Qu-Bit Nebulae V2

i do have kontakt and soundtoys complete bundle, but cycles is way too expensive. im sure its worth it though. Kontakt is worth it for the Output engines and una corda alone. but its also great for doing granular sampling across a mapped keyboard with its own sample engine. i like the time stretch options it has (time machine, etc).

that stereo panning is a really good idea. i love echo boy. crystallizer would be cool for this too i think. i used that one all the time for a while.

I’ve been trying to figure out how to order my effects chain. the thyme and ventris are stereo and the ventris has an echo engine and an “offspring” engine that sort of has a weird tremolo spring effect that might be cool. right now i have Digitone (with 0-coast and cue out from Octatrack running through it’s two external audio inputs)>Thyme>Outward>Ventris. I had the Ventris after the Outward before i got the Thyme because I didn’t really want the Clouds-like smearing effect and i thought that granulizing everything in the signal chain would be cooler than washing it all out with reverb. but if i go Outward>Thyme>Ventris (or even swap Ventris with Thyme) I may be able to do something like this.

Thyme has a reverse polarity for the stereo outs which might produce some cool results with the grains from outward. it also has external clock sync and i can p lock the preset sequencer to my octatrack midi channel to control it rhythmically. then if i want, i can further spread them out with some careful ventris routing. I realize this could be much easier using them as send/return effects but my cue outs are full and i get nasty feedback when trying fx send with the octatrack. maybe something like the vfe klein bottle would be nice in the future.

i love the bitwig sampler. the internal modulation is great and the ability to modulate the layers or whatever they’re called (voices maybe?) is incredible. freeze mode is very nice on that too. yeah i love the bitwig sampler. I have almost every software granular plugin and standalone that exists, i just want to have a nice hardware equivalent. but this is a very good point

@kveye yeah it seems like the magic from cycles is the way it sequences and modulates your grains automatically, which is kind of hard to do manually. I like the Outward pedal because it has self modulating lfos, which is very nice. its like having a slightly more limited clouds without the need for external modulation, but you can do midi or cv modulation via expression port or midi trs port.

The clouds module for vcv has taught me a lot about how to make it really work to get nice results. i usually put a kind of erratic but slow lfo on the position, have a slow random gate on the freeze, sometime modulate density and texture, but not much, heavily attenuated envelope on the size and its pretty nice to just play into with like a random freeze loop with automated playhead scrubbing

so i understand that they need modulation to really add movement and make them come alive. I see your point though, and i think that’s a good thing to remind myself of. that i can probably achieve the results i am after if i just utilize the gear i already have in a more effective way. need to experiment with some different chain routing arrangements


Well it is certainly doable with a sequencer and a couple of other modules. But, with cycles you get immediacy of presets and mangling your source instantly and exploring a ton of ways to do that in a short amount of time. I think it fits good into a process where you compose a piece from a big number of elements and want to spend too much time crafting each sound while in the zone.
That does not work for me yet, I like to focus on nuances and generally make less complex music elements wise hence the love for modular :slight_smile:


I tried re-routing the Outward to the beginning of my effects chain and it does work nice. I also wanted to point out for anyone considering the Thyme that I don’t think it was a mistake to buy one. It is extremely versatile and can do things no other effects unit can do (to the best of my knowledge). It takes a light touch but if you learn how the internal signal path works and how the parameters affect them and are able to slowly modulate certain knobs, you can get the kind of results you may expect from scripts like cheat codes or modules like clouds. It really fills up the stereo field and can automatically add random interesting dynamic shifts to anything you feed it. I think I’m happy with the what I have for now.

Eventually when I do decide to get a small 4ms case, I will probably go with the nebulae according to the advice I’ve gotten here. But it is definitely possible to get some crazy expansive granular generative sounding wide ambience with the Outward and Thyme, even just with the thyme, but especially when you add the Ventris in the mix (which even seems to be overkill a lot of the time)

Thanks for the feedback on this. I think it really helps to get real world anecdotal evidence from people with similar goals & taste, so I appreciate that


You’ve encouraged me to want to return to my Thyme machine (time machine? who!). I found it quite disappointing when running material through it, but I think you are right regarding the light touch and need to understand the internal signal paths. Demos I’ve watched have demonstrated that it can do so much more than I was able to get out of it, so thanks for the encouragement!

1 Like

Amazing work!! The blend knob with live record was a big issue for me, so this would be great.
I’ll try to give it a test today.

1 Like

Tried this out now, it really transfroms live looping, for me at least. Once you get the hang of switching between blending the input in and recording, and then blending between the granular and phase modes, it’s amazing.

Plus, really like the alt Pitch variation mode, creates a really different and unique effect. Is it correct both pitch variations don’t work togther, it’s one or the other?

Again, great work and thanks for sharing.


Thank you! Yeah, you can’t do both at once. The pitch of a grain is decided when the grain is created, and the two modes differ in that one assigns a random pitch (the range is decided with the knob) to every grain, and the other assigns a probability of the grain being generated an octave or two up. No grain can play at two pitches simultaneously, without duplicating the signal.


Cool, thanks.

Also have you changed the recording mode sightly? It seemed to default to circular recording, and when I changed it, it still seemed to circular record just without the red lights. I’ll test again incase I’m going mad :smiley:

If you see the red lights, it shouldn’t be circular recording. I DID disable the red lights when circular recording is enabled, yes, I forgot to mention that! It’s because the speed controls are not locked when circular recording is enabled, per the manual. However, the red lights don’t make this obvious, so I disabled them.

The recording should otherwise remain unchanged. The default recording is a little bit funky as is. It doesn’t record/overdub the in signal, it records a combination of the in signal and the vocoded/granular mix. This means that it’s normally recording what you’re hearing, or something close to it.


