Fluid Corpus Manipulation Toolbox (was Fluid Decomposition) v.1.0.0 Release Candidate 1 is here!

Code Sharing
1

Dear all

After many months of hard work, we have at last a release candidate ready for your pleasure! Many, many things have changed, so we have made a curated changelog to focus on what matters. There is three small bugs left as far as we can tell, and a few help files to improve still, but let us know whatever you find (good or bad)

https://www.flucoma.org/download/

I hope you enjoy!

PA, Owen and Gerard

1.0.0-RC1

date: 31 March 2020

New Features:

  • BREAKING CHANGE: (buf)melbands how has normalised amplitude output option, on by default, which is independent of window and fft size
  • (buf)NoveltySlice now has a minimum slice length parameter
  • BREAKING CHANGE: (buf)sines now have new parameter names, and a new option to select which algorithm is used to track the sines. It also has improved sound quality and more refined thresholding.
  • BREAKING CHANGE: (buf)AmpSlice is now 2 objects, (buf)AmpGate for the absolute, and (buf)AmpSlice for the relative
  • (buf)AmpSlice and (buf)AmpGate has 0Hz highpassfreq to bypass the high pass filter.

Bug Fixes:

  • (Pd on Windows) now works :wink:
  • BREAKING CHANGE: buf* processes now use the buffer’s sampling rate (or the SR attribute for Pd)
  • BREAKING CHANGE: (buf)ampslice now works much faster and consistently but thresholds and times will have to be tweaked
  • (Pd) BREAKING CHANGE: all non-real-time (buf) objects have lost the tilde (~) in their name to be consistent with the language conventions.
  • BREAKING CHANGE: (buf)OnsetSlice latency reporting (rt) and offset (buf) are now more accurate but will have changed for similar threshold
  • BREAKING CHANGE: (buf)NoveltySlice latency reporting (rt) and offset (buf) are now more accurate but will have changed for similar threshold
  • (buf)HPSS speed is improved
  • Many edge cases were found and sorted.
  • (Max) parameters updates now behave all the time in real-time
  • (SC) BREAKING CHANGE: bufnmf’s audio output destination buffer is renamed properly to resynth

Improvements:

  • (Pd+CLI) documentation is now more accurate and consistently formatted
  • cross reference between documents (see also)
  • credits, references and acknowledgements are streamlined and consistent
  • (Pd) better code for pinknoise abstraction

New Example:

  • (SC+Pd) most Max examples are now ported to the two other CCEs
  • (SC) Gerard’s GUI demos of algorithms are now available

Known Bugs/Issues:

  • (Mac) Notarization is not implemented yet, since most CCEs are working on workarounds…
  • (SC) Current implementation of non-realtime objects does not work with remote servers
  • (Win - SC) calling cancel on threaded non-realtime processes crashes scsynth
2 Likes
2

:clap: i think Max and SC links not working for me… ?

3

I’ve used them on Mac, I have not tried other platform. @weefuzzy any idea?

4

lemme try a different browser. just not clickable. Mac w/chome.

5

Just tried them again, on Chrome on Mac, without being logged, and they worked…

6

oh odd. works fine on safari. my version of chrome Version 80.0.3987.149 i can’t click any of the Max or SC links. shrug

7

Oooh, weird. I’ll look into that, but it’s only using pretty vanilla stuff.

8

I am using Edge, which is Chromium and I didn’t have an issue. None on FF either.

9

Just tried here on 80.0.3987.149 on Mojave, all seemed click-happy. Maybe there was a cache issue: I did mess around with the css a bit (but it’s still good ol’ fashioned <a> doing the links, so it would have looked awful but still worked. I think…)

1 Like
11

This release has now been patched with fixed documentation for AmpGate (thanks @rodrigo.constanzo for catching this)

1 Like
12

I don’t know if there’s a box.com setting or something, but when trying to download this with a slow internet connection, if it takes more than x amount of minutes, it times out, 100% of the time. I’ve tried like 5+ times to download it on my studio computer (where I have slower internet) and it just times out over and over.

13

Box is really not good indeed, sorry for that. Nothing we can do except, soon, move to github for RC2 I reckon. @weefuzzy and @groma are the authority in such matter in the team, so I will defer, but we all hate box equally and, imagine, are stuck to work with it on a minute-basis as we have been working remotely for a few years now and that is what the uni provides…

14

Scratch that, even on my normal connection I’ve been stuck at 30mb for 40min.

Screen Shot 2020-05-08 at 2.32.32 pm
Screen Shot 2020-05-08 at 2.32.32 pm862Ă—198 169 KB

15

4hours later… (literally the same download)

Screen Shot 2020-05-08 at 6.47.28 pm
Screen Shot 2020-05-08 at 6.47.28 pm878Ă—208 170 KB

16

man this is bad. I’ve never seen it that bad though… we have had a few hundred downloads of RC1, maybe the system is capping us… I’ll investigate

17

I’ve always had issues downloading the full packages. I remember one time you even uploaded a wetransfer in the interim as I wasn’t able to download it from the link.

(for posterity, that download was still going a few hours later and I finally killed it)

18

@rodrigo.constanzo
I’ve put the Max release onto Github. Haven’t updated the links on the web site yet (still need to do the others), but see how this works for you
https://github.com/flucoma/flucoma-max/releases/download/1.0.0-RC1/FluidCorpusManipulation_Max_Darwin_Windows_1_0_0_RC1.zip

1 Like
19

i am digging already… digging…digging

1 Like
20

:laughing: I was about to say: this implies that the code is now public too, but you’re way ahead of me.

1 Like