| Author |
Understanding Frequency Ratios, tuning, any tips? |
| Damo303 |
I've been learning FM synthesis, i'm trying to understand Frequency ratios.
If I want to tune my oscillators a fifth apart, thats seven semitones right?
So for example if I tune osc 1 to a 'C1' (using the Jones o'Tool) then osc 2 should be tuned to 'G1'.
Going by the TipTop Freq/Note chart C1 = 33Hz and G1 = 49hz, as these are not mathematically even intervals this will give me an Inharmonic overtone, correct?
Or a Inharmonic overtone is only created by subtracting or adding a constant number? (i've googled the definition of 'constant number' and the explanations are confusing to the say the least, anyone explain it in simple terms?)
Basically i'm interested in exploring the different tones I can get out of FM synthesis and subtractive synthesis, I like bass/lead sounds where artists seem to tune their oscs a fifth apart so I wanted to understand it fully and try the same techniques.
cheers |
|
| VortexRanger |
| The harmonic overtones based on frequency ratios start with one octave, then an octave and a fifth. So if you want to tune to the second ratio of "C1" it should be "G2". |
|
| VortexRanger |
|
| VortexRanger |
| Damo303 wrote: | I like bass/lead sounds where artists seem to tune their oscs a fifth apart so I wanted to understand it fully and try the same techniques.
cheers |
This usually is just two oscillators mixed together before the filter, as on a Minimoog or its descendants, rather than modulating each others' frequency, if I'm understanding what you're saying.
FM ratios (and even more so, additive synthesis) are based on harmonic series because different levels of various harmonics over time are the essence of what gives musical sounds their timbre. |
|
| Mans |
|
| felixer |
| Damo303 wrote: | | TipTop Freq/Note chart C1 = 33Hz and G1 = 49hz |
they are just off: 33*1,5=49,5 ... the difference would be slow chorusing@0,5Hz.
that would be 'just' intonation, beware that the fifth on an equaltempered keyboard (like a synth or organ) is slightly (2 cents) flatter.
this is a problem and it gets worse with the major third and minor seventh. you can get very deep into theory and go pretty crazy (some pianotuners have suffered serious damage), but in the end it just has to sound good and you tweak until it does 
forget numbers and displays (or use 'm to get in the ballpark) and use your ears for finetuning. |
|
| LeFreq |
I just listen!  |
|
| Frankenzappa |
| LeFreq wrote: | | I just listen! |
+1 don't over think fm. just use your ears and have fun  |
|
| lego |
G1 to C1 approximates a ratio of 3/2.
G1 ~= 49.00
C1 ~= 32.70
G1 / C1 ~= 1.498
A general rule of thumb is that simple ratios like 3/2 can produce more harmonic FM sounds than complex ratios, which can create more enharmonic or clangorous sounds.
Fifths sound consonant to us and show up all over the place. Stringed instruments like violins and cellos are tuned in fifths. A fifth can be used in a lead or bass patch without contributing to whether a chord is major or minor, and without creating beats and taking up too much sonic space like thirds can. They're used in "power chords" on guitar and bass. I think fifths can be easy and intuitive to dial in when tuning two oscillators to each other.
To answer your question, yes, a fifth is 7 semitones. You can remember a fifth by counting 5 notes of the scale starting with 1 (C D E F G), and you can count 7 semitones but start with 0 (C C# D D# E F F# G). Yay music terminology.  |
|
| ndkent |
In a nutshell, for FM you want math numbers like 3 to 2 frequency ratios for a fifth.
Don't look up frequencies of notes in Western music. All scales are compromised somehow because you either get pure harmonic intervals and need a new tuning for every key you write music in. Meaning a melody based in the key of C will sound right but start playing the same melody in say F# and it will sound sour or wrong. Or do what western music has done in the last couple hundred years and use Equal Temperament which means all scales are equally a bit off and harmonically compromised but an equal amount so anything interval sounds the same in any key.
Finally when using frequencies to get things in the right ratio, I think going for ones comfortably high is to your advantage. Lower frequencies are a little harder to work with imho and the frequency count may be rounded more when down low. Obviously you hope your VCOs then track |
|
| StoneLaw |
|
| Nantonos |
| Damo303 wrote: |
Going by the TipTop Freq/Note chart C1 = 33Hz and G1 = 49hz, as these are not mathematically even intervals this will give me an Inharmonic overtone, correct?
