAudio, Korg,Alesis, Audix, Shure, Mackie, JBL, EV, Taylor, Gibson, Ovation, Ephiphone, Ibanez, Fender, Sunn 2000S

Top




#468246  11/07/03 09:04 AM
Re: Varispeed

fullcity
Planeteer
Registered: 05/28/03
Posts: 52
Loc: New Jersey

This is interesting discussion. I do realize that sampling rate (number of times per second that the source audio is 'read', or sampled) is different from fundamental frequency (number of times per second that a waveform repeats its cycle).
I think this is something that you can easily try on your machine. How many Hz do you bump the sampling rate in order to raise the pitch of the musical content by one semitone? Try it and tell us. If I wasn't at work, I would do it immediately!
_________________________
My newest CD , "Looking Past The Present" and other VSrecorded tracks

Top




#468247  11/07/03 10:37 AM
Re: Varispeed

SteveDWalker
Planeteer
Registered: 06/01/02
Posts: 3036
Loc: Clearwater

fullcity, Timedoman and frequency domain techniques are the two main approaches to time stretching and pitch shifting. It appears that the 2480 does Vari Pitch using timedoman technique by varying the sampling rate in the range of 16khz to 50khz. The frequency domain technique uses DSP algorithms and does not change the sampling rate.
_________________________
Steve
Roland VStudio 100, Sonar Platinum, VS2480HD, MAudio, Korg,Alesis, Audix, Shure, Mackie, JBL, EV, Taylor, Gibson, Ovation, Ephiphone, Ibanez, Fender, Sunn 2000S

Top




#468248  11/07/03 11:00 AM
Re: Varispeed

Wish
🌴 VSP Moderator 🌵
Planeteer
Registered: 09/09/00
Posts: 8604
Loc: Phoenix, AZ

Here's a conversion chart someone had posted awhile back...other people commented some of the values were slightly off.
Use at your discretion!!!
Rate = Key = Semitone 49.67 = B = +2 46.72 = A# = +1 44.10 = A = 0 41.59 = G# = 1 39.41 = G = 2 37.00 = F# = 3 34.93 = F = 4 32.97 = E = 5 31.33 = D# = 6 29.58 = D = 7 27.84 = C# = 8 26.31 = C = 9 24.78 = B = 10 23.47 = A# = 11 22.05 = A = 12
_________________________
Q: Is it good music? A: I Wish Art

Top




#468249  11/07/03 01:21 PM
Re: Varispeed

Alterboy
Planeteer
Registered: 11/28/01
Posts: 437
Loc: Vancouver, Canada

The multiplier value that I recommended for calculating semitone variations can be taken to two more decimal points ie 1.0594613 to help you more accurately calculate semitone frequency changes.
Here are the new calculated values from 22050 [one octave below 44100] all the way up to 50K.
22050.........A one octave below 44.1K 23361.161.....Bb 24750.288.....B 26222.017.....C 27781.26......Db 29433.22......D 31183.41......Eb 33037.672.....E 35002.195.....F 37083.534.....Gb 39288.635.....G 41624.86......Ab 44100.003.....A 46722.326.....Bb one semitone above A 440 49500.58......B
Another way of roughly approximating the required sample rate change would be to take the pitch [in hz] that you wish to shift to, and add two zeros...
Having said that, here's an interesting sidebar for all interested in sample rates and there possible musical implications: In the chart above, if you place the decimal point two places to the left you will discover that the frequencies are almost identical to the standard of tuning for our musical instruments. Instead of tuning your instruments to A440, if you tune to A441 [a negligible change], your music will “resonate” with the standard sampling rate of 44.1khz. [not that anyone necessarily cares or hears such things...]
This may lead you to wonder why Sony and Philips did not choose a sampling rate of 44,000 when they developed the standard for the audio cd. The answer lies in the way that 16bit digital audio was being stored at the time...onto video tape, a high bandwidth, robust and inexpensive media. [Don't forget that Sony developed the Beta video format and used it for storage with their early pcm recorders.] As it turns out, 44.1Khz works neatly with the framerate and lines of video in the TV signal and it also provides us with a sampling rate that is at least twice that of the highest frequency we need to sample [Nyquist limit, Shannon's theorem]. Perhaps some of you remember that in the early days of cd manufacture [the 80's], digital audio was encoded onto threequarter inch broadcast format video cassettes prior to the making of a glass master at the plant. It's all quite interesting ...

Top




#468250  11/07/03 05:45 PM
Re: Varispeed

SlySi
Planeteer
Registered: 10/15/03
Posts: 38
Loc: Hertfordshire, UK

Thanks. The effect i'm trying to get is similar to The Cure's Pornography album. It appears the drums (and some keyboards) were recorded at a higher pitch so that when slowed down the drums sounds heaver and deeper and the keyboards had a darker tone.
Not tried it yet but I hope it'll work.

Top




#468251  11/07/03 07:06 PM
Re: Varispeed

LakeStone Karl
Planeteer
Registered: 06/26/02
Posts: 15821
Loc: Eastern PA, USA

The playback frequency of successive notes is derived from the 12th root of 2.

Top





21236 Members
25 Forums
144801 Topics
1671090 Posts
Max Online: 290 @ 11/19/19 07:32 PM


