Help - Search - Members - Calendar
Full Version: Ogg/ Mp3 Gain
Hydrogenaudio Forums > Hydrogenaudio Forum > General Audio
degarb
To get a listenable file in the sp/imp-100 one must make the files louder with http:// www.geocities.com/mp3gain/ This helps even with my imp-250 unit. Unless mp4s and ogg files can be made LOUDER, the ogg sound would be a "practical" step down in quality--at any bit rate! In everday practice ogg/mp4s would not be very listenable in a loud gym, vacuuming, or mowing, or driving in a loud work vehicle. Yet. the ogg gain program has no dang help at all, and "vorbisgain.exe -g10 file.ogg" only makes the files 1 db quieter.

Also, one often needs a parser that can create id3 tags from the file names--such as http://www.cyberianz.com/ahmet/ I personally name files Artist-album-side- tracknumber-name.ext when pulling my lps into mp3s.
Garf
I recommend getting a better unit.

Vorbisgain can never solve this, since it doesn't alter the actual sound data. It gives hints to the playback system, and if that playback system is so dumb it can't produce a decent volume to start with, I wouldn't expect it to ever work.
degarb
I think my winamp is too dumb to increase volume with vorbis gain. However, one would expect opensource snobs not to include a help file with the exe.

Ogg bothers me in two ways. One, the vbrs are too restrictive. I can set lame to wildly grab bits from 96 to 256, which it does. While ogg varies bit rates only slightly, seldom grabbing high rates on a 160 kps encoding. Two, simply the lame files sound better even on winamp and a loud player, because it cheats on reequalizing (as mp4) and can be made noticeably louder by default. Lame also exaggerates the parts of music that people like to listen to (highs and bass.) This is one reason I go with lame over wma.

Ogg needs to get the tools together in gain/eq/and tag encoding with ogg drop. Otherwise, I will stay with lame, and happily so. (though hoping for ogg, when ready.)


I do have a sp 250 and a 4 db gain improves mp3 sound immensly as well.
kjoonlee
QUOTE
vorbisgain(1)     VorbisGain    vorbisgain(1)



NAME
      vorbisgain - calculate the replay gain for Ogg Vorbis files


SYNOPSIS
      vorbisgain [ -acCdfhqrsv ] [ -g album gain ] input_files ...


DESCRIPTION
      vorbisgain  calculates  the  ReplayGain values for the named Ogg Vorbis
      input files and writes back the result in the form of  tags  (comments)
      in the file. These values can be used by a playback program to maintain
      a uniform sound level during play. (See http://www.replaygain.org/  for
      more  information.)   vorbisgain  uses a default target level of 89 dB,
      rather than the 83 dB recommended by the ReplayGain standard, when cal-
      culating  the  gain to apply. (Some players include a pre-amplification
      setting with which the target level can be changed.)


      vorbisgain input files must be Ogg Vorbis I files with 1 or 2  channels
      and  a  sample  rate  of 48 kHz, 44.1 kHz, 32 kHz, 24 kHz, 22050 Hz, 16
      kHz, 12 kHz, 11025 Hz or 8 kHz. If  an  input  file  contains  multiple
      streams (i.e., it is chained), the streams must all have the same for-
      mat, in terms of sampling frequency and number of channels.


      All streams in a chained file are processed, but  the  ReplayGain  tags
      are only written to (or removed from) the first stream.


OPTIONS
      -h, --help
      Show command help.


      -a, --album
      Activates  album mode, in which the album gain (sometimes called
      the audiophile gain) is calculated in addition to the track gain
      (sometimes called the radio gain).


      -g n, --album-gain=n
      Sets the album gain value to use. Default is to calculate it, if
      -a is specified. Implies -a in the  sense  that  the  audiophile
      gain tag is written. This is a relative value, in dB, specifying
      the change in volume that should be applied.


      -c, --clean
      Remove any ReplayGain tags from the specified files. If  a  file
      does  not contain any ReplayGain tags, the file is not modified.


      -C, --convert
      Convert old format ReplayGain tags to a new format (see  section
      TAG  FORMAT  below  for details). If a file does not contain all
      ReplayGain tags that are needed for a conversion,  the  file  is
      not modified. The album peak tag is only created if -a is speci-
      fied, and the album gain value is then checked for  concistency.
      Otherwise any album gain is converted without any checks.


