Jump to content
IGNORED

FM Synthesis (techniques, anecdotes)


Guest skibby

Recommended Posts

 

oversampling does nothing to smooth things out really, because everything has to be downsampled again anyway, so its like working on a TIFF in photoshop and saving it as a JPEG anyway. Unless of course your DAW is running at 196, in which case its cool if the FM synth is running at that SR.

disagree. calculating at a higher sample rate will always yield better quality. you just need to know if it's different enough to bother with the extra cpu. using the image example again, imagine working on a photoshop file that will be saved to 512 x 512. if you're doing repeated processing at this resolution it will not look as nice as if you were doing it at 2 or 4x that size.

I totally respect you but, it depends what you mean by 'quality' since audio is sort of like ear heiroglyphics.

Link to comment
Share on other sites

 

 

 

oversampling does nothing to smooth things out really, because everything has to be downsampled again anyway, so its like working on a TIFF in photoshop and saving it as a JPEG anyway. Unless of course your DAW is running at 196, in which case its cool if the FM synth is running at that SR.

disagree. calculating at a higher sample rate will always yield better quality. you just need to know if it's different enough to bother with the extra cpu. using the image example again, imagine working on a photoshop file that will be saved to 512 x 512. if you're doing repeated processing at this resolution it will not look as nice as if you were doing it at 2 or 4x that size.
I totally respect you but, it depends what you mean by 'quality' since audio is sort of like ear heiroglyphics.
i mean mathematical precision that the user will ideally hear as a smoother and less aliased signal

 

 

i know u know your stuff, and u know how deep i've been dipping the old wick in phase modulation too. I do agree that there are cases in which oversampling are useful in the process, but just where to put the oversampling is really a good question. firstly, oversampling is i think going to become obsolete, because i believe the samplerate of audio in the future will be edging its way higher and higher as a standard. we are fast approaching a place where oversampling won't be needed because there won't be anybody who can hear the difference between 192khz and higher, etc.

 

the only real reason to use oversampling with FM is when someone foolishly applies a high frequency operator to another high frequency operator at high levels, and the resulting signal goes beyond nyquist. it's sort of like an arm throwing a ball that has another arm growing out of it throwing another ball. the thing is, if that happens with or without oversampling during the modulation, the output will still sound like noise anyway once it exits the algorithm. I don't always like to hear aliasing, and believe me I will break out a scope to look at waveforms if I hear problems.

 

I had massive problems while learning to code my own phase modulation software until I unlocked the concept of interpolation. Interpolated delay is the key to smoothness because oversampling itself has no contribution to offer in the actual generation of new samples to smooth out digital stairs. I found that the stairs are most noticable at lower frequencies, when the original signal is being stretched out, since that is when the D to A converter is outputting the values at the samplerate of the device. I spent loads of time fiddling with 2X, 4X, 8X oversampling at practically every step of the way until I realized that the main problem with slowing down audio is the fact that there are missing values that do not equal the previous or next queued sample.

 

oversampling, i believe has it's rightful place when calculating hypernyquist stuff. But the only time I can hear hypernyquist stuff is when it's slowed down. There are of course several algo's for that as well, and they often munch CPU with, at least in my experience, nominal returns especially when it's in a mix.

 

In which place do you find that oversampling is best?

Link to comment
Share on other sites

Never got that CTRLR to work properly

 

yeah it took me AGES to get it to work. I had to go pester the dev, and he actually fixes usually any bugs that pop up, as well as adding features if they are no brainers.

Link to comment
Share on other sites

  • 7 months later...

Which hardware FM synth do you people think has the best programming interface?

 

I think the Reface DX looks pretty well laid out and direct. Other than that, the Monomachine is very hands on and quick (it has a reduced parameter set, take it or leave it). The Stereoping programmer opens up the TX81z pretty well, but lacks a few of the parameters and is a bit unresponsive due to the MIDI traffic throttling.

Link to comment
Share on other sites

Guest Chesney

Always wanted a DX7 as a kid but since getting into making music again later in life I was never bothered about trying FM, the internet is geared to put you off because most of the time the people talking on the internet are old guys with analogue elitsm or stupid people who are just looking for stuff to make the music for them with no effort (huge generalisation I know). Anyway, I bought a DX100 on a whim just because I have always loved the sound but was put off by the interface. I was surprised that when i got it, it was so easy to program. Of course, to a person new to synthesis overall, it would be daunting. The interface is quick and simple.

So, any DX is easy (never tried rack TX's) The monomachine is really easy too and a bit more visual.

I say just dive in.

Sorry for the tldr rubbish.

Link to comment
Share on other sites

Yeah, if you know what's going on and start to acquire some muscle memory on the buttons/slider, the DX synths can be quick to program.

 

But it would definitely help if they had more than one data slider and maybe a visual presentation of the EGs. The reface lets you edit four related parameters at a time, and gives you a visual clue on what's going on other than a value between 0 and 127. =)

 

Recently saw this DX21 series on Youtube, which makes it look pretty quick and easy to work on:

 

 

 

 

 

Link to comment
Share on other sites

  • 3 weeks later...

Which hardware FM synth do you people think has the best programming interface?

 

I think the Reface DX looks pretty well laid out and direct. Other than that, the Monomachine is very hands on and quick (it has a reduced parameter set, take it or leave it). The Stereoping programmer opens up the TX81z pretty well, but lacks a few of the parameters and is a bit unresponsive due to the MIDI traffic throttling.

 

Link to comment
Share on other sites

  • 3 months later...

Hello,

 

Has anyone had any recent experience with the FS1R editor called ZeeEdit?

 

Both in terms of recently purchasing the editor, and/or recently using it?

 

 

I was given an old HP desktop box that I was thinking about using as a dedicated box to edit the FS1R with.

 

I would be using an M-Audio MidiSport 4x4 to communicate with the FS1R.

 

Also, it would be nice to send the FS1R simple musical sequences while editing the sound parameters, so I would imagine I would need some sort of midi merger thingie?

 

Thanks for reading,

Al

Link to comment
Share on other sites

I love fm so much! FM8 is probably my favorite plugin(with absynth 5 maybe) i made entire drill n bass tracks with only FM8 for synths.

Like 20-30-40 instruments tracks of only FM8.Its wonderful,expressive and can get sounds that are out of this world!

 

Also i just got a DX7 few weeks ago in a trade for a Minibrute with a friend,and i love it with passion.Its amazing. Hardware takes the FM/PM sound to an even higher level! It feels more alive than FM8.You really feel its coming from a machine,sounds are more organic and powerful!

 

When you go beyond the usual boring presets and dig into the machine a wonderful world of FM lushness opens to you!

 

It can go really wild and experimental!

 

2 recordings with 2 cool presets i made with Dexed.

Syroish bass.mp3

dx7 insects mastering mp3.mp3

Edited by fxbip
Link to comment
Share on other sites

  • 7 months later...

i honestly really like operator

How does it compare to the Yamahas or the Monomachine (just cos those are the ones I know), or any other FM you've farted around with?

 

Dave Noyze made a little tool for PreenFM2 to generate Farey sequence family ratios. It could be used to manually program Yamaha stuff also.http://ixox.fr/forum/index.php?topic=69272.0

Thanks for reminding me about Farey ratios. I was thinking about this stuff and I wasn't sure whether it had a name and then I remembered talking to Entorwellian about that awesome Truax site about a year ago: https://www.sfu.ca/~truax/fmtut.html
Link to comment
Share on other sites

It's in between mnm and Yamaha in terms of fm complexity.

 

It has 4 op and around 12 algs, but you can set each op to any wave. Sine is best for classic fm but sometimes it's nice to have a triangle in there for a harmonic boost.

 

The envelopes are pretty deep. The lfo goes well into audio rate, and the kicker is it has a feep filter section so you can use it as a 4 osc subtractive synth.

 

Also has a mod matrix and operator feedback (only positive feedback, no negative).

 

My favorite control is the time parameter which globally affects all envelopes plus or minus the current setting.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.