Elektron-Users

another vote for a dedicated granular machine.
I have an MDUW and that feature would make me get an OT in a snap.
granular with jitter settings, lfos and parameter locks, combined with the scene function would make the OT such a killer for experimental sound design...

JES wrote:
I was all excited to see how people were getting cool granular fx with the OT but those videos were, well, not the OT.

The main thing I want is some ability to freeze audio, scrub, add jitter, add density and control grain size. I basically use it to make clouds--polyphony isn't a big issue. The Monolake granulator is cool but a granular machine could be simpler like audio ease Riverrun or density. OTOH, adding simple FM -- one step up from the comb filter -- would be amazing.

Also would really like a granular machine, just having control of grain-size and density instead of the retrig parameters would be great. Perhaps one fx slot could be traded for more extensive controls like pitch envelope, size, amp and pitch jitter. Would be sooo awesome. This was really what I expected the OT to have.

"Radical Audio Manipulation Possibilities" kind of implies, radical, not, vanilla (which is what I consider it to be)

Another thing that would be cool and totally doable would be a phase-vocoder buffer-sampler machine. That is, a machine like static machines that play back a pre-fft analysed sample, so you could smear and process it in various ways (you could have a special set of effects for this). FFT-analysis takes time (introduced unavoidable latency) but if you already do it on the entire sample when you load it up then you get around this issue.

Also, I think elektron should purchase an elastique or other formant-preserving (and modulating) time-stretching algorithm license. Although this old skool 90's akai stretch is kind of cool, so many modern daws and software support this PSOLA approach, I find it kind of ridiculous not to have it in a machine which is only a sampler.

DSP dimension also has some cool algorithms that would suit the "radical" description:
http://www.dspdimension.com/technology-licensing/mcpt-synthesis/

Also, I like the reverb, but I think we could benefit from some more types, that is, just different allpass topologies, the one they have is great for medium size rooms but for large lushness something with more smearing would be great. I think inserting a pitch-shifter for shimmer-type effects would be great too. Perhaps some kind of reverb/multitap delay hybrid with frequency or pitch-shifting feedback. That way all manner of strangeness and or shimmering could be achieved (something like valhallaübermod).

And another thing, delay with tape-emulation (tape age and such would be an extra parameter). This is really so simple to achieve really, just delay with interpolated slewed delay time adjustments (to simulate tape speed change) and some random-modulated slow and fast modulation of delaytime (to simulate fast and slow mechanical modulation of tape speed) insert a biasing error on the write pointer and some eqing that's also a bit random modulating and you have a very convincing tape emulation.

If you haven't seen this I think you should:
http://www.youtube.com/watch?v=_Xy0szkt7ao
(the father of granular synthesis)

Veqtor wrote:
JES wrote:
I was all excited to see how people were getting cool granular fx with the OT but those videos were, well, not the OT.

The main thing I want is some ability to freeze audio, scrub, add jitter, add density and control grain size. I basically use it to make clouds--polyphony isn't a big issue. The Monolake granulator is cool but a granular machine could be simpler like audio ease Riverrun or density. OTOH, adding simple FM -- one step up from the comb filter -- would be amazing.

Also would really like a granular machine, just having control of grain-size and density instead of the retrig parameters would be great. Perhaps one fx slot could be traded for more extensive controls like pitch envelope, size, amp and pitch jitter. Would be sooo awesome. This was really what I expected the OT to have.

"Radical Audio Manipulation Possibilities" kind of implies, radical, not, vanilla (which is what I consider it to be)

Another thing that would be cool and totally doable would be a phase-vocoder buffer-sampler machine. That is, a machine like static machines that play back a pre-fft analysed sample, so you could smear and process it in various ways (you could have a special set of effects for this). FFT-analysis takes time (introduced unavoidable latency) but if you already do it on the entire sample when you load it up then you get around this issue.

Also, I think elektron should purchase an elastique or other formant-preserving (and modulating) time-stretching algorithm license. Although this old skool 90's akai stretch is kind of cool, so many modern daws and software support this PSOLA approach, I find it kind of ridiculous not to have it in a machine which is only a sampler.

DSP dimension also has some cool algorithms that would suit the "radical" description:
http://www.dspdimension.com/technology-licensing/mcpt-synthesis/

Also, I like the reverb, but I think we could benefit from some more types, that is, just different allpass topologies, the one they have is great for medium size rooms but for large lushness something with more smearing would be great. I think inserting a pitch-shifter for shimmer-type effects would be great too. Perhaps some kind of reverb/multitap delay hybrid with frequency or pitch-shifting feedback. That way all manner of strangeness and or shimmering could be achieved (something like valhallaübermod).

And another thing, delay with tape-emulation (tape age and such would be an extra parameter). This is really so simple to achieve really, just delay with interpolated slewed delay time adjustments (to simulate tape speed change) and some random-modulated slow and fast modulation of delaytime (to simulate fast and slow mechanical modulation of tape speed) insert a biasing error on the write pointer and some eqing that's also a bit random modulating and you have a very convincing tape emulation.

If you haven't seen this I think you should:
http://www.youtube.com/watch?v=_Xy0szkt7ao
(the father of granular synthesis)

Curtis Roads did all that by hand or wrote software that took forever to run to accomplish his goals...By turning it into some cheesy effect on mass market hardware it becomes "vanilla" the more polished out of the box and effect is the more quickly it becomes boring and cliche.

