Clouds Parasite: an alternate firmware for Clouds feat. the Oliverb and Resonestor


#21

@BennelongBicyclist I answered to you on Muff Wiggler about the naming! (ha, we should really settle on one forum for our discussions :D)
About Frames: I’m also planning a release of the Frames firmware soon, we should really unite forces! It would really be too bad to have two separate versions… What are your plans? I have my sequencer modes, plus saving of multiple presets.

@rupertlally Yes, I agree, it would be really useful to have the original firmwares somewhere.


#22

@mqtthiqs wrote on MW:
bq. Sorry, I wasn’t aware of the pattern here (which is Tides and Bees-in-Trees, right?)… There is definitely deference in my naming, and deserved one, but I intended more a reference to the “biological” nature of open source, each project occupying its own ecological niche. I have to say I have a conflicted relationship with Greenaway… I haven’t seen Drowning in Numbers in ten years, and remember feeling pretty strongly against it, so I’d feel a bit uneasy to name this work after his… Can’t we have our own pun? Plus, seen your job, I thought you would appreciate smile Ok ok, you won, I’ll give Greenaway another shot soon smile

Greenaway is not to everyone’s taste. Anyway, if you want to adhere strictly to the naming convention, the games listed in the film are:

  • Bees in the Trees
  • Dawn Card Castles
  • Deadman’s Catch
  • Flights of Fancy (or Reverse Strip Jump)
  • The Great Death Game
  • Hangman’s Cricket
  • The Hare and Hounds
  • Sheep and Tides

However, I think other references to the film are also fine, such as to the soundtrack, which adds:

  • Trysting Fields
  • Wheelbarrow Walk
  • Fish Beach
  • Wedding Tango
  • Crematorium Conspiracy
  • Knowing the Ropes
  • Endgame

And then there are the 100 greatest magnitude stars named by Skipping Girl in the film (she only gets up to 58, however).

The problem with “parasite” is that it has negative connotations for most people (possible but as yet unproven therapeutic uses notwithstanding). It makes me want to recommend an antihelminthic to treat your firmware. The biological analogy isn’t really correct, either. I was going to suggest epiphyte, or even epibiont. Phages springs to mind, or plasmids. Or something from genomics Alleles (not quite right)? Polymorphisms (but not really single nucleotide) is probably closest, and is a term used in object-oriented programming, of course.

Then for Clouds, you have a rich meteorological taxonomy to chose from.

But a game from Drowning by Numbers would be best.


#23

mqtthiqs> About Frames: I’m also planning a release of the Frames firmware soon, we should really unite forces! It would really be too bad to have two separate versions… What are your plans? I have my sequencer modes, plus saving of multiple presets.

Well, I was thinking of implementing multiple presets in Frames, which I think I originally suggested (see 5th post in this thread ) - however, I’m busy with other things, including implementation of presets in Braids, so I am happy if others work on that. Other ideas for Frames include:

  • porting of the Lorenz attractor code from Streams - it would be possible to tweak all three parameters via the top row of knobs, and output values for each of the x, y and z dimensions on individual outputs.
  • addition of other chaotic algorithms, including perhaps Chua’s circuit and/or the circuit used on the NonLinearCircuits Sloth module by AndrewF.

Other ideas I have captured here. Please observe the moral rights of those who came up with the ideas.


#24

As mentioned in the wiki, for Frames, implementing a 4 channel version of this (minus the low-pass gates, but perhaps with envelopes) seems like low-hanging fruit. It may be possible to borrow the vactrol code from Streams (which also uses an STM32F1 MPU, like Frames) to implement “pingable” envelopes in Frames, either as part of a MakeNoise RxMx implementation, or separately (or both).

Frames could provide 4 pingable envelopes, sharing a single trigger or strike input, but each with an adjustable delay. The big knob could vary the shape of the pinged vectorial envelope, as on Streams.


#25

This is mirrored from MW (two forums is one too many):

mqtthiqs> I already tested going back to the factory firmware, it works without a problem. I’ll double check tonight to be sure, and also to see whether settings are kept, but it should be fine since I’m not modifying save/load routines neither the data structures that are saved.

BennelongBicyclist> I don’t (yet) have a Clouds, but presumably it saves the mode and other state it was last in so that it comes up in the same mode and state on rebooting? And presumably your extra reverb mode is implemented by adding an extra number to an ENUM which identifies the state or mode of the module. If that number is stored, what happens if a user selects the add-on reverb mode, powers down, and then re-installs the official Clouds firmware? The Clouds firmware will read the value from flash, but it will be outside the range of values it expects. What happens? Does it just revert to defaults, or does it crash? Those are the sorts of scenarios that needed to be thought about and tested for. As I said, Olivier added code to the latest Braids v1.7 firmware to handle such scenarios, but Clouds firmware may not include such code. It might be a good idea for Olivier to routinely add such settings sanity-check code to the official firmware for all his modules, in due course, if hacking becomes commonplace.


#26

I’ll think about the names BennelongBicyclist, thanks for the suggestions (I’m still not really convinced, sorry). I can probably tell you already that it’s not going to be “Crematorium Conspiracy” if I’m going for less disturbing :slight_smile:
Let’s talk about Frames ideas elsewhere soon! I’ll release properly the small hack I already posted as soon as I get them right, and then I’ll have some free time for more! (I also really need to make some music though…)

> Other ideas I have captured here.

