I think I changed my mind

[Berk]

I think the pads in this one is FM. Definitely worth

 

[dingformung]

1 hour ago, chim said:

All algos aren't equal, plus DAC's and transformers play a part in the sound. Goes for the old ROMplers (wavestation, d-50, JV) vs reissues/VST's as well. 

Sure, even the most miniscule rounding error can change the sound audibly when it comes to FM but you tweak the sounds to your liking anyway, so that isn't really important. If you do it with software you have full control over the algorithm, make your own algorithm etc. There really are no advantages in having a digital hardware FM synth. If you like touching knobs (hrhr) just use a MIDI controller to control the software. But what do I know

[Silent Member]

[Braintree]

The NI's FM8 and the Digitone are two of my favorite synths. FM is usually an integral part of my process.

[dingformung]

The Digitone is sexy, admittedly

[xox]

Monomachine’s FM machines look quite limited but they sound amazing to me.

[Ginerpri]

FM is super cool but to me it always needs some "organic" to be added into, like DAC color, distortion, resampling from speakers, etc.

At the moment I like to make patches on Ableton's operator, mostly bass, drums and pads, then sample them and play with filtering, distortion, sample start and so on on the Model Samples. Thinking of buying a hardware fm synth, though. There are many cool options.

There was this Synset by Fingersonic in the works which looked promising but if I remember correctly they guy behind the company was hired by Behringer and ended the project.

With some patience you can achieve a lot of cool stuff, like this (not me):

 

 

[digit]

On 5/17/2020 at 12:36 PM, zkom said:

Is there actually some reason to have separate FM hardware gear and not just run software on your PC?

have a listen to a FS1R and you will have the answer to your question.

 

 

Edited May 20, 2020 by digit

[zkom]

23 hours ago, dingformung said:

On 5/19/2020 at 9:15 AM, chim said:

All algos aren't equal, plus DAC's and transformers play a part in the sound. Goes for the old ROMplers (wavestation, d-50, JV) vs reissues/VST's as well. 

Sure, even the most miniscule rounding error can change the sound audibly when it comes to FM but you tweak the sounds to your liking anyway, so that isn't really important. If you do it with software you have full control over the algorithm, make your own algorithm etc. There really are no advantages in having a digital hardware FM synth. If you like touching knobs (hrhr) just use a MIDI controller to control the software. But what do I know

Yes, well basically if the sound is colored by hardware imperfections I see the point. Porting the algos to run on a PC shouldn't be much of a problem with modern PCs. You could probably even do them with some high precision 128-bit floating point arithmetic to get an ultra-clean and precise sound. But if the point is to add some grit then I can see the point of using hardware.

Anyway, the clean and mathematically accurate way of making music would be an interesting aesthetic to try sometime. Usually it comes out as sounding very dry and jarring, but I think this is because not many people have really explored the possibilities and how far that particular rabbit hole goes. Just use pure waveforms, white noise, FM and additive synths, and things like that, with no distortion, compression or analog modeling. Get it sounding interesting some other way than adding analog sounding imperfections.

[Limo]

1 hour ago, zkom said:

Just use pure waveforms, white noise, FM and additive synths, and things like that, with no distortion, compression or analog modeling.

I might be wrong, but I think that is the reason why Max and Pure Data sound as bland as they do by default: because they use mathematically pure wave forms and white noise. Reaktor is much better in that regard because they strove to implement sound generation that mimicks what happens in hardware.

Going for an ultra clean, mathematically precise aesthetic is not a new rabbit hole to explore, it's something that has been tried and found lacking.

Unless, of course, you're Autechre and your music is so complex and detailed it doesn't matter that the basic sound generation is bland.

 

[zkom]

Just now, rhmilo said:

I might be wrong, but I think that is the reason why Max and Pure Data sound as bland as they do by default: because they use mathematically pure wave forms and white noise. Reaktor is much better in that regard because they strove to implement sound generation that mimicks what happens in hardware.

