Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[ken-tajalli]

I have used "Advanced Tools" & "analyze Audio" tools to measure the peak levels of a track.
If I do not agree with the results, is it possible to modify them?
I have searched for an answer and so far it seems I can not.
Any Ideas?

[Hendrik]

This is a measured and calculated value, how would you not agree with them?
Because they are calculated, they are not editable.

[ken-tajalli]

I know, I know, but humor me, is it possible.

[ferday]

No, it's not.  These are calculated values, not mere tags

The only way to modify is to use a program like Audacity to actually drop/raise the peak level and modify the file itself

[sKiZo]

Been a perennial pet peeve with me too ...

Calculated or not, in many cases, the volume levels are just plain wrong. IMHO ... "volume leveling" would suggest level, as in volume, as in not having to adjust the playback between albums.

I see that now (as of MC19 anyway) we can edit the values in the analysis tables ... but ... (big but here) they don't seem to affect what I'm hearing. Example ...



Top table shows the edited values, bottom the original set by the analyzer. One would expect to hear a SIGNIFICANT difference (50db??) and nada. Here you can see that, yes, the system is configured to use the analysis tables on playback ... unless I'm missing something here ...



Maybe I'm just targeting the wrong values when I edit the record??

What would be PERFECT is if there were a MASTER volume setting per album. Playback is consistently too high or too low across the tracks, so it'd be killer if I could just tweak that as needed and have that saved for an album.

PS ... the edited values stick in the table even after a full MC exit and restart, so I know the editing function works. Sort of ...

The only way to modify is to use a program like Audacity to actually drop/raise the peak level and modify the file itself

Certainly an option, but we're talking guesswork there. Even noting the db drop on the volume slider in MC and matching that to a corresponding db drop in Audacity had inconsistent results, and there's no similar ability to gauge a db INCREASE required to balance the material with the majority of your media.

[Hendrik]

Volume Level (Replay Gain) is not used for playback in MC, only Volume Level (R128) is used, since its more accurate.
There is absolutely no reason to manually "tweak" any values, the measured values are very accurate as it analyses every single second of the audio.

If you want to hear an equal volume level across all tracks and albums, do NOT enable "Adaptive Volume", it'll destroy the benefit of perfect EBU R128 volume leveling. Only enable Volume Leveling, and disable Adaptive Volume.
Adaptive Volume is designed to maximize volume, not level volume, which can even result in an opposite effect.

[Zhillsguy]

Volume Level (Replay Gain) is not used for playback in MC, only Volume Level (R128) is used, since its more accurate.
There is absolutely no reason to manually "tweak" any values, the measured values are very accurate as it analyses every single second of the audio.

If you want to hear an equal volume level across all tracks and albums, do NOT enable "Adaptive Volume", it'll destroy the benefit of perfect EBU R128 volume leveling. Only enable Volume Leveling, and disable Adaptive Volume.
Adaptive Volume is designed to maximize volume, not level volume, which can even result in an opposite effect.

Excellent information. The VL R128 tag can be edited, and I had to do it once or twice for concert videos. The one that comes to mind atm is Sarah McLachlan Mirrorball... from what I gathered the 5.1 track was the one analyzed but since I use the stereo track the R128 tag needed to be adjusted to bring the resulting playback volume up to be consistent with others (my setup is 2.1). I believe this was discussed in a thread a while back - being able to choose which audio track is analyzed for video(?).

[Hendrik]

I believe this was discussed in a thread a while back - being able to choose which audio track is analyzed for video(?).

You could play the file once, switch the audio track and stop playback. MC will remember your choice. Then re-run analysis and it'll analyse your selected track.
There isn't really a better solution at this time, unfortunately. The assumption is that different tracks on one video would be mastered at least somewhat similarly.

[mojave]

If you want to hear an equal volume level across all tracks and albums, do NOT enable "Adaptive Volume", it'll destroy the benefit of perfect EBU R128 volume leveling. Only enable Volume Leveling, and disable Adaptive Volume.
Adaptive Volume is designed to maximize volume, not level volume, which can even result in an opposite effect.

If Volume Leveling is checked, then Adaptive Volume still allows for equal volume level across all tracks and albums in Playing Now. However, different Playing Now's can have different volumes.

With convolution I lose some volume, so I prefer to maximize it in a playlist by using Adaptive Volume. For those with low output voltage, such as a laptop, it can be very beneficial. Also, Adaptive Volume also prohibits anything from exceeding 0 dbTP if  Peak Level (R128) is higher than 0 dbTP. It is rare that I am listening at 100%, but others might find themselves pushing the limits more often.

Regarding multiple movie audio tracks, if one wants all analyzed then I suggest creating a Particle(s) for the other audio tracks. This way you have an entry in the Library for each track and can select whichever one you want and playback with Volume Leveling.

[sKiZo]

OK then ... turned off "adaptive volume" ... and now everything dropped. I had been running more or less at 93% on internal volume, and now have to boost the EQ in DSP by 1.5db to get the same approximate SPLs out of the system with the internal volume slider at 100%. Huh. I could live with that if there isn't a better way. Probably a bit more boost in the EQ to get a wider "fudge factor" for tweaking the volume if needed.

One point of confusion. Example track - Looking at the ANALYZE AUDIO table, I see:

- Volume Level (R128) is -11.2
- Volume Level (ReplayGain) is -6.7

In the file list on the main screen, I also see a R128 column with a value of -10.2db. Huh?

