Jump to content

Question about Replay Gain and Inter-sample peaks


Epyon07

Recommended Posts

Hi there,

I know this is isn't the right place to discuss audio data encoding but I don't know elsewhere to post.

Anyway I applied some Replaygain tags to both my Flac/Mp3s (EBU R128 Standard, -18 LUFS loudness target) but I noticed but when I encoding the output of flacs and mp3 from the same source have different peaks and most of the mp3 go above 1 dB, I've tried to search about this and found out that this is called intersampling that happens due to data compression and it can be avoided during the final mastering steps, what I've read is that some players offer some headroom to avoid that clipping. does Poweramp offer that or should I do something before encoding the lossy files myself?

I'm always applying (prevent clipping according to Peak)  Althought I've read here that it should be applied only to genres with moderate to high peaks.  http://eolindel.free.fr/foobar0.9/Replaygain.php (I exclusively listen to metal). I don't know if that option is the answer to this .

So I'm curious about this because I listen to a lot of metal records with already annoying highs (hi hats, cymbals) all the way through albums in the lossless format, but now as I am moving to MP3 320 CBR i'm paranoid about my hearing and worried that replay gain might aggravate those highs gains.

And about this particular issue I know headphones play some role in that but is there some method to lower that issue on the fly without using EQ? The treble knob is sitting at 0 % by default and it can't go lower. Some of my favorite albums are hell to go through https://www.youtube.com/    watch?v=ZtvP80tsCWk 

Thanks in advance.

Link to comment
Share on other sites

I thought ReplayGain aimed to set the peak level of any given track to -14dB from digital full-scale maximum, precisely to allow for post-processing headroom? This is one of the things people seem to misunderstand about GR, it is not a volume 'maximiser' (a "turn it up to 11" tool) and can quite often result in quieter listening levels - albeit consistently so - rather than louder.  

Andre

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...