Forum index > Feature requests

individual step length per layer in voice sequence


New to this forum, just want to say the Essence FM looks like an amazing implementation of fm! Been very close to purchasing one... but have been waiting out for more information on the new keyboard synthesisers/synthesis engine.

In the meantime I've been reading the manual and watching videos of people using the voice sequencer to try and understand how it works. Is the voice sequencer is limited to the 4x4 grid, or is it possible to have individual lengths per layer in the voice sequencer?

(i.e.... layer 1 loops@ 3 divisions, layer two@ 5 divisions, layer three@ 8 divisions, layer four @ the full 16 divisions etc...)

If individual layer loop lengths aren't currently possible, would that be a viable feature request? It would allow for a really quick way to build up cross-rhythms or to do other timings that aren't divisible by 4 or fit into 128 steps whilst still being quantised.

(conditional trigs would be amazing as well.... but thats maybe asking a lot!)

All the best, and look forward to hearing more about the new synthesisers!

The voice sequencer is a single grid of 128 steps (or 4 divisions, or 8, 16, 32, etc. depending on how you see it). Individual length per layer is not possible, but you can pretty much do the same by stacking multiple patches inside a Performance, with different tempos for each voice sequencer. (the voice sequencer isn't global, it's per patch)
Some conditional trigs can be achieved with the modulation matrix or with layer properties (round robin, exclusion groups, velocity ranges and key ranges)
I really want to see 4 pages in the voice sequencer for longer/varying patterns. I see it as a 16 step sequencer with micro timing.
Jazz Paladin
It would also be cool to adjust velocity per note in the sequencer
Yes I have often wanted to do that too.
Jazz Paladin
Quote : (the voice sequencer isn't global, it's per patch)

So far , I seems to me that it would also being nice to have the voice sequencer per performance instead. If I want to use the same voice in another song it essentially means I have to creat another patch for it just to be able to have a separate sequence for it if it is already tied to a voice sequence in another song .
Yes. Sequences could have been stored separately in a bank so any element (patch, perf, voice) could use them, but that kind of modular approach also increases complexity for the user.

Try to think about Patches being the real "Voices" of the EssenceFM, not just an annoying step you have to go through. "Voices" are really meant to be designed using multiple 6-OP sounds with various volumes, pannings, tunings... So if you need to use the same "Voice" (Patch) in another song, all that requires is doing a copy-paste of the patch. That will keep the reference to the same voices (because voices aren't stored inside patches, they're only references), while creating a new sequence (because the sequence is stored inside the Patch)
Thanks so much for the detailed reply on this. 🙂

Not having an essence FM does really limit the scope of imagining what the voice settings and multiple layers at different tempos can achieve. Going to keep saving up, also patiently waiting for updates on the new instruments!

That being said, it does seem complex to have to jump back and forth into separate voice sequencers to achieve pseudo alternate time signatures, as 128 will never divide into 3 without programming 'off the grid' (Just thinking about how this could become messy when tempo syncing the essence FM to other instruments!).

Realise that this is all hypothetical, but maybe there is a way to alter the sequencer to have an option to loop(reset) the entire sequence @ 3x4 i.e. 96 steps? (though any step length from 1-128 would open up so many possibilities!!!). 96 steps would unlock time signatures in 3 within one single sequence without having to play around with offsetting tempos in separate voice sequencers.


Tangential, but just thought I'd share as this discussion is reminding me of a web based 'multiplayer' '16 step' sequencer some of my friends developed for 6 fairly basic FM 'voices' last year, though its more like a variable 1 to 16 step sequencer. It was really fun to perform and compose with and opened up many musical paradigms; a lot of this was done with fairly simple controls over the sequencer:

adjustable step range controls (any step could be the start of the pattern and the loop could be any step length),

fractional multipliers on the tempo (.125,.3,.5,.75,.833,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15) so steps skipped and double played in really interesting ways depending on the sequence step length

forward-reverse-rebound(brownian) directional control for the sequence

Here's a link to a short performance

link to the website...doesn't seem to be working at the moment :(

github of the project: