Hi all, we've released v1.0.6 of Collective which we hope will fix a few issues and improve performance a bit. Please update and inform us of any problems.
https://marketplace.tracktion.com/shop/ ... collective
V1.0.6
- [Linux] Fixed the directories in which Collective searches for Instruments, Sounds and Samples:
Public content (for all users of a machine): /usr/local/share/Tracktion/Collective/
Private content (per user): ~/Tracktion/Collective/
Please move your content folders to one of these locations if the don't already reside there.
- Fixed rare crash in the equalizer control.
- Possible recovery after Collective going silent by selecting a new instrument.
V1.0.5
- Preset browser remembers last selection upon closing and reopening the plug-in view.
- Preset list can be scanned with Up and Down Arrow keys. Pressing Return loads the selected preset.
- Further improvements and fixes.
- [Mac AU] Audio Unit validation finishes successfully.
V1.0.4
- [Mac AAX] Number of automation parameters in AAX limited to the first Sound layer.
- [Mac] Possible crash on some Macs fixed.
V1.0.3
- Fixed possible crash while loading.
Collective v1.0.6 Released
-
- KVRAF
- 1790 posts since 30 Dec, 2012
-
- KVRAF
- 37490 posts since 14 Sep, 2002 from In teh net
Is the published params issue fixed?
-
- KVRAF
- 1674 posts since 13 Jan, 2014
Just trying 1.08 beta on an existing track as a performance test with collective on evry track.
I then realized (if I did not get something wrong) that I have to reduce the number of voices with evry new sound I try again on the instruments page. I would really love if there would be an additional general option to set the number of voices as a default (in the settings for example) no matter which instrument I choose cause with the default 40 voices and the general cpu-hungryness of collective thats really annoying to set again for evry instrument you try while hoping it will NOT crash evrything if you forget one time.
I then realized (if I did not get something wrong) that I have to reduce the number of voices with evry new sound I try again on the instruments page. I would really love if there would be an additional general option to set the number of voices as a default (in the settings for example) no matter which instrument I choose cause with the default 40 voices and the general cpu-hungryness of collective thats really annoying to set again for evry instrument you try while hoping it will NOT crash evrything if you forget one time.
-
- KVRAF
- Topic Starter
- 1790 posts since 30 Dec, 2012
1.0.8 is the same functionality-wise but has a new update checker which is currently in testing.U4Ya wrote:What about 1.0.8 beta, what updates are in it?
We've now split the content download from the plugin binary to reduce the size of downloads.
The same download manager will be present in the next version of Waveform so you'll be notified whenever there are updates for any of your Marketplace products.
-
- KVRAF
- Topic Starter
- 1790 posts since 30 Dec, 2012
Is this that the names of some parameters were incorrect? I think that was only for some presets (as parameter names can change for each preset). I think that was fixed in the last content update but I'll double check with Wolfram.aMUSEd wrote:Is the published params issue fixed?
-
- KVRAF
- 1674 posts since 13 Jan, 2014
NO, sorry, this is STILL so far away from a sample-workstation-workhorse for all day use and STILL that buggy (I am actually trying the 1.08 beta), no, maybe 2.0.....
As I said: I have an 11 tracks song filled up to the brim with ram-and cpu consuming mostly kontakt libs (sometimes 2 or 3 libs on one track parallel, lots of modulating soundscpae and sequencing/arpeggiating pluck stuff, my ram with 8 GB is nearly filled), then something on the master track and it WORKED. I can play the song without crackles or drop outs. Of course you have to be patient sometime and I KNOW I am on the border BUT....
Now the same song, 7 from 11 track filled with one instance from collective, each reduced to max. 8 voices (just as an estimation: in the 11 track song most kontakt instruments are set to 32 voices and use them, I would suggest four-time as many voices is realistic...), NOTHING on the master track and it just has real ongoing sound dropouts (have closed waveform, opening again, other projekt no problems), cpu indikator differs between waveform (there evrything LOOKS fine < 40 %) and task manager which has spikes going up to 100 % (I think there the dropouts will come from).