I assUme this is a calculated field, and the one I should be adjusting? I don't see that field as available in the ANALYZE AUDIO table even as an option.

Which begs the question ... is R128 set on a sliding curve, and is there any way to do a macro adjustment on that? I'd say a majority of the tracks in my library are in the -8db to -10db range in that field. Not a biggie ... just seems strange that I wouldn't be seeing results that are more zero-centric ...

[mark_h]

I have used "Advanced Tools" & "analyze Audio" tools to measure the peak levels of a track.
If I do not agree with the results, is it possible to modify them?

It is possible to modify such fields but you have to use a little trick to do so...

Add a User Field to your library, eg call it:

Peak Level Edit

Make it a calculated field and in the expression box put the name of the field you want to edit, eg

[Volume Level (ReplayGain)]

Now, in your view show your Peak Level Edit field and you can then F2 edit it, which will in turn write the value into the field you choose.

[sKiZo]

There's an option ...

Let's say ALL the values (with few exceptions) are consistently low by "6" ... How would I go about setting the field to add a fixed amount to the value, yet still allow me to edit the value manually if desired?

Or am I doing basically the same thing by bumping up the master slider in the EQ section?

One other question comes to mind ... I've seen mention of whether Volume Leveling needs to be checked on when ANALYZING AUDIO. Seems to me I had a whole bunch of stuff in the library before that option became available. Don't suppose it'd be all that difficult to just rerun the entire catalog if it does make a difference.

PS ... I just did a couple test files with and without VL checked, with no value changes. Maybe I just got lucky?

[mojave]

Using Adaptive Volume > Peak Level Normalize is much safer than bumping up the master slider in the EQ section. When used in conjunction with Volume Leveling, Adaptive Volume makes sure that the volume is reduced as little as necessary in a particular playlist (Playing Now). Using the master slider can result in clipping.

Here are two examples of playlists. In one, Volume Leveling reduced volume by 18.1 dB and Peak Level Normalize increased it by 2.6 dB for a net reduction of 15.5 dB. In the second playlist, Volume Leveling reduced volume by 3.5 dB and Peak Level Normalize increased it by 6 dB for a net increase of 2.5 dB.

The two playlists are 18 dB apart. The second can play louder because the Peak Level (R128) value of songs is lower and the intrinsic volume difference among the songs is a lot less.

[sKiZo]

... which closes the great circle once again to where I started ... 

Some good stuff here though, and a few more options I hadn't been aware of. Back to playing with it again and hoping to stumble across the magic combination that works in MY world ... I'll let ya know if I trip over the Holy Grail along the way ... 

[sKiZo]

OK then ... let's try this a different way.

I have both Adaptive Volume and Volume Leveling enabled, and plan to leave it like that, as that works quite well for 99% of my collection. I still have the occasional album that screams at me - for whatever reason, MC doesn't have a clue what to do to make it right.

Recent example ... Alabama Shakes, Sound and Color. Flat out a speaker breaker after running the analyze tools. I usually run MC at around 93% internal volume, and I could make these tracks good at around 80%. That's -10db, so I went into Audacity to manually drop the volume 6db on each which should have solved the problem. Of course, once I analyze the new files in MC, they're back to screaming. dOH!

So ... next step ... manually edit the tags? I tried the [Volume Level (ReplayGain)] trick mentioned earlier, and that doesn't work. Maybe just editing the wrong tag? Here's a screenshot from Audacity with the current settings for one of the problem tracks. Any one tag that screams FIX ME to anyone?



Hep me, please! This should NOT be this complicated!!

[6233638]

Do you have the same problem if you disable Adaptive Volume? (as a test)

[sKiZo]

Been there. Done that. Disabling adaptive volume DOES "fix" the screaming issue on the AS track pic'd above, but ... I lose a whole bunch of volume on the majority of my catalog.

However ... that did lead me to something that does seem to be working ... for now anyway. I disabled Adaptive Volume, left Volume Leveling engaged, and added a "volume increase" filter of +6db in the Parametric Equalizer settings. That 6db seems to work quite well overall, with a db or two exception based on the selection playing. Master internal volume is set at 93% to allow a fudge factor (I do need to boost some earlier vinyl rips that were set too soft) and also allows me to adjust as needed - good fer instance once again, the AS track pic'd still needs to be reduced a couple db for comfort, but they DID really overpeak it when recording, so not a surprise there.

Thanx for pointing me in the right direction ... still ... apparently adaptive volume isn't for everyone.

In the meantime, I'll keep my finger handy on the trigger in case something goes sideways and the whole catalog tries to shred my speakers ... <G>

PS ... Just noticed I'm in the MC20 folder ... still running MC19 (latest stable) here. Any changes in how volume adjustment is implemented that would make a difference? Not a fan of upgrading when I don't have to as I've had my share of problems with a lot of New! Improved?? software.

[6233638]

Been there. Done that. Disabling adaptive volume DOES "fix" the screaming issue on the AS track pic'd above, but ... I lose a whole bunch of volume on the majority of my catalog.

So your issue is Adaptive Volume then, not Volume Leveling, and not the tags.
 
Yes, Volume Leveling intentionally lowers the volume so that the average loudness is -23dB.
This is the only way to ensure that you have a consistent level across all tracks.
 
As soon as you increase the target or start boosting the volume, you are going to introduce uneven leveling, where some tracks will sound quieter than others because Media Center will not allow tracks to clip.
 
The real solution is to enable Volume Leveling and then turn up the volume on your amplifier.