      -d, --display-only
      Display  the  result  only;  do not write anything to disk. This
      applies to all options.


      -f, --fast
      Only calculate the gain for files that do not contain all replay
      gain  tags needed (the album gain and peak tags are only consid-
      ered if -a has been specified).


      -q, --quiet
      Do not display any output while processing. Only error and warn-
      ing messages will be printed.


      -r, --recursive
      Enter  directories (recursively) and search for files, if direc-
      tories or file patterns are specified.  Note: Only available  if
      vorbisgain was configured with --enable-recursive.


      -s, --silent
      Sliently skip  any  non-Vorbis  files  found. Vorbis files that
      can't be processed for some reason are skipped as  well, though
      not  silently.  Default  is  to stop when such files are encoun-
      tered.


      -v, --version
      Display the version of the program.


EXAMPLES
      Simplest version. Calculate the track gain and peak only.

      vorbisgain somefile.ogg


      Note that the following examples are only possible  if  vorbisgain  was
      configured with --enable-recursive.


      Calculate  the  album  gain and peak, in addition to the track gain and
      peak, for all .ogg files in the directory "music" (and all  subdirecto-
      ries).  All files in one directory are treated as belonging to the same
      album. Files that already have ReplayGain tags are  not re-calculated.
      Note  the  quotes, as they cause the shell to not do any filename glob-
      bing:


      vorbisgain -a -f -r "music/*.ogg"


      Calculate the album gain. The  files  specified before the  directory
      "album" are treated as one album, the files in the directory "album" as
      another album and the remaining files as a third album:


      vorbisgain -a -r a.ogg b.ogg c.ogg album d.ogg e.ogg f.ogg


      Remove all replaygain tags from a collection of oggs:


      vorbisgain -c -r "music/*.ogg"


TAG FORMAT
      vorbisgain creates tags like these (when in -a mode):


      REPLAYGAIN_TRACK_GAIN=-7.03 dB
      REPLAYGAIN_TRACK_PEAK=1.21822226
      REPLAYGAIN_ALBUM_GAIN=-6.37 dB
      REPLAYGAIN_ALBUM_PEAK=1.21822226


      Gain specifies how much the volume should be changed  before  playback,
      in  dB.  Peak  is the maximum sample value of the file before any gain
      has been applied, where 1.0 means  "full  sample  value"  (32,767  when
      decoding to signed 16 bit samples).


      Earlier versions  of  vorbisgain  (before  0.30) created the following
      tags:


      RG_RADIO
      RG_PEAK
      RG_AUDIOPHILE


      When -c is  used,  RG_RADIO  is  converted  to REPLAYGAIN_TRACK_GAIN,
      RG_PEAK   to REPLAYGAIN_TRACK_PEAK,   RG_AUDIOPHILE   to   REPLAY-
      GAIN_ALBUM_GAIN and REPLAYGAIN_ALBUM_PEAK is calculated as the  maximum
      of all RG_PEAK tags in the album.


AUTHORS
      Program Code:
      Gian-Carlo Pascutto <gcp@sjeng.org>
      Magnus Holmgren <lear@algonet.se>


      ReplayGain Analysis Code:
      Glen Sawyer <glensawyer@hotmail.com>
      Frank Klemm (http://www.uni-jena.de/~pfk/mpp/)


      Man Page:
      Magnus Holmgren <lear@algonet.se>


BUGS
      None known.


SEE ALSO
      http://www.replaygain.org/
      Contains detailed  information  about  ReplayGain and how it is
      calculated.


      http://www.hydrogenaudio.org/
      Discussion forum for  audio  compression  and  related  issues,
      including Ogg Vorbis and VorbisGain.



     2002 Jule 4    vorbisgain(1)


Kind of off-topic, but..

IMHO Winamp2 together with in_vorbis 1.2.8 and 'vorbisgain'ed oggs is pretty much perfect as far as PC playback is concerned. (You can set pre-amp separately for oggs and mp3s, and you can even choose to apply pre-amp to oggs only if replay gain tags are available.)

