INTERACT FORUM

Please login or register.

Login with username, password and session length
Advanced search  
Pages: [1]   Go Down

Author Topic: Using replay gain when crossfading cause dissortion  (Read 1510 times)

zirum

  • Galactic Citizen
  • ****
  • Posts: 403
  • still learning.
Using replay gain when crossfading cause dissortion
« on: October 05, 2007, 10:12:59 am »

Hi all!

I use replay gain. Since most songs are set pretty low (-12 to -16 db) I have set it to add 12db to output. The problem is when a un-analized song is play, this as well adds the gain. Fortunatly, clip protection may handle this, making the overall track volume pretty the same.

But there is one problem. When crossfading between the tracks without replay gain, the songs get's really disorted.

I was hoping for one of the two:
- Default gain level for tracks without replay gain info.
- Disable replay gain for tracks without replay gain info.

It's not a biggy, but a bit annoying. Any plans for adding such feature JRiver?
Logged
Note to myself: Read, think, write - Read, think, write - think, read, write - think, write, read - think, write, read... Aahhw, i always mess that up...

Matt

  • Administrator
  • Citizen of the Universe
  • *****
  • Posts: 42002
  • Shoes gone again!
Re: Using replay gain when crossfading cause dissortion
« Reply #1 on: October 05, 2007, 11:54:47 pm »

Hi all!

I use replay gain. Since most songs are set pretty low (-12 to -16 db) I have set it to add 12db to output. The problem is when a un-analized song is play, this as well adds the gain. Fortunatly, clip protection may handle this, making the overall track volume pretty the same.

But there is one problem. When crossfading between the tracks without replay gain, the songs get's really disorted.

I was hoping for one of the two:
- Default gain level for tracks without replay gain info.
- Disable replay gain for tracks without replay gain info.

It's not a biggy, but a bit annoying. Any plans for adding such feature JRiver?


Unanalyzed tracks will use the average gain of the analyzed tracks in Playing Now.

The theory is that you normally play similar loudness tracks at the same time, so this is the best guess we can make.
Logged
Matt Ashland, JRiver Media Center

raym

  • MC Beta Team
  • Citizen of the Universe
  • *****
  • Posts: 3583
Re: Using replay gain when crossfading cause dissortion
« Reply #2 on: October 06, 2007, 03:52:33 am »

Unanalyzed tracks will use the average gain of the analyzed tracks in Playing Now.

The theory is that you normally play similar loudness tracks at the same time, so this is the best guess we can make.

But when there's no other tracks in playing now, and you play unanalyzed tracks (like a cd for eg) you end up with the problem zirum is talking about as well. Distorted tracks (at the start mostly).

I've always wondered if it's possible for MC to auto bypass replay gain when all tracks are unanalysed??

Thanks.
Logged
RKM Smart Home - www.rkmsmarthome.com.au
Z-Wave Home Automation

Alex B

  • MC Beta Team
  • Citizen of the Universe
  • *****
  • Posts: 10121
  • The Cosmic Bird
Re: Using replay gain when crossfading cause dissortion
« Reply #3 on: October 06, 2007, 04:30:26 am »

I don't quite understand this:

But there is one problem. When crossfading between the tracks without replay gain, the songs get's really disorted.

Do you mean that if you set the playback option to Gapless the problem does not exist?

Quote
Since most songs are set pretty low (-12 to -16 db) I have set it to add 12db to output.

Instead of using a constant positive gain value you could use the "Automatic..." option. It automatically applies an adjustment value that makes the maximum peak level of the tracks in Playing Now to be 100%.

I usually have the "Album Gain" and "Automatic..." options enabled and normally I don't need to adjust the playback volume levels even when playing a various tracks playlist.

Though, I don't know how well this would work with unanalyzed tracks. I use only imported and analyzed library tracks in my playlists.
Logged
The Cosmic Bird - a triple merger of galaxies: http://eso.org/public/news/eso0755

Matt

  • Administrator
  • Citizen of the Universe
  • *****
  • Posts: 42002
  • Shoes gone again!
Re: Using replay gain when crossfading cause dissortion
« Reply #4 on: October 06, 2007, 08:20:36 pm »

Instead of using a constant positive gain value you could use the "Automatic..." option. It automatically applies an adjustment value that makes the maximum peak level of the tracks in Playing Now to be 100%.

I usually have the "Album Gain" and "Automatic..." options enabled and normally I don't need to adjust the playback volume levels even when playing a various tracks playlist.

Though, I don't know how well this would work with unanalyzed tracks. I use only imported and analyzed library tracks in my playlists.

This is good advice.  Automatic mode will do all the math for you to make sure you get the maximum gain with no clipping.

You'll need to adjust the volume for each new Playing Now, but not inside a given Playing Now.
Logged
Matt Ashland, JRiver Media Center

zirum

  • Galactic Citizen
  • ****
  • Posts: 403
  • still learning.
Re: Using replay gain when crossfading cause dissortion
« Reply #5 on: October 09, 2007, 11:00:20 am »

I don't quite understand this:

Do you mean that if you set the playback option to Gapless the problem does not exist?

I haven't tested around with the diffrent features what does and not, but I think it applies the boosted sound to the fading in (or out) song. Since the song is boosted to much, auto-clip adjust down, but it fading makes it stronger, and then it clips and so on. I have so much workload currently so i am not able to test right now.

The advice will probaly work for now, even though i suspect it will be a problem on my wireless (if the playlist overall volume is high, but the current song are to low, it will maybe hibernate). I'll test a bit later.
Logged
Note to myself: Read, think, write - Read, think, write - think, read, write - think, write, read - think, write, read... Aahhw, i always mess that up...
Pages: [1]   Go Up