If you spend time mapping out parameters to LFOs and scenes you can create unique sounding effects. For example, you can use multiple narrow band focused EQs, re-sampling and comb filters linked to one tracks LFO designer, but each tracks LFO inter modulated by it's own LFOs to get quivering, smeared, spectralish sounds.

If Elektron focuses on having the "best" quality effects it will quickly be superseded by mass market computers as CPUs, HDs, etc. increase in performance. On the other hand, if Elektron focuses on interface, ergonomics and playability, they are creating a device that has timeless usability. I mean, Übermod is awesome, Melodyne is rad and I "sequence" both with the OT.

I don't mean to be such a grumpy dick, but I'm just slightly frustrated that people aren't willing to abuse the shit out of the OT to accomplish what's already possible. I know it's personal preference that I appreciate the building blocks more than the pre-formed kit and if Elektron had the resources to implement both I wouldn't even bother writing, but as it is I'd rather see lower level solutions....

But, yeah, I would like to see audio input followers and analysis that could be used to modulate parameters...which could be used as a kind of vocoder, but also a million more things besides. How cool would it be to link envelopes from internal and external audio to things like playback slice, tempo, and various effect parameters!

I also agree that sample to analysis for things like FM and AM modulation would be highly neat.

Anyway....

Hi Jonah,

You keep saying granular synthesis is already possible on the OT. Perhaps you could post some demos of its existing granular powers? I realize it's a powerful machine (while Vector has lots of good points, I wouldn't call it "vanilla"). I've just had zero satisfaction making decent grain clouds with it and it seems like the engine is perfectly suited to the task with a few small modifications. I will try the one recipe you posted when I get home in a few weeks.

Alright, I'm a Octatrack owner for 5 days now and I already have feature requests
The main reason I bought the OT was for looping my guitar and my vocals, both signals stereo because of effect processing before entering the OT. Unfortunately there are 2 things that keep the OT away from being perfect for this purpose (and therefore keeping me from looping at all. I'm torn between selling it again and keeping it for learning more about sampling which I discovered can really be fun, but it's quite an expensive (and advanced) "Toy" )

1) AB to Main Out, CD to Cue Out
The Input CD cannot be routed directly to Cue Out, so I'm not able to have both stereo-signals monitored on different stereo channels.
2) "Combine" two Pickup Machine (Tracks)
There is always one Pickup machine which is the "master" and all other pickup machines slave themselves, did I understand that right? Well, sometimes I want to loop both signals (guitar+vocals) at the same time. I can do this by taking an extra pickup machine and configure it to get Input from AB and CD but then they are getting summed in the track of course, and that's what I want to avoid. So, it would be awesome to be able to "connect" or "combine" two pickup machine (tracks) and record and control them at the same time.

These first two things would turn the OT into a real advanced looper because you can work with two stereo signals completely independent.

Furthermore I have some more ideas to improve the OT:
3) More advanced control over the Pickup machine
For example: I'd like to record a new loop and STOP it without it starting to loop, playback or overdub. So I can for example play my first verse and my pickup machine is recording it. Then I play my chorus, where I stop the machine and finally, when I come to verse 2 I play the loop back so I can overdub and therefore make verse 2 sound a little bigger than verse 1.
4) Midi Control in general:
I'm using 1.2c at the moment, and I ran into problems with Midi controlling the pickup machines. I sent a view note commands with Ableton Live to test it, and I was able to start Recording but then I always got a "Dub aborted" error message. Is it my fault or is there something wrong with Midi Control?
It would be nice if the OT gets even more Midi controllabilty (not only Pickup Machines). I'd like to activate scenes, patterns, parts, .... everything with midi, will that be possible one day?

Thanks for your attention and sorry for my English

Best regards
ben

EDIT:
One more thing came to my mind this morning:
will it be possible to make the pickup machines record to the CF-Card instead into the RAM to increase the recording lenght by a million times? The Electro Harmonix 2880 Super Looper for example does that so it should actually be possible?!

When setting a pattern length which is less than 16 steps(for example 5) I would definitively like to see an option to copy these steps across the 64 steps or an an arbitrary number. Its so cumbersome to repeat a 5 step sequence.
Some of you might ask why would you want to do it since the five steps keep looping in sync with the other tracks. Merely because I would like to introduce some variation inside the looping steps.

Being able to use a midi controller like the MPD32 to trigger slices.

The ability to record from the main And cue outputs at the same time. An all-outs record. For sure!

Brought up earlier...we need the ability to adjust the start marker in the audio editor without shifting the end marker, if that's not already available and I'm somehow missing it.

c0rpse wrote:
It would be nice if when you held down a p-locked trig key the pages with locked parameters lit up. Similar to how scenes work.

Noisebuddy wrote:
Just a Send-Machine

Isn't this the cue ?

no ....

cue outs for other things

i thought a more general fx-machine with different send-return-levels to the parts could be usefully .... internal!

but anyway ... it´s not that important

the midi-thing is very interesting for me .... i enjoy it to sequence my two shruthis and my blofeld

but ..... to blend in the midi-LFOs with the fader (or generically midi-scenes) would be nice .... even with the fader which has a much higher data-resolution compared to midi, in theory it should be possible to calculate the fadermovement-data down to a midi-event in a specific time-window

oh yes and a few more FX

btw i love the Delay it sounds great