Or a Inharmonic overtone is only created by subtracting or adding a constant number? (i've googled the definition of 'constant number' and the explanations are confusing to the say the least, anyone explain it in simple terms? |
Forget intervals (subtracting) and adding. Instead think of doubling and halving, ie multiplying and dividing. An octave is a double. Go from there. |
|
| Nantonos |
| lego wrote: |
To answer your question, yes, a fifth is 7 semitones. You can remember a fifth by counting 5 notes of the scale starting with 1 (C D E F G), and you can count 7 semitones but start with 0 (C C# D D# E F F# G). Yay music terminology. |
An octave, from the Latin for eight, so called because it has twelve notes  music terminology  |
|
| Samwise |
| lego wrote: |
An octave, from the Latin for eight, so called because it has twelve notes music terminology |
Uh...if you're talking about scales, then yeah, it's eight notes per octave. |
|
| Captain Coconut |
|
| wednesdayayay |
|
| usw |
| Samwise wrote: | | lego wrote: |
An octave, from the Latin for eight, so called because it has twelve notes music terminology |
Uh...if you're talking about scales, then yeah, it's eight notes per octave. |
7 notes per octave actually  |
|
| Captain Coconut |
| And the eighth is the octave. |
|
| felixer |
| usw wrote: | | Samwise wrote: | | lego wrote: |
An octave, from the Latin for eight, so called because it has twelve notes music terminology |
Uh...if you're talking about scales, then yeah, it's eight notes per octave. |
7 notes per octave actually |
some may think that the chromatic scale (all 12 notes) is the basis of all scales. actually it's the other way around: if you stack all scales on top of each other you'll get the cromatic scale. so start thinking in terms of harmonics. then it becomes clear that the octave is a close relative to the fundamental. the fifth is still family, the major third starts to get a bit off and the minor seventh is no more then a cousin. western harmony doesn't use any more harmonics. the others (minor third, fourth etc) are the result of 'a fifth minus major thrid' or 'octave minus fifth' etc
in chords etc (anything you play on a keyboard) you can often get away with equal tempered but within sounds (like fm of additive) it sounds lousy. just intonation sounds right.
so there's the problem. pythagoras found out over 3000 years ago and since then many clever people (like werckmeister, kirnberger et all) tried to solve the puzzle, but all they could come up with were various compromises.
but yes, do study all those things. it's fascinating, but it will not give you 'the answer' so that everthing will be in tune for all times. because that is impossible. not only practical but also theoretical. so we all end up doing our version of some compromise that sounds 'good enough' for us ... we live in an imperfect universe. but you knew that already  |
|
| Tim Stinchcombe |
|
| n_m |
I love threads like this one! 
I know what I'll be reading the upcoming weekend. Thank you Tim Stinchcombe  |
|
| Don Erskine |
| I just think in integer multiples of 0.0833333333 Volts. |
|
| dkcg |
| i just use my ears and a little experimentation, aka knob twisting. |
|
| felixer |
thanks
music used to be an exact science along with mathematics, astronomy and physics. but beware of greek theories as they sound good but are invariably wrong listening to radiotelescoop output will give you a taste of the 'harmony of the spheres'. sound more like john cage then mozart ... |
|
| Naive Teen Idol |
| Question: is there a Scala scale that aligns to the harmonic series for "easy" ratios in a modular context? If so, I'm thinking you could use, say, Disting's microtonal quantizer mode (algorithm K-5) to set it up. |
|
| toy |
| Frankenzappa wrote: | | LeFreq wrote: | | I just listen! |
+1 don't over think fm. just use your ears and have fun |
Easier said than done. Does this just come with time? |
|
| oberdada |
| Naive Teen Idol wrote: | | Question: is there a Scala scale that aligns to the harmonic series for "easy" ratios in a modular context? If so, I'm thinking you could use, say, Disting's microtonal quantizer mode (algorithm K-5) to set it up. |
Do you want the actual harmonic series, or just any scale that contains the harmonics as a subset? Try just intonation, it's all about ratios. |
|
| Naive Teen Idol |
| Good point! I’ll try that. |
|
| PhineasFreak |
| 4ms smr is the single most amazing accurate tuning source for harmonic, microtonal ad other scales - theres hundreds of scales built in and it'll drive u to 6 vcos polyphonically |
|
| wechard |
| Also, keep in mind that the theory behind FM synthesis and the theory behind tuning systems are two different things, even though the way they both use ratios can make them look similar. It’s possible to make connections between them, but when you’re first learning it helps to keep them distinct. |
|
| starthief |
There are a lot of non-harmonic ratios that can sound great with FM. Really it comes down to using your ears. Octaves or other integer ratios are just the guaranteed easy ones.
I like it when the ratio produces strong undertones which become the new perceived fundamental. Just-slightly-off ratios are also great for creating movement within the timbre, somewhat similar to chorusing. |
|
|
FAQ & Terms Of Use
MW Radio
Twitter
Patreon
Search
Sign up
Log in
