View Issue Details

IDProjectCategoryView StatusLast Update
0002843ardourbugspublic2020-04-19 20:14
Reporterardmac Assigned Tocth103  
PrioritynormalSeverityminorReproducibilityalways
Status closedResolutionfixed 
Product Version2.7.1 
Target Version3.0-beta1 
Summary0002843: new region gain ^ doesn't work if new region was part of previously normalized region
Description- select new track, normalize
- create new region that contains amplitude well below max
- select the new region, ^ does not work

for ref from IRC:
[07:03] <drmacro> las: yeah, we went through this before, but the thing I didn't get was the new region with amplitude normalized, but still below max still remembers it has been normalized
[07:05] <las> drmacro: please file that as a bug
[07:05] <las> drmacro: your summary just now was the key part
TagsNo tags attached.

Activities

olaf

2009-09-03 12:45

reporter   ~0006640

it is not about forgetting the normalisation or not it is about that the amplitude with ^ will not get higher than 1
see editor_ops.cc:
            fraction += 0.05;
            fraction = min (fraction, 1.0);

cth103

2009-09-09 16:44

administrator   ~0006656

Hi, when you say "select new track, normalise", what do you mean by "normalise"? Normalise a region on that track?

ardmac

2009-09-12 17:55

reporter   ~0006660

- Select a whole track or region, then normalize.
- now create a new region, in the track or previous region where the max level is say -20.
- this new region can be decreased in amplitude, but not increased. Even though there is no part of the amplitude anywhere near max

ardmac

2009-10-13 13:57

reporter   ~0006698

I have noticed in some cases, the region in question can't be reduced as well.

I have not been able to reproduce this at will.

Also, when the region in question has been decreased, it can't be increased back to what it was before the decrease.

It also appears that if a region level is below some threshold before normalization, it will not increase.

olaf

2009-10-13 15:29

reporter   ~0006701

"- now create a new region, in the track or previous region where the max level is say -20."

what do you mean with create a new region?

ardmac

2009-10-14 11:12

reporter   ~0006703

- split 'S' a track to create a region that is time 0 to 20 where the level between time 0 to 10 is level -20 and the level between time 10 to 20 is -1.
- normalize this region, now time 0 to 10 is -19 and time 10 to 20 is 0
- now split this region at time 10
- select region that is now between time 0 to 10 (level, as per above, is -19)
- shift^ does nothing
- shift& will reduce this level
- shift& again will reduce this level further
- shift^ will not increase

olaf

2009-10-17 05:16

reporter   ~0006715

Last edited: 2009-10-17 05:17

ok your description is absolutely what i see in the code of editor_ops.cc.
there is a maximum amplification for ^ but not for normalisation.
i think there are two solutions:
1. no maximum border for amplification with ^
2. the maximum is set to 1 (or higher) and when there is a normalisation performed this maximum is set to the amplification factor used for the normalisation.

ardmac

2009-12-03 12:46

reporter   ~0007221

Additional note: when a split with very low audio level is normalized and then shift& is used it actually increases the (now max level) peaks into clipping rather than reducing the level.

It does not happen at first in a session, but once it has it happens it seems the inverse function of shift& is sticky for that track.

If the split is expanded to include some audio that is not so low level, the normalize/shift& reduces the level.

I have seen this while working on splits as described previously with this report and add this note just for detail.

olaf

2010-12-07 15:45

reporter   ~0009567

my suggestion would be to remove the check of the maximum volume for ^.
or at least put the maximum really high (100) I don't see a benefit for this border.

cth103

2010-12-12 01:03

administrator   ~0009599

This situation should be improved in A3 SVN. Feedback welcome.

cth103

2011-02-16 18:26

administrator   ~0010122

Two months without feedback, so I'll mark this as resolved. Please re-open if it is still a problem. Thanks!

system

2020-04-19 20:14

developer   ~0021985

Issue has been closed automatically, by Trigger Close Plugin.
Feel free to re-open with additional information if you think the issue is not resolved.

Issue History

Date Modified Username Field Change
2009-09-03 12:02 ardmac New Issue
2009-09-03 12:45 olaf Note Added: 0006640
2009-09-09 16:44 cth103 Note Added: 0006656
2009-09-09 16:44 cth103 Status new => feedback
2009-09-12 17:55 ardmac Note Added: 0006660
2009-10-13 13:57 ardmac Note Added: 0006698
2009-10-13 15:29 olaf Note Added: 0006701
2009-10-14 11:12 ardmac Note Added: 0006703
2009-10-17 05:16 olaf Note Added: 0006715
2009-10-17 05:17 olaf Note Edited: 0006715
2009-12-03 12:46 ardmac Note Added: 0007221
2010-04-24 10:28 cth103 Category bugs => bugs2
2010-04-24 10:31 cth103 Category bugs2 => bugs
2010-12-07 15:45 olaf Note Added: 0009567
2010-12-07 23:50 cth103 cost => 0.00
2010-12-07 23:50 cth103 Target Version => 3.0-beta1
2010-12-12 01:03 cth103 Note Added: 0009599
2011-02-16 18:26 cth103 Note Added: 0010122
2011-02-16 18:26 cth103 Status feedback => resolved
2011-02-16 18:26 cth103 Resolution open => fixed
2011-02-16 18:26 cth103 Assigned To => cth103
2020-04-19 20:14 system Note Added: 0021985
2020-04-19 20:14 system Status resolved => closed