On topic now: http://winvorbis.stationplaylist.com/ Have you tried this frontend? It can scan filenames, do vorbisgain, and can rename your oggs for you.
Garf
QUOTE
However, one would expect opensource snobs not to include a help file with the exe.


Why, thank you. As someone else already noted, there is a manual, but apparently you just didn't download it nor did you try the '-help' option.

QUOTE
Ogg bothers me in two ways.  One, the vbrs are too restrictive.  I can set lame to wildly grab bits from 96 to 256, which it does.  While ogg varies bit rates only slightly, seldom grabbing high rates on a 160 kps encoding.  


You are completely wrong. It is not because the reported bitrates do not fluctuate wildly that the actual VBR encoding algorithm does not. The reported bitrate in WinAmp is averaged over a section of time, which is why it appears relatively stable.
degarb
Kjonlee, thankyou. However this is too confusing (what does one care about relative volume for my interest, or peak vol?), what is the simple setting to tell the player to CRANK UP the volume on all tunes several db? mp3 gain simple constant gain of 3-4.5 db works perfectly on winamp and the sp-100/250.

So far any setting, I hear no diff on winamp lite. Do I need a better player? I don't desire winamp 3 until the lite comes out.


I also see a new oggdrop with a gain, but again, hear no difference with any setting there.
Garf
QUOTE (degarb @ Oct 16 2002 - 06:03 PM)
So far any setting, I hear no diff on winamp lite.  Do I need a better player?  I don't desire winamp 3 until the lite comes out.  

Did you even bother to read my post?

Vorbisgain calibrates the volume so all music sounds equally loud. It requires support from the player. It doesn't change the volume level of the data in the ogg itself, you should use your players volume knob for that.
kjoonlee
degarb: To simply raise the playback volume in Winamp2, configure the Ogg Vorbis plugin to use pre-amp.

If you want proper Replay Gain, go install winamp2.81, better yet, install wa2update.exe too. (You can get it here, from the awesome Peter Pawlowski's page.) Then you'll have access to all sorts of goodies, such as selective pre-amp.

As Garf, the creator of vorbisgain.exe, has already mentioned, the replay gain tags are only for players with vorbisgain support. If you want equal loudness on your portable, you're on your own for now.
degarb
With MP3 gain, you can increase volume on the rio and winamp. This is done. This is possible. So is it superior to ogg? :-)

EXample: If I record a parent wav at 70-90 percent peak. (standard for 1950 to 1990 lps which can vary wildly) I can use mp3 gain simply add 3-4 db and get a listenable file on ALL players. (See rio file will never be loud as amp would make possible if at 100 percent possible.) (No raising wav volume IS NOT an option for many reasons, from time to perserving, to reversability.)

No, sorry, these directions make no sense. They loose me at words "peak, and relative" because I care not for either issues. Suggestion for an idiot setting? I say this not just for self but with out going into major geek mode, I've no time for that explanation. Nor do most people. Most people want to simply crank up the volume. While some want to make all songs same volume. All other intentions are greek to us.

Downloading and updating winamp, if possible. Thanks.
ManyFaces
:?

AFAIK, mp3gain makes all track equally loud.

VorbisGain works in the opposite way: if you make tracks sound too loud, you will be screwing the higher parts, so the trick is to lower the loud track to match the lowest tracks.

Al least, that is what i read in ReplayGain site.
Garf
QUOTE (ManyFaces @ Oct 16 2002 - 06:58 PM)
AFAIK, mp3gain makes all track equally loud.

VorbisGain works in the opposite way

No, they don't, they work exactly alike.
Garf
QUOTE
Most people want to simply crank up the volume.


They should not use Vorbisgain, because it's not meant to do that. There's already an excellent tool for that, named the 'volume knob'. Devices from 50 years ago already have it. If whatever you are using doesn't have a good one, I suggest you drop it in the trashcan.
degarb
[QUOTE=ManyFaces,Oct 16 2002 - 08:58 AM]:?


>VorbisGain works in the opposite way: if you make tracks sound too loud, you will be screwing the higher parts, so the trick is to lower the loud track to match the lowest tracks.
----

mp3gain has an alternate mode called constant gain, so I can make a 1948 album recorded at 57 percent peak, sound full strength.

