[dizzyorange]

I play tracks on shuffle so replaygain is a necessity, but I find it "overcompensates".  For example, heavy metal will be very quiet and then a soft acoustic track will be overwhelmingly loud.  I want some volume equalization, but I still want tracks are supposed to be loud to be loud, and tracks that are soft to be soft.  Is there any solution to this?

 

[cel4145]

I've read of a way to manually edit the replaygain tag so you can set it how you like it. But now I can't remember where. LOL

 

[dizzyorange]

Thanks for the helpful reply 

 hehe

 

[cel4145]

Thanks for the helpful reply   hehe

LOL

I can find a starting point for research. See this thread, and this post with a solution for Foobar: http://www.hydrogenaudio.org/forums/index.php?showtopic=76059&pid=666905&mode=threaded&start=#entry666905

I don't use Foobar2000, but that should get you started. It's free to download and use

 

[StudioSound]

JRiver has switched from ReplayGain to R128 in v19, and I find it does a much better job. You might want to try that.

 

[RonaldDumsfeld]

Why not report your findings on the official foobar support forum.
 
http://www.hydrogenaudio.org/forums/index.php?showforum=28
 
That way you get to talk directly with the devs. It's a win:win situation for you. If you have spotted a genuine issue which can be independently verified by accepted experts in the field then everyone benefits because there will be an updated version. If you are doing something wrong then you still gain because you have learned something new.
 
With all due respect you are peddling an unsubstantiated personal anecdote as fact. There are dedicated people trying hard to get this right for all of us (for free).
 
Why not become part of the solution?

 

[cel4145]

Doesn't sound like there is a problem that he needs to report to Foobar. That sounds exactly like how replaygain works. It has to do some kind of averaging/norming of the songs volume. It's an algorithm that has no sense of music genres and definitely WILL make a song that should be louder sound the same as one that should be softer.

 

[RonaldDumsfeld]

Not if you use album gain intelligently it doesn't..
 
http://wiki.hydrogenaudio.org/index.php?title=Replaygain
 
ed: added link. To save reinventing the wheel.

 

[cel4145]

Not if you use album gain intelligently it doesn't..

Not everyone listens to their collection by album.

Not gonna help if someone is using a playlist with songs from different albums, such as one album that is all acoustic vs. one that is all heavy metal, to borrow from the OPs example.

 

[RonaldDumsfeld]

RTFM or take it up with the experts. PM if you wish to continue please.

 

[cel4145]

RTFM or take it up with the experts.

No need to get rude if you don't know understand how replaygain normalizes well enough to have a discussion about it. In fact, I'm not sure you understood the point that the OP was trying to make :rolleyes:

 

[dizzyorange]

LOL

I can find a starting point for research. See this thread, and this post with a solution for Foobar: http://www.hydrogenaudio.org/forums/index.php?showtopic=76059&pid=666905&mode=threaded&start=#entry666905

I don't use Foobar2000, but that should get you started. It's free to download and use

 
Cool, thanks for that.  It's not an ideal solution (which would be automatic) but if I hear a particular song that too loud or too soft I can change the value on the spot so when it plays again it'll be better.
 

  JRiver has switched from ReplayGain to R128 in v19, and I find it does a much better job. You might want to try that.

 
Hmm interesting but JRiver is payware right?  Plus I love the simplicity of foobar, I have it stripped down so it's just a list of songs and nothing else.
 

  Not if you use album gain intelligently it doesn't..
 
http://wiki.hydrogenaudio.org/index.php?title=Replaygain
 
ed: added link. To save reinventing the wheel.

 
That wouldn't work because soft albums will still be too loud and vice versa.  It would be helpful to someone who listened to albums one at a time instead of songs on shuffle.  Plus a lot of my songs don't have an album in the tags.
 
An ideal solution for me would be an "intensity of effect" variable that could be set for replaygain.  For example, say a soft acoustic track gets assigned -1.00dB and a loud rock track gets assigned -9.00dB.  Perhaps, by setting the "intensity" variable to 50%, the soft track would be assigned -0.50dB and the loud track -4.50dB.  To avoid clipping a certain global dB reduction could then be made to all songs after the track-specific changes are made.
 
^hope that makes sense. 

 

[cel4145]

An ideal solution for me would be an "intensity of effect" variable that could be set for replaygain.  For example, say a soft acoustic track gets assigned -1.00dB and a loud rock track gets assigned -9.00dB.  Perhaps, by setting the "intensity" variable to 50%, the soft track would be assigned -0.50dB and the loud track -4.50dB.  To avoid clipping a certain global dB reduction could then be made to all songs after the track-specific changes are made.

^hope that makes sense. 

That would be useful. Or at the very least, would be nice to have a media player implement it as a per genre adjustment that is stored in the player db (instead of the mp3 file).

 

[StudioSound]

Hmm interesting but JRiver is payware right?  Plus I love the simplicity of foobar, I have it stripped down so it's just a list of songs and nothing else.

 
It is, but they have a fairly long demo period - I think you get a month to try it.
One of the things I like about the program is that it's highly customizable - you can have as much or as little interface as you want. Depending on your computer proficiency, you may find the learning curve for that quite steep though.
 

  An ideal solution for me would be an "intensity of effect" variable that could be set for replaygain.  For example, say a soft acoustic track gets assigned -1.00dB and a loud rock track gets assigned -9.00dB.  Perhaps, by setting the "intensity" variable to 50%, the soft track would be assigned -0.50dB and the loud track -4.50dB.  To avoid clipping a certain global dB reduction could then be made to all songs after the track-specific changes are made.

 
An ideal solution would be to replace ReplayGain with a better method of analysis...

 

[dizzyorange]

   
It is, but they have a fairly long demo period - I think you get a month to try it.
One of the things I like about the program is that it's highly customizable - you can have as much or as little interface as you want. Depending on your computer proficiency, you may find the learning curve for that quite steep though.

 
I've been wondering for a long time: does JRiver really sound better than other media players?  I wouldn't use crossfeed or anything like that.  I mean, compared to WASAPI foobar (which is supposedly bit-perfect), does it really sound better?  Is it even theoretically possible for one software player to sound better than another if both are bit-perfect?  I know much is made of the 64-bit processing but I keep volume at 100% all the time anyway and don't use any type of sound processing.
 
^if the above sounds skeptical it's because I am, but I don't mean to be antagonistic by any means.  I'd like to hear your opinion if you have one.  But please don't say something like "FLAC is nearly as good as CD but a notch below" because stuff like that makes me disbelieve the speaker instantly   I'm only saying that because one proponent of JRiver said something like that in another thread