Going for an ultra clean, mathematically precise aesthetic is not a new rabbit hole to explore, it's something that has been tried and found lacking.

Unless, of course, you're Autechre and your music is so complex and detailed it doesn't matter that the basic sound generation is bland.

I agree it's not a new avenue but I think there might be something there because people gave up too early. Generatively add lots of tiny details for example? Or adding a ton of minor parameter automatization to create movement? Complex sidechaining hierarchies or cyclic sidechaining graphs? Doing some frequency domain fuckery that is not possible with realtime plugins. It would probably sound alien as fuck but maybe also really cool.

[Limo]

15 minutes ago, zkom said:

I agree it's not a new avenue but I think there might be something there because people gave up too early. Generatively add lots of tiny details for example? Or adding a ton of minor parameter automatization to create movement? Complex sidechaining hierarchies or cyclic sidechaining graphs? Doing some frequency domain fuckery that is not possible with realtime plugins. It would probably sound alien as fuck but maybe also really cool.

Isn't that just called "IDM"?

"IDM" always struck me as "synthesizer music - but not as boring". Lots of detail and trickery to maintain interest. 

[zkom]

9 minutes ago, rhmilo said:

Isn't that just called "IDM"?

"IDM" always struck me as "synthesizer music - but not as boring". Lots of detail and trickery to maintain interest. 

It could be simply to add movement to sound so it doesn't sound static and feels more organic. What the non-linear analog components and things like distortion do is basically change the harmonic content of the sound based on the amplitude of the waveform. DSP effects like low/band/hi-pass filters, amplifiers, delays etc have linear response and the harmonic content does not change based on the amplitude of the waveform. So the digital delay can sound more boring than the tape-delay because it's just adding sounds linearly and does not saturate on higher amplitudes which would create new harmonics and it also doesn't have the frequency filtering when the signal is fed back. Now what if we created completely new ways of creating harmonic changes digitally and not just try to model the analog components?

Just thinking while writing.. Sorry for derailing the thread.

[dingformung]

I thought the same. You need less harmonic distortion on FM sounds because usually they are rich harmonically already in their pure form. Adding FM to the kick naturally makes it resonate over a larger frequency spectrum and can be used as a replacement for distortion, makes it sound thick. A kick thats just a sine wave plus some filtered noise sounds much better with harmonic distortion, an FM kick on the contrary doesn't necessarily need it. Also, what if you have a kick with a bass sound at 80 Hz and instead of creating a harmonic of that exact same sound at 160 Hz you place a completely new sound in that frequency region that complements the bass sound but in a more interesting and less static way? And yes, jitter can help adding interest. Though for mixing and mastering saturation is essential, no reason to not use it. Very pure and clean sounds usually sound good when saturated. But again, you don't need hardware for it and you can use other forms of distortion than the good old tape saturation or other analogue style sorts of saturation, e.g. phase distortion. A stylistic choice. 

 

[modey]

1 hour ago, zkom said:

people gave up too early

What? There are still tons of artists doing super precise, clean digital stuff. Ryoji Ikeda and Mark Fell, for example..

[zkom]

6 minutes ago, modey said:

What? There are still tons of artists doing super precise, clean digital stuff. Ryoji Ikeda and Mark Fell, for example..

Thanks, would you recommend some particular releases to check?

Actually, I also think probably Frank Bretschneider's releases and Cylob's Ambient News project kind of fit this category.

I'm getting curious now what kind of stuff I can find that is pure digital and doesn't even try to sound analog.

[Freak of the week]

24 minutes ago, zkom said:

I'm getting curious now what kind of stuff I can find that is pure digital and doesn't even try to sound analog.

some of my favs:

frank bretschneider - rand

ambient news - zweite sendung

ryoji ikeda - supercodex

mark fell - multistability, ul8

alva noto - prototypes, transform

 

academic/noisier stuff:

xenakis - gendy3, s.709

herbert brun - sawdust

barry truax - riverrun

hecker - recordings for rephlex

yasunao tone - solo for wounded cd, mp3 deviations #6+7

curtis roads - point line cloud

haswell & hecker - blackest ever black, kanal gendyn

 

also, you might wanna check my sig :)

Edited May 20, 2020 by Freak of the week

[Nil]

@zkom you're in for a treat. As already mentioned, Ryoji Ikeda and Alva Noto make fantastic music, but are more about raw test tones and turning all sorts of data into audio than synthesis per se.