This is just (STILL) no fun even to tryout.
I can accept the high use (which is no fun eager, BUT...) on something like biotek which I may use one instance on one track but on a "sampler" ?????
As I said: I have an 11 tracks song filled up to the brim with ram-and cpu consuming mostly kontakt libs (sometimes 2 or 3 libs on one track parallel, lots of modulating soundscpae and sequencing/arpeggiating pluck stuff, my ram with 8 GB is nearly filled), then something on the master track and it WORKED. I can play the song without crackles or drop outs. Of course you have to be patient sometime and I KNOW I am on the border BUT....
Now the same song, 7 from 11 track filled with one instance from collective, each reduced to max. 8 voices (just as an estimation: in the 11 track song most kontakt instruments are set to 32 voices and use them, I would suggest four-time as many voices is realistic...), NOTHING on the master track and it just has real ongoing sound dropouts (have closed waveform, opening again, other projekt no problems), cpu indikator differs between waveform (there evrything LOOKS fine < 40 %) and task manager which has spikes going up to 100 % (I think there the dropouts will come from).
This is just (STILL) no fun even to tryout.
I can accept the high use (which is no fun eager, BUT...) on something like biotek which I may use one instance on one track but on a "sampler" ?????
-
- KVRer
- 28 posts since 12 Dec, 2016
I have just installed 1.0.8... with the hope that after all reports and even the call for experience, "MIDI silence" is fixed. But no, still the same.
Windows has many alternatives for "all rounder" sampler, but on Linux I have Collective only. Unfortunately Collective on Linux is almost unusable (standard Ubuntu). The problem can be reproduced within 1 minute: new project, one track with Collective, loaded Drum Kit, pattern dropped to the track, a bit manipulated/looped and BUM... sound appears and goes randomly.
I use 2 workarounds:
1) restarting waveform and rendering / freezing required track before the effect appear
2) removing all effects from Collective preset (as for other, some "enable/disable/change" effect also can work for a while, but that is too unstable).
Both workarounds in combination are close to 100% reliable for me, but boring: use "effectless" version of the preset as long as I want manipulate MIDI, then render/freeze the result with Waveform restart. If I want to modify MIDI/the sound later, I have to repeat the procedure.
As other, I also observe huge CPU use by Collective. Like 30-40% of one i7 core. But with the first problem it makes no sense to run many of them, so I will start begging for CPU fix only after "silence bug" is fixed
PS. Waveform by itself also consumes quite some CPU... And while Collective has mentioned problems with GUI closes, remembering "baby time" of some plug-ins from other company, can it be that graphic driver (in particular OpenGL) can contribute there? I have not tried to install NVidia Linux drivers yet, so working with stock opensource intel (I have dual graphics notebook).
Windows has many alternatives for "all rounder" sampler, but on Linux I have Collective only. Unfortunately Collective on Linux is almost unusable (standard Ubuntu). The problem can be reproduced within 1 minute: new project, one track with Collective, loaded Drum Kit, pattern dropped to the track, a bit manipulated/looped and BUM... sound appears and goes randomly.
I use 2 workarounds:
1) restarting waveform and rendering / freezing required track before the effect appear
2) removing all effects from Collective preset (as for other, some "enable/disable/change" effect also can work for a while, but that is too unstable).
Both workarounds in combination are close to 100% reliable for me, but boring: use "effectless" version of the preset as long as I want manipulate MIDI, then render/freeze the result with Waveform restart. If I want to modify MIDI/the sound later, I have to repeat the procedure.
As other, I also observe huge CPU use by Collective. Like 30-40% of one i7 core. But with the first problem it makes no sense to run many of them, so I will start begging for CPU fix only after "silence bug" is fixed
PS. Waveform by itself also consumes quite some CPU... And while Collective has mentioned problems with GUI closes, remembering "baby time" of some plug-ins from other company, can it be that graphic driver (in particular OpenGL) can contribute there? I have not tried to install NVidia Linux drivers yet, so working with stock opensource intel (I have dual graphics notebook).