Cool. Yeah, so makes sense now, normal recording works fine. Really enjoying it, makes the live looping so much more powerful. Thanks!

1 Like

Hmm, my nebulae no longer appears to be responding to the neb_update files. I’m running @TheTechnobear’s firmware and attempted to try out @alext’s new firmware but it didn’t seem to change the controls. I don’t think think it ever changed firmwares to @alext’s version because it was just white LEDs and never cycled green/aqua, as well as booted seemingly too quickly to change the firmware. Then I tried to use @TheTechnobear’s neb_update file as well as Qubit’s stock firmware and neither seemed to register. I also tried to use qubit’s revert to factory firmware trick and that didn’t work either so I’m kind of stuck on this one. I used 3 different brands of USB sticks - all empty and have been successfully used with Nebulae before.

Any ideas for how to get unstuck? Has the method of updating firmware changed (turn on with neb_update.zip on root of USB drive)? I am able change sample sets via usb as well as access the wifi/web portal.

if you started the update procedure with @alext firmware, then it will have unpacked it, and started to do the update… and started overwriting files, which now are being used even if you try to install a new firmware :frowning:

there are a few ways to try to get ‘unstuck’

a) if you know linux
then you can attempt to ssh into nebulae, and see what process is sticking… what is mostly likely (in my experience) is that the main nebulae.py is failing, and keeps getting restarted.
that said, given you have not seen the other green leds etc, that implies the calibration may be failing for some reason… but impossible to know.

(note: you may find you need a few attempts to ssh into it… if the main nebulae process is failing the constant failing/restarting - means the neb has little cpu left to process the ssh request - also I find using an ethernet cable can be more likely to work)

unfortunately, I cannot walk you thru this, as there are too many ‘moving parts’
but, if you can do this to debug, then this is useful, since hopefully the underlying ‘issue’ could be fixed,
but you need to know your way around linux, read scripts - and be able to track down what is going wrong.

b) revert firmware
the nebulae looks for a file on the usb stick called
(no extension)

this attempts to revert to the original firmware that shipped with the nebulae.
however, Ive heard its not always successful.

c) re-flash the sdcard
this will work… basically take the sdcard and use an sdcard read/writer to put the original image back on it.
see : this link when someone has uploaded the firmware.

obviously b/c will put you back to the beginning, so if you want my firmware again, you will have to re-apply it.

edit: yeah, not surprising the wifi stuff still works, as it runs in parallel to the nebulae process - so if its die-ing the wifi/browser will still work ok (assuming enough cpu left over to process requests)

1 Like

oh, one more ‘less drastic’ thing you can try … assuming you can get into the nebulae (via browser or ssh)

go to /home/alarm/QB_Nebulae_V2/Code/config/
and remove the file bootinstr.txt

this would ‘solve’ the issue, if the changes made in the firmware are somehow incompatible with an ‘alternative’ instrument, that was loaded at the time you did the firmware update.
this could possibly make the instrument keep crashing, and so keep restarting
so deleting above file, will stop it loading that instrument - and instead load the default granular which hopefully will work.

I don’t know if this is the case, as Ive not tried this firmware - so I don’t know what the underlying cause is.

1 Like

github lets you look at changes across forks, and so here is the set of differences between @TheTechnobear’s and mine. I couldn’t immediately see anything that would be to blame, but maybe someone else can.

Hmm, I don’t suspect @alext’s firmware is the culprit here as I’ve reflashed the SD card to Qubit’s firmware (which worked) but since then I still haven’t been able to update firmware (either @TheTechnobear’s or @alext’s). So the problem persists (now without a web browser option to do diagnostics). Also not able to add .instr files via usb.

I’ve reached out to Qu-Bit. Hopefully this isn’t a hardware problem - is there any part of the system software-wise that wouldn’t be affected by reflashing the disk image?

1 Like

No, if you reflash sdcard , it’s ‘as new’

I assume you don’t have usb stick inserted, usb sticks are not very reliable.

There’s a small chance the sdcard is failing, but they are pretty reliable - but it’s inexpensive to get another one to try.

When you say issues happen when swapping audio files - is this when you are trying to put them on nebulae ( via usb/WiFi) or when you reload into granular?

Are the audio files particularly big? Have you used them before with success on neb?

I’d try to replicate this in the ‘factory firmware’ as I’d assume this is what qubit would want.

sorry, i missed this. i had a hard time with it too. i almost returned it in the first few days. but now im so glad i didn’t. another good tip is to spend a lot of time building those meticulously crafted “sounds”/presets in banks so that you can trigger them via midi later on when performing/recording. I have gotten to the point where it sounds almost exactly like the Cheat Codes script, but with more flexible speed param modulation and stereo phase trickery. its insane. i’ve been using my 0-coast’s lfo for the cv input so i can modulate stereo phase and playhead 3&4 spacing

@robinrimbaud not a musical application really, but examples of the audio manipulation techniques:

And some concrète style with a textured sample running through the Thyme, a more granular demo:

1 Like

Excellent, would love to hear what you’ve done!

1 Like

I keep the USB sticks out unless swapping files. I have about 6 sticks each with a different collection of ~70 mb of .WAVs. Over the past year I’ve used these to swap audio file collections without issue. If it’s the USB sticks that are failing then at least 2 or 3 of them have failed simultaneously.

I considered the SD card might be failing yesterday. I flashed the firmware to another card I had lying around and just got solid white LEDs. This might have been a formatting issue - I noticed the factory card was formatted FAT16 while my card was FAT32. Not sure if that matters. Disk utility on Mac only has the option to format MS-DOS (FAT) so that’s what I did.

The issue with swapping audio files happened when I tried to put them on nebulae via wifi. The collection of files I was sending to nebulae were all ones I had used previously with success - maybe 50 wav files each about 1 mb.