Chances of recreating an FS1r?

What do you think? Is it possible to create an ensemble emulating an FS1r? I had one but it was really hard to use for me. Already the MIDI management with voice and performance modes very quite confusing. But I think this synth is soo great that it needs to be revived.
Or is there already an ensemble? Or is there already someone working on an ensemble? Is anyone interested at all?

thanks
Stiller

 

Stiller,
I still own and use the FS1R, and consider it priceless, surprising, and best controlled by an external controller. Have you got a bank of MIDI faders and knobs you can dedicate to it? It has a unique alchemy of formant and phase modulation that might be hard to emulate. Good luck in your search.
aip

 

I sold mine. Even with SoundDiver I had no fun using it.
But if the experts say formant synthesis is hard or impossible to create with Reaktor then maybe I have to leave the plan. Who else thinks so?

 

Stiller,
I'm no expert, but I'm just saying that of all my hardware synths, the FS1R always surprised me with its incredible versatility. There were times when I would have sworn there were samples of a woman's voice, etc. - but, no samples, just incredible algorithms. Maybe someone else here can help you, but since I've still got the original, I never thought of trying to build it in R5. It is a bitch to program, though!
aip

 

Formant modulation on sawtooth chord.

 

Afaik the fs1r uses the same phase modulation as the DX synth - only more of it and controlled/configured in a clever way (formant sequencing).

Now there are several DX imitation in the UL so I would assume it could be done, but you need to understand the inside of the formant sequencing. I'm not sure how much is explained in the manual (I don't have access to that at the moment).

 

Yeah, I'm sure the difficulty is not so much the FM but the FS section. As far as I remember even the manual didn't go too much into detail on FS. but I just downloaded them and will see, what's in there.
My main question is just, whether formant synthesis/sequencing can be done at all with Reaktor.

@ Jedinhopy
Where is that example from? The Virus TI can sound like that. Or is it an .ens?

@ aip
Yes, the FS1R is amazing! I was also impressed how this thing could sound without any samples. That's why I'd like to recreate it somehow.

 

Formanted strings:

 

This thread comes to mind:

http://electro-music.com/forum/viewtopic.php?highlight=fs1r&t=12628

 

How would you compare the FS1r to FM8? Personally I just LOVE the latter; incredible and fantastic sounding synth (I have no experience with the Yamaha, btw).

 

KyleT,
I haven't got the FM8, but it doesn't have the formant synthesis, if I remember correctly.
The FS1R is an unknown jewel!
aip

 

I have no clue about the fs1r but of course reaktor is capable of "formant synthesis"

a formant is just a resonance with more or less bandwidth and there are several ways to synthesize formants and reaktor is capable of any method, with some built in or part of the factory or user libary

the simplest way is just a bandpass filter, in FM you create a formant when you set the carrier to an integer multiple of the modulator, with the modulator being the fundamental frequency and the modulation depth the formant bandwidth and the carrier frequency the formant frequency of course

there are also PAF, FOF and VOSIM which allow smooth changes of the formant frequency, bandpass filters or
granular enveloping of a wavetable like in the pitchformer sampler module for instance

since the fs1r is from yamaha and incorporates fm i assume the formants are created with fm or bandpass filters
or fof or vosim or pitch shifted wavetables or paf - in that order

exercise :
- explore the formant osc in reaktor>core cells> formant osc
- create a little 2 ops fm synth with modulator pitch set to fundamental F, carrier pitch quantized to nearest integer multiple of fundamental F
- determin which method allows more smooth changes of formant frequency
- extra bonus : find all ensembles in the legacy library or user libraray which relate to "vowels" or "formants"

basically you need 2 or 3 formants to make a vowel like sound

 

For an ensemble that is "inspired" by the FS1r as opposed to an emulation of it I would recommend checking out FMonics v1.3 by Christopher Jon. I've had an FS1r and find that FMonics is a nice compliment whilst not necessarily being able to do the exact same sort of things.

It's probably one of the most used ensembles that I've ever downloaded from the UI. I actually find it much more user friendly than the FS1r even when using one of the dedicated editor programmes.

John

 

jeye_musik,
User friendly, the one term that was missing when they started the FS1R project!
aip

 

Totally agree with Fmonics...also miGraine / Marlene v2.0 I think is the best use of formants I've heard in a synth beyond fake vocal noises.
I believe FS1R was an attempt to bring a kind of FOF + FM synth to the hardware market. Its a nice synth with as mentioned a terrible user interface. The formant sequence part of it though seems under developed
I never really liked the basic sound though either vs older yamaha fm synths, its a lot more "shiny" sounding or something.
The used price keeps going up on them too as no one is ever going to bother making something like that in hardware again.

 

since it seems appropriate in this thread to post somewhat unrelated sound snippets, heres a snippet of an FM based piano I'm just about to finish


100% synthetic reaktor - no samples are used and no phys mod either
its a surprisingly simple 3 sine FM ops patch plus some dsp tricks and careful "pysically informed" parameter choice

this was actually inspired by this thread. I started by exploring fm formants
( but it doesnt use them )
and also by this video [ame="http://www.youtube.com/watch?v=SVY5whlKplk"]YouTube- Alesis Micron Piano FM Patch[/ame]
still I am not sure whats going on there
I like my patch but I would also like to recreate whats going on in that video, cause you cant get those nice synthetic variations on my patch

 

I've been browsing the manual of FS1r trying to understand the concept. It appears to be a classic FM but more... (8+8 operator) but each operator consists of a formant/consonant oscillator/noise with controllable bandpass filters. Then there are some pre-configured sequencers controlling the formant filters to get some human sounds where much research seem to be put into.

A starting pint could be to use a FM ensemble and transplant the ensemble Vowels replacing the sine.

I think the formant sequencer is the challenge. To set up some event table is not the problem but to make them sound human...

 

does this sound human ?
its from an abandoned attempt to create a singing synth vox
one trick here is the pitch glide which is different from normal glide more like real voices behave

 

hey man what is that video?
is that you in the video or something?
explain that video to me

thanks man

 

I dont know this guy i just found the video when I googled for piano & fm
I thought maybe some one finds inspiration in this

right now I building my own fm synth
theres a good one in the library btw by thomas goebel but still not up to my taste
also way too tricky to edit

anyways this thread has been an inspiration for me
I always hated fm just till the other day

I wasnt aware of its potential when you do it right

also I have some sort of sound fetish about broken artificial piano sounds
not the usual dx epianos though, but I did several piano patches on my supernova for instance
its like I m looking for a sound that doesnt exsist
the blue flower of piano timbres

but I dont want to hijack this thread