Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Scanner find different gain for two track in same volume (in track gain) (Read 1699 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Scanner find different gain for two track in same volume (in track gain)

Hello

I recently scanned my musics to set replaygain using foobar2k version 1.3.15 (latest version) in windows 10 x64 with original package (no added plugin or anything else)

After applying gain I found that times that I change volume became more than without replaygain!

I tested and for example I uploaded two of my files that are in same level of loudness but foobar2k recognized and wrote different track gain value to them.

(The files that uploaded here are original files without replaygain tag but if you test will see that foobar2k will recognize different track gain values for them + if you hear to music without replaygain volumes are same)

Because they are in same album when I use album gain the problem solved but I need track gain because I usually play random tracks.

Also I scanned my files several times but problem didn't solve.

Please help to fix this or please help for another tool for scanning replaygain.

Thank you very much

Re: Scanner find different gain for two track in same volume (in track gain)

Reply #1 –
The first track has much more energy in the mid frequencies where ear is the most sensitive. Without ReplayGain it also sounds louder to my ears. It seems intentional that the mixer and/or artist intended the fast guitar track to be louder than the next slower track. And album gain is meant to preserve such artistic choices.

Simple RMS and the original ReplayGain algorithm also agree with the loudness differences. Original ReplayGain algo estimates the difference even higher than the ITU method used by foobar2000. Personally I'm very pleased with the results foobar gives.

Edit: Do you use EQ that alters the frequency response lowering mids? That would be one explanation to your perception difference.

Re: Scanner find different gain for two track in same volume (in track gain)

Reply #2 –
Thank you very much for your test and detailed answer.
I have increased midrange in EQ however May be this is because my headphones have weaker and a little veiled midrange.

May you please test these tracks also?

(These aren't in same album but when I use track gain and play with track gain applied, 03 is loader/harsher in my ear (violin parts) but 01 is weaker because of voice of singer and I should increase volume in 01 after I listened to 03 with my optimal volume But without replaygain when I jump from 03 to 01 or reverse, both have good volume and I don't need to change volume)

(I mean without track gain both these tracks are in same volume but after applied track gain to them and play them with track gain applied, I need change volume when I jump from one to another)

(These files I attached also are original version without replaygain tag)

Thank you very much

Re: Scanner find different gain for two track in same volume (in track gain)

Reply #3 –
Quote from: Case on
Simple RMS and the original ReplayGain algorithm also agree with the loudness differences. Original ReplayGain algo estimates the difference even higher than the ITU method used by foobar2000. Personally I'm very pleased with the results foobar gives.

This means when we play music with applied track gain the loudness of all tracks doesn't become same as other ?

Re: Scanner find different gain for two track in same volume (in track gain)

Reply #4 –
Quote from: Alidl on
May you please test these tracks also?
Without ReplayGain the 01 track sounds much louder than the 03 track.
But as a test I boosted midrange with foobar's EQ and played the tracks with default DirectSound output. Since the quieter track has plenty of headroom it was able to gain amplitude while Windows mixer's clipping protection prevented the louder track from getting louder. This kind of EQ and clipping protection abuse gave the illusion that the tracks could have similar loudness.

If you look at the spectrogram of the two files you can see that the first file has much more energy in all frequencies:

01 - TREASURE BOX.mp3:


03 - Tadoritsuki Tai Basho wa Doko.mp3:


With these tracks different loudness estimations have quite varying opinions. foobar's current scanner thinks track 01 is 6.98 dB louder, original ReplayGain scanner thinks it's 4.71 dB louder and simple RMS calculation tells the difference is 9.67 dB.

Note that playback volume affects how loud different frequencies appear. If I keep volume very quiet I think original ReplayGain's adjustment changes these files to be closer to each other but when playback level is more reasonable I think the current scanner used by foobar2000 offers better results.

Re: Scanner find different gain for two track in same volume (in track gain)

Reply #5 –
Thank you very much for your detailed answer.