This page has been most useful already, thanks for maintaining it! That’s where I got inspired to do the shift register sequencer, and the multiple presets things. Naturally, I’ll acknowledge it!


#27

i wanted to report that a couple times a sine-like tone manifested playing with oliverb, and the only way i could get rid of it was to power cycle. it could be a bug, but i’m not sure how i could reproduce it yet, maybe if it happens again i will shoot a quick iphone video… but just wanted to mention.


#28

I also had the same feedback/ sine wave tone happening - not just in Oliverb mode - last night whilst I was processing my Bass Guitar through Clouds (using a Doepfer A-119). It happened a couple of times, the first time when I tried to switch from granular mode through to looping delay, later on just randomly. It’s always the same frequency and from what I can remember, lower than 1khz. It’s only happened processing the bass, I’ve not had it happen at all processing either Eurorack signals or guitar. Like Skyon, I’m not sure if I can manage to reproduce it but if so I will try and record it…


#29

Hey guys, thanks for the report! It sounds like a serious bug. I’ll investigate this in the next few days. In the meantime, any additional info is more welcome! (a recording, steps to reproduce…)


#30

I had it again today, but didn’t have the modular connected to the computer so, couldn’t record it unfortunately. It was using Oliverb mode and with several inputs (from within the modular) going into the module in mono. I had a bunch of modulations going in from a Double andore’ envelope gen, a wogglebug and some of these then sent through a clock divider and so unplugged each at a time and it seemed to been to do with the modulation of “Position”, as once I unplugged the source modulating that it went, though that I realise that’s not very scientific :). The first time it happened (yesterday) I turned the modular off and then back on again and the tone was still there, which would seem to rule out any sort of feedback. Once again it’s a static frequency, under 1khz a bit like an additional harmonic or locked frequency that starts to bleed through - however altering the frequency of the sound sources or the pitch or the size parameters does not change the frequency of the tone… Sadly I’m away until Monday now, but will try and re-create the patch once I’m back…


#31

Rupert, I managed to reproduce the bug, by switching away from OliVerb mode while it is fed with sound. You then get a tone in the built-in reverb, if it is on. It is nasty.
For the mean time, you can actually eliminate the tone by just switching back to Verb and again back to your mode of choice (you don’t have to turn power off and on again). Did you get that without switching mode at all?
I’m busy today and tomorrow, but I’ll try to figure it out by the end of the week, and hopefully release a fixed version this week-end. Would you be interested in beta-testing? Anyone else?
Thanks again for the report.


#32

Ah, well done Matthias! I think that’s exactly what happened the first time with my Bass guitar, I had a fairly long reverb going and tried to switch to another mode was the reverb was still tailing out… I’m not sure what caused it yesterday, though it wasn’t through switching modes. But I had a lot of mods going in from the Double Andore whose wavetable env generators shape was being modulated so that could have a more obvious cause: changing wavetable shape too fast whilst plugged into a combination of SIZE, POSITION, and DENSITY, for example…
I will gladly Beta test, though as I said I’m away from the modular now until Monday…


#33

You’re welcome - thank you for all your great work!


#34

Hi skyon and rupertlally (and any other courageous user)
I hopefully fixed the constant tone bug, among others. Would you mind testing this new firmware (v0.95beta)? If you can confirm that it’s gone, I’ll release this bug-fix version in a few days.

Here is what changed:

  • fixed a bug producing a constant tone in the built-in reverb when switching away from OliVerb
  • consolidated clock multiplication/division code to avoid reported module freeze (although I could not reproduce it).
  • improved a bit the quality of the Oliverb: less distortion, less damping.

Thanks in advance!


#35

I’ve downloaded it and have been throwing all sorts of modulations at it and changing between modes whilst doing so for the past hour and not had a repeat of the “locking up/feedback problem” so far… Will let you know if anything crops in further testing…

In the meantime came up with a cool track using it and Sheep in mid range PLL mode:
https://soundcloud.com/rupertlally/dark-echoes


#36

Thanks Rupert, I appreciate your help!


#37

@mqtthiqs been absent from modular landia because of work, i’m glad that you’ve been able to reproduce it, i will install it tonight, and i suppose no news will be good news. thanks for your hard work!


#38

skyon, just a heads-up: have a look at this thread on MW, and you might want to check withmqtthiqs before re-installing - something weird seems to be going on with the Clouds audio bootloader, not clear what. https://www.muffwiggler.com/forum/viewtopic.php?p=1875138#1875138


#39

Yep @skyon, you might want to wait a bit before updating, and watch the MW thread. We’re waiting for Olivier’s thoughts on a user’s module which got bricked during the update procedure. The problem also happened to me, and I had to go buy a FTDI cable to un-brick my Clouds. It seems to be very rare, but it happened twice so if you want to be extra-careful, wait a bit.


#40

mqtthiqs, when your module was bricked, have you tried getting a dump of the flash content?

python stm32loader.py -p PORT -b BAUDRATE -r -l 1048576 image.bin

If it ever happens again, please do it, this is very valuable information for me!