Mark Fell (venturing as well in additive synthesis for Editions Mego, dope Manitutshu EP !), his son Rian Treanor (Ataxia is a marvelous record, full of crystal clear FM tones and proper mind bending musical mathematics), his collaborator Gabor Lazar (the additive synthesis don) are top notch digital musician. I'd also recommend wholeheartedly Kindohm (livecoding + aditive synthesis) and William Fields (algorithmic x FM voodoo) as amazing limit-pushers as well. Both release records on Conditional, which catalogue is getting more awesome (and then some) with each and every new release.?

On a ambient tip, my personal fav is Ssaliva (We never happen being my fav record of his), but I suspect he's more about additive synthesis, sample mangling and resynthesis than FM.

[pizza]

9 hours ago, digit said:

have a listen to a FS1R and you will have the answer to your question.

 

 

do you have experience with one? would love to hear some thoughts

[garbage_burner]

+1 for the megafm. Definitely not for everyone but it does have a magical quality due to the dac distortion. The cc integration and MPE are solid. I’m using a sonoclast plastic pitch plus and I can tune every note. Instant lp5 sounds 

[TubularCorporation]

9 hours ago, zkom said:

Anyway, the clean and mathematically accurate way of making music would be an interesting aesthetic to try sometime. Usually it comes out as sounding very dry and jarring, but I think this is because not many people have really explored the possibilities and how far that particular rabbit hole goes. Just use pure waveforms, white noise, FM and additive synths, and things like that, with no distortion, compression or analog modeling. Get it sounding interesting some other way than adding analog sounding imperfections.

There was a lot of academic electronic music like this in the 80s and 90s especially. Lots of stuff done in synthesis languages like M, CSound, proprietary stuff, things coming out of places like IRCAM and The Columbia-Princeton Electronic Music Center, etc. I don't really have many examples I could give off the top of my head since most of what I've heard doesn't do much for me and usually sounds more like tech demos IMO, but it's out there, and a lot of digital synthesis techniques that are common now came out of it.

 

Paul Lansky's CMix stuff is pretty good.

 

 

Edited May 20, 2020 by TubularCorporation

[sheatheman]

One of the coolest things about idm FM sounds is a lot of them were made on the mnm, and fm synths sound amazing with parameter locks. Lots of cool AE stuff sounds like it does because of the mnm.

course they can do all that in max now

which brings me to operator. Truly a wonderful synth, but ableton’s parameter locking capabilities are quite terrible. They are on a step basis and not a note basis, basically each parameter has a global automation curve per track. You can mess with step length to try to get it to sound like a monomachine, but it’s not the same.

which brings me to my last point:

there’s a new iPad app called Drambo, which is like a combination of an octatrack with a fully modular environment. I’m getting some of funnest FM stuff I’ve ever gotten from this app. Sounds good too. The parameter locks are just like elektron, but the sequencer is perhaps even more advanced. Highly recommend. I might even upload a video. 

Edited May 20, 2020 by sheatheman

[YangYing]

fart modulation

[modey]

2 hours ago, sheatheman said:

One of the coolest things about idm FM sounds is a lot of them were made on the mnm, and fm synths sound amazing with parameter locks. Lots of cool AE stuff sounds like it does because of the mnm.

They only used the monomachine on two albums. I think most of their FM stuff came from either the Nord Lead or DX100. 

Nord Lead FM is amazing btw! Only 2op but it has a really unique character. 

[sheatheman]

Quaristice and what else?

I’ve never used a nord lead.