Sounds like some accidental MIDI learn, can you check the MIDI table page for any assignments?
Zebralette 3 Bug Reports
-
tasmaniandevil tasmaniandevil https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=62450
- KVRAF
- Topic Starter
- 1737 posts since 22 Mar, 2005 from a planet called u-he
That QA guy from planet u-he.
-
- KVRAF
- 2016 posts since 23 Jan, 2022
i just checked again, it is doing it only loaded inside Chord Prism, maybe is a problem of the Chord generator hosttasmaniandevil wrote: ↑Mon Feb 26, 2024 6:10 amSounds like some accidental MIDI learn, can you check the MIDI table page for any assignments?
aliasing plugin owner
-
- KVRist
- 40 posts since 4 Sep, 2011
Settings of big OSC editor and little one are not "synced" I.e., grid harmonics in big editor is not reflected in the little guy. Bug/by design? Is a bit jarring to me when going back and forth.
You do not have the required permissions to view the files attached to this post.
-
- u-he
- 28065 posts since 8 Aug, 2002 from Berlin
-
- KVRist
- 330 posts since 1 Jul, 2023
Had Zebralette3 crash Cubase, although I had finished work and was shutting Cubase down. Windows generated a crash report which I can upload if needed.
Finding that some patches sometimes trigger random noise when playing via my midi keyboard. Very random, not on all patches but notably on the additive piano patch and often when playing multiple notes in succession.
Finding that some patches sometimes trigger random noise when playing via my midi keyboard. Very random, not on all patches but notably on the additive piano patch and often when playing multiple notes in succession.
-
- KVRist
- 40 posts since 4 Sep, 2011
Harmonic grid and x/y setting aren't always working. In attached image X is set to 4 but grid clearly shows more than 4 harmonics. In this state, changing X to 32 or any other value had no effect on the grid.
You do not have the required permissions to view the files attached to this post.
-
- KVRist
- 40 posts since 4 Sep, 2011
I am going in and out the big OSC editor a lot and noticed that at some point, the audible waveform is fixed i.e. does not change no matter what I draw or change to the curves. Closing ZL3 and reopening seems to fix it (I don't have to exit the DAW). (Working with Logic 10.8.1, on a Mac Studio M2Max)
-
- KVRAF
- 25463 posts since 3 Feb, 2005 from in the wilds
That makes sense. Just maybe the X axis value should be greyed out when Harmonic Grid is active.
-
tasmaniandevil tasmaniandevil https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=62450
- KVRAF
- Topic Starter
- 1737 posts since 22 Mar, 2005 from a planet called u-he
Could it just be a mismatch of curve morph position and the currently selected waveform snapshot?
Happened to me more often than I'd like to admit.
If this can't be the reason, please save the state of the preset when it happens the next time and attach it here, so we can have a look.
That QA guy from planet u-he.
-
- KVRer
- 13 posts since 18 Dec, 2018
In the OSC wave shape graph editor UI:
When playing notes on the computer keyboard in live, one of the letters (F, G or H I think) calls up a secondary curve of some sort, that overlays the oscilator shape editor. I don't really know what it does, or why it's there. Could be that it is the morph line?
Anyway, that keyboard assignment is a bit problematic.
(otherwise, I can't wait to get my hands on zebra 3. this is amazing)
When playing notes on the computer keyboard in live, one of the letters (F, G or H I think) calls up a secondary curve of some sort, that overlays the oscilator shape editor. I don't really know what it does, or why it's there. Could be that it is the morph line?
Anyway, that keyboard assignment is a bit problematic.
(otherwise, I can't wait to get my hands on zebra 3. this is amazing)
-
- u-he
- 28065 posts since 8 Aug, 2002 from Berlin
So, when the editor has key focus (after any mouse click into it), it currently iterates through Guide Curves when pressing 'g'. With shift ('G') it swaps between editing Guide and actual Curve.
I guess this is less of a bug and more so of a UI/UX thing.
Anyhow, I've revisited our long time stalled key control project to see if I can remove some quirks and improve it somewhat. It's looking very good for that. Our software is getting more keys these days, and somehow we can do a lot of things now that we couldn't do a few years ago. With some tweaks I could even get our GUI-Editor to fully work on Windows, in most hosts. Yay. We'll beta test this soon, too.
I guess this is less of a bug and more so of a UI/UX thing.
Anyhow, I've revisited our long time stalled key control project to see if I can remove some quirks and improve it somewhat. It's looking very good for that. Our software is getting more keys these days, and somehow we can do a lot of things now that we couldn't do a few years ago. With some tweaks I could even get our GUI-Editor to fully work on Windows, in most hosts. Yay. We'll beta test this soon, too.
-
- KVRer
- 4 posts since 13 Dec, 2018
I did some more detailed inspection on the high frequency "vinyl" crackling noise behavior in wavetable rendering mode that I and few others have noticed. Tests were mostly done playing octaves 4 and 5 with the regular -12 transpose.
I used FL studio 21, Windows 10 on both my laptop and desktop, regular sample rate being 48khz. Hopefully I don't sound like a broken record or blabble obvious stuff too much, I'm trying to list pretty much everything I've noticed.
The noise is at it's most noticable when doing sounds with a very limited harmonic spectrum, the quickest way of reproducing it noticably is using the high pass effect on the init saw and cutting aggressively, value at around 85-90. When doing similar high passing with another plugin, it's not noticable. I tested with zebrify vcf-filters and fabfilter pro q3.
As showcased before, it also is easy to make clearly noticable when using curve spectrum mode and creating a strong peak at the highest octaves.
However, even when using regular waveforms (saw/square) with no oscillator effects in curve geometry, a crackling/pulsating behavior can be seen with fabfilter pro q 3 spectrum analyzer, although the volume of the artefact is too low for it to be heard. The behavior does not look similar to regular aliasing produced by for example, xfer serum draft-mode, and it also does not sound like it, whenever audible.
The artefact becomes more noticable with higher resolution settings, while higher sample rates reduce it. The behavior is similar on both my laptop and desktop, and it's clearly unrelated to things like cpu-load.
I also made a few tests with sine waves.
When creating a single sine shape in the curve geometry and playing some very high pitched note, for example 11-16khz, the effect is basically gone. Spectrum analyzer still reveals it to be there somewhat. However, when creating for example, 16 sines withing a single frame and playing it in similar registers, there are very noticable artefacts. Similarly, in curve spectrum, pitching up a sine that is the fundamental on the spectrum, the artefact is only barely seen in analyzer, not heard. However, when creating a single sine higher up the spectrum, say harmonic 11 or so, and pitching that to a similar frequency, very noticable artefacts happen.
I have not noticed a significant amount of regular aliasing in the plugin, nothing out of the realm of expectations. My layman intuition says that this is something else than just things going over the nyquist-limit, although sample rate does have an effect on it.
Hopefully something I wrote here is helpful in finding things out. Finally, I wish to say again that Zebralette 3 is such a lovely creation already, it's absoludely mindbending what this "single oscillator" is capable of achieving!
I used FL studio 21, Windows 10 on both my laptop and desktop, regular sample rate being 48khz. Hopefully I don't sound like a broken record or blabble obvious stuff too much, I'm trying to list pretty much everything I've noticed.
The noise is at it's most noticable when doing sounds with a very limited harmonic spectrum, the quickest way of reproducing it noticably is using the high pass effect on the init saw and cutting aggressively, value at around 85-90. When doing similar high passing with another plugin, it's not noticable. I tested with zebrify vcf-filters and fabfilter pro q3.
As showcased before, it also is easy to make clearly noticable when using curve spectrum mode and creating a strong peak at the highest octaves.
However, even when using regular waveforms (saw/square) with no oscillator effects in curve geometry, a crackling/pulsating behavior can be seen with fabfilter pro q 3 spectrum analyzer, although the volume of the artefact is too low for it to be heard. The behavior does not look similar to regular aliasing produced by for example, xfer serum draft-mode, and it also does not sound like it, whenever audible.
The artefact becomes more noticable with higher resolution settings, while higher sample rates reduce it. The behavior is similar on both my laptop and desktop, and it's clearly unrelated to things like cpu-load.
I also made a few tests with sine waves.
When creating a single sine shape in the curve geometry and playing some very high pitched note, for example 11-16khz, the effect is basically gone. Spectrum analyzer still reveals it to be there somewhat. However, when creating for example, 16 sines withing a single frame and playing it in similar registers, there are very noticable artefacts. Similarly, in curve spectrum, pitching up a sine that is the fundamental on the spectrum, the artefact is only barely seen in analyzer, not heard. However, when creating a single sine higher up the spectrum, say harmonic 11 or so, and pitching that to a similar frequency, very noticable artefacts happen.
I have not noticed a significant amount of regular aliasing in the plugin, nothing out of the realm of expectations. My layman intuition says that this is something else than just things going over the nyquist-limit, although sample rate does have an effect on it.
Hopefully something I wrote here is helpful in finding things out. Finally, I wish to say again that Zebralette 3 is such a lovely creation already, it's absoludely mindbending what this "single oscillator" is capable of achieving!
-
- KVRAF
- 25463 posts since 3 Feb, 2005 from in the wilds
What I've noticed in Curve Spectrum mode is that there can be very loud individual partials. So the overall level of the sound is not high, but there are individual partials/transients that are. Maybe the overall level is -18 but some bits are say +20NeonBuddha wrote: ↑Mon Mar 11, 2024 12:07 pm As showcased before, it also is easy to make clearly noticable when using curve spectrum mode and creating a strong peak at the highest octaves.
If you make a strong peak at the highest octaves, lower the heigh of that peak and the crackles go away.
-
- u-he
- 28065 posts since 8 Aug, 2002 from Berlin
So my bet is, the higher the partials, the louder interpolation noise. I think for extreme spectra with very loud partials in the top three or four octaves, we either need to use a more CPU intensive interpolation method, or we need to use a larger internal wavetable size. Both are manageable options.
When it happens without and OscFX, one can export the Wavetable and check if it's the same thing in Hive. Hive uses the same playback algorithm, pretty much.
When it happens without and OscFX, one can export the Wavetable and check if it's the same thing in Hive. Hive uses the same playback algorithm, pretty much.
