Incorrect reflection of the scope of actions (Limiter) in View - History.

You see: Limiter - 2 Mb, and Cut to the clipboard - 59.2 Mb. It is wrong. Indeed, everything is In contrary, Limiter - 59.2 Mb, and Cut to the clipboard - 2 Mb. What has happened?
Capture.PNG

Immediately after applying Limiter everything is correctly:
Capture.PNG

But after some other effects the picture from the first post appears.
Capture.PNG

It’s telling you how much space will be reclaimed (how much additional disk space will become available) if you discard some of the history.
For example, if you discard the first three levels, you will gain 59.2 + 2 + 2.5 = 63.7 MB of free space on your hard drive. The amount of space used by the project will go down from 132.9 MB to 132.9 - 63.7 = 69.2MB

Thank you, but I’m not asking that a bit. I’m interested in figures against the Limiter effect. Why in the process of working in the same project the value in the second column is 59.2 MB, then 2 MB?

What is your interest? Is that you need to know for a particular purpose (in which case, what is the purpose), or are you just wanting to understand what it is doing behind the curtains?

Since for the other effects the value in the second column (Reclaimable space) doesn’ t change, no matter how many operations would you do, I decided that this may be an error. Is not that so? I apply about 10 effects, did not count, the value was 59.2 MB, it remains, and for Limiter dropped in the table for no apparent reason from 59.2 to 2 MB. I’ve downloaded Limiter from:
https://wiki.audacityteam.org/wiki/Nyquist_Effect_Plug-ins#Limiter
Before it was built-in - the problem repeated.

No it’s not an error.

I assume that you are asking out of interest (because you want to understand) rather than for a specific practical purpose, so I shall try to explain on that basis. It’s probably easiest if I give an example.

Note that I’ve approximated the values in MB because I can’t be bothered to work it out exactly, and also I’m only taking account of the audio data. In real life there is also some overhead due to the XML data and XML backup data.

  1. Starting with a new empty project.
    The amount of data used in the project = 0 MB, and there is nothing to undo.

  2. Generate 30 seconds mono tone @ 44100 Hz sample rate.
    30 seconds
    4 bytes per sample (default 32-bit float format)
    44100 samples per second
    = (about) 5 MB of data
    The project size is about 5 MB.
    There is one “undo” step (two items). The first item is a “null data” action of creating the empty project. Discarding this step will not remove any audio data, it would just prevent you from “undoing” generating 30 seconds of audio.

  3. Delete 5 seconds of the audio.
    The active project now contains 25 x 4 x 44100 = (about) 4MB
    The 5 seconds that were cut are retained as “Undo” data: 5 x 4 x 44100 = (about) 1 MB.
    There are now 3 items in the Undo history:
    1 - Creating the empty project
    2- Generating the tone
    3 - Deleting 5 seconds
    As before, discarding the first item will recover no disk space.
    Discarding the second step will delete the 5 seconds of data (about 1 MB) that were cut and are currently retained on the Undo stack (but not used in the current project state).

  4. Apply the “Amplify” effect to a 10 second selection.
    The original data in that 10 second selection is pushed onto the Undo stack. That’s about 2 MB of data.
    The selected region is replaced by new data from the Amplify effect, so the amount of data in the current project remains at about 4 MB.

Note that the figures in the right hand column do NOT represent the amount of data “used” by the effect. They represent the amount of data that the command pushed onto the Undo stack. Thus the numbers represent the amount of disk space that will be released if undo levels are discarded (undo levels are discarded in order starting from the top of the list).

Sometime the numbers can look surprising, but if you monitor the size of the project data file as you work, you can see that the quoted figures do actually match the measured sizes.

Clearly! I’ve already realized that this is not a mistake. Thank you!