Anyone experienced anything like this (Windows and Linux Ubuntu)

[rickpress]

Create an edit and the CPU maxes out. Move the same edit over to a new tab with the same vsts whatever and it works fine and settles in? The audio engine does this on both Windows and Linux versions. Disabled all background processes. Truly annoying.

This is a great DAW and i would like to use it but it seems there are bugs flying around in Waveform which make using it a pain. A real pain.

Has anyone else experienced anything else similar?

[Maarid]

I have met something similar, but I will not lie, I did not check what I saw as in the video (in the new edit).
I didn't like attractive by itself, another meaningless sampler, I won't check, sorry.
I will support bugtracking the way I would like to be supported in this position.
Most often I observe the choking of the processor with some specific plug-in, most often this is due to a mixture of delay and reverb, complex effects plugins in the style of izotop iris, GuitarRig, but it happens very rarely when seemingly adequate plug-ins go crazy, as for example it was recently with FabFilter ProL - at some point he began to eat up the entire processor and only a complete reboot of it from the circuit helped.
Oddly enough, due to my interests, I have been connecting many plugins through the jbridge gasket for a long time. Back in version 11, Waveform supported plugins at the processor level very poorly. Any sane synthesizer caused him heartburn, but through jbridge the problem evaporated.
Right now, in the process of writing an answer to the translator, I launched Waveform and pressed the play of the combination of Ample Guitar GuitarRig in its purest form, Waveform shows me two plugins as 44% of the processor, but the task manager is 9.5%. I honestly don't know what to do with this information. And I honestly share my experience that it is more convenient to run all sane plugins outside the program. After all, disabling the sandbox usually leads to the closure of the program, which is even more inconvenient.
On Linux, I met several jbridge-style wrappers on githab, vstwrap vst3wrap, etc., I didn't check personally, sorry again.

yes, I almost forgot, I tell everyone not to forget myself - please do not use default gain on track, especially with automation, it is he who adds static low-level noise to each track.

[Peter Widdicombe]

Waveform CPU use is sort of misleading - the percentages for plugins show how much of the USED CPU is being allocated to each plugin. For instance, the SYSTEM (i.e. Windows) might be allocating 20% of the processing capability to Waveform, but you have 2 plugins each reporting 44%, and another at 10%. This shows you the ones "using the most juice" whether the system itself is loaded or not.

Now, if you find Waveform is taking 70% of the CPU, those same 2 plugins at 44% are each taking 30% or so of the actual processing capability...

[Maarid]

Now, if you find Waveform is taking 70% of the CPU, those same 2 plugins at 44% are each taking 30% or so of the actual processing capability...

You didn't understand, sorry for my clumsy translation.
The percentage of plugins in the opening window inside Waveform has nothing to do with 44% of the CPU indicated as currently involved at the top of the DAW window, the Windows Task Manager itself shows at the same time that only 9-10% of the CPU is involved in the overall Waveform. Do you understand?

[dysjoint]

yes, I almost forgot, I tell everyone not to forget myself - please do not use default gain on track, especially with automation, it is he who adds static low-level noise to each track.

Off topic, but yes,
I read about this and briefly tested it recently (as far as my abilities took me) and Waveforms automation does seem to add harmonics pretty significantly, at low levels mostly. Here's a Spectrogram of a 1k sine tone with a volume automation in the middle third, and edge fades at the beginning and end of the clip. In actual use in the daw I never noticed or heard it, but it is there.

[rickpress]

With respect to what you are both saying when the CPU meter in Waveform reaches anything around 80 percent the edit will freeze and the audio engine stops, as shown in the video. So I would imagine the Waveform meter is fairly accurate regards to how Waveform behaves.

[dysjoint]

With respect to what you are both saying when the CPU meter in Waveform reaches anything around 80 percent the edit will freeze and the audio engine stops, as shown in the video. So I would imagine the Waveform meter is fairly accurate regards to how Waveform behaves.

Absolutely. I have had edits behave like your first example, which is really annoying as I bought an i9 to not have cpu problems, but Waveform does max out and quit the audio engine well before my cpu gets anywhere near maxed. I don't know any solutions, watching this thread to see if anyone else does. (I'm on Windows 11, which is supposed to be good with multi core processors, as is Waveform 12)

[Maarid]

So I would imagine the Waveform meter is fairly accurate regards to how Waveform behaves.

With all due respect, I hardly understand why evidence of obvious things is required, but well, doubt is everyone's right.

1.png

2.png

Does this look like an adequate state reading in your opinion? Perhaps it is I who have problems with iron? And I don't know. With all due respect.

[rickpress]

Regghardless the fact is when the waveform meter hits around seventies to eighties it will max out regardless of how that ties in with what windows task manager displays as cpu readings. The unfortunate fact is that the audio engine seems to be a bit quirky and maxes out at times for no reason using certain plugin combinations and at other times has no problem with the same plugin combinations which makes working in it frustrating.

[Maarid]

Regghardless the fact is when the waveform meter hits around seventies to eighties it will max out regardless of how that ties in with what windows task manager displays as cpu readings

That's why I wrote to you above that I don't know what to do with this information. The humor of the situation is that regardless of the complexity of the project, regardless of the processed frequency and bit depth, I will consistently get 100% CPU in the program, but rarely go beyond 50% CPU in the system. But I can load the processor with any application or test up to 100% directly and get cursor fading. It's just a fact without interpretation.
The sound engine in the program is the JUCE sound engine. No comments.
Still, as in my first message, I would just recommend that you try to reproduce your mistake from the video by taking out the Attractive process outside the DAW using jbridge. I think the problem is in the processing of the reverberation of the plugin itself. If the error repeats in jbridge, then Attractive is broken, if not, then I'm right and Daw initially strangles plugins by default.
That's my thought, I tried to write quite simply

[nowgad]

No comments.

I also wrote about this several times. When a DAW starts to choke, it's very rare that the CPU usage rises to 65%. Basically, the CPU load is no more than 35% (why exactly a third is unclear), while the figures in the DAW are about 80%. I asked the developers about this when they still knew how to talk, but did not receive a clear answer.
The funny thing is that the JUCE engine and plugins are built using JUCE, but all together it doesn’t work properly))

[Maarid]

why

no comments, because we get proprietary and closed code based on a good framework that is being developed right now. Neither you nor I are able to control the version of the framework on which this program is being built, respectively, its performance may depend on both the author himself and his incorrect choice of tool. How to comment on it?

[lmv]

1. The big problem with your post is that you are claiming that the exact same edit will behave differently in separate tabs. But then in the video you say, "the only difference is that..." Here we go.

So it's not the exact same edit. The one that has more instruments crashes. You are comparing two different things as equal. Your comparison is worthless. Sorry.

2. Waveform won't let you drag anything across tabs so how did you load "the same edit" in another tab? What was this procedure like?

3. The thing about having to reload drum presets is interesting. I only have that problem with the shortcircuit sampler. Add it, load preset, save, close, reopen. Shortcircuit complains. Browse to the allegedly missing sample, save the edit again, close, reopen, problem solved.
But it's only with shortcircuit and I had that problem in Mulab too a long time ago. What samplers are you using?

[dysjoint]

2. Waveform won't let you drag anything across tabs so how did you load "the same edit" in another tab? What was this procedure like?

You can copy/paste stuff across edits. Clips, plugins, tracks...

[rickpress]

How am. I using more plugins? You can count them they are the same edit. I opened a new tab and loaded the same vsts with the exception of the micro drum sampler for the drum sampler and copy and pasted the same. Clips over.

I did exactly as dysjoint has said, copy and paste and just load the same vsts. There are only two instrument vsts loaded, two instances of Attracktive. The edits are practically the same. The reason I said the only difference is using micro drum sampler over drum sampler. They are both playing the same kit. There is no case of either edit having more insturments. You can count. It's not hard to see. Both edits have the same clips, so they are almost identical. Changing micro drum sampler for sampler makes no difference so I don't see how the comparison is worthless. What the video shows is that waveform is a bit sketchy sometimes with cpu usage. Not an issue I can use fl studio instead.