The notion that cranking the 12mw headphone to all the way up, and it supposed to be loud or throw it out, is stupidity. Nor is it necessary. Encode in mp3 and use mp3 gain. MP3 gain is a nice way of getting the most out of all portable mp3 players. All of them are weak , or at least the high end ones I have tried.

And so, I would like to do with ogg. If not possible, I will stick to mp3s, and push ogg developers for the easiest of all tricks to make music sound better without a large amp or high bit rates.
ErikS
Hahaha. This is just too funny!

degarb, don't you understand what people say? You're totally missing the purpose of replaygain.
ManyFaces
ErikS: you didn't answered him as well...

Degarb: The deal is most albums nowadays are making all tones sound equally loud, including whispers, thus screwing the sound.

Instead of screwing the sound of even more albums, ReplayGain does the opposite: lowers the output of newer (and screwed albums) to match the level output of old ones.

...what you are asking for would be something like pre-amp, or kinda. I don't know.
sven_Bent
QUOTE
To get a listenable file in the sp/imp-100 one must make the files louder

looks to me that you amp is weak...

QUOTE
with http:// www.geocities.com/mp3gain/  This helps even with my imp-250 unit.  Unless mp4s and ogg files can be made LOUDER, the ogg sound would be a "practical" step down in quality--at any bit rate!

Digitally increasing volume will decrease quality as you run out of bit to describe the volume (clipping)
just like when you are altering the ligth level on pictures. if you keep on it will alle be white (1bit).

QUOTE
In everday practice ogg/mp4s would not be very listenable in a loud gym, vacuuming, or mowing, or driving in a loud work vehicle.

Funny i don't have this problem with my amp.
i throw my parties with AAC encodet music.
ErikS
One more way of rephrasing what everybody has already said:

Replaygain is not a volume knob and should not be used as such.




By the way, can the iriver players play Oggs already?
degarb
WINVORBIS IS NICE.

Since many albums I record are only 55 to 78 percent clipping level, adding a modest mp3 gain cannot cause clipping. Moreover, 3 db--my normal gain--seldom causes it in the most aggressive cd rips.

Note also, I have a 12mw amp in the imp-250 (formerly sp-100, yet couldn't take the skipping when jogging), yet my gym is louder than when I mow the lawn. Yet, my backup player is the paltry 7mw sp-100.

Yes, iriver will support ogg in a few months, and yes, for practical and political reasons, I would like to use ogg. Sooner is better than later. yet, I NEED a constant gain!
Hanky
QUOTE (ErikS @ Oct 16 2002 - 08:34 PM)
By the way, can the iriver players play Oggs already?

No, not yet.
Latest news is still the 2002-09-24 post from Emmett Plant (CEO Xiph.org) on the IRiver user forum telling us that IRiver's chief engineer confirmed to him that IRiver downloaded the 'Tremor' fixed point Ogg Vorbis decoder and they are currently in the process of porting it to their hardware.
And to get this post on topic , let's hope IRiver doesn't forget to implement support for Vorbisgain tagged files, which I believe they are not required to by the standard.
Lets keep fingers crossed rolleyes.gif
degarb
Downloaded and installed winamp 3. First 5 minute impression: Sounds far weaker and putrid than winamp lite 2.78, and alt 3 is crippled for ogg files.

Now, for the gentleman who threw parties with aac. I have tried aac, got the aac drop and psytel fastencoder engine. But the drop front end fails to control the quality settings. So I must have a missmatch of encoder to front end.

I am turned off at the restricted encoder issues with aac already. Drink from a hose that can be turned off at will, and find your self addicted to something you must pay for regularly.

My vote is with ogg. But mp4 is stunning/loud at 160 abr. Ogg just ressembles the original wav a little better. Perfect at 224 nominal, while acceptable at 96 (minimal artifacts at even 88).
rjamorim
QUOTE (degarb @ Oct 16 2002 - 06:28 PM)
Now, for the gentleman who threw parties with aac.  I have tried aac, got the aac drop and psytel fastencoder engine. But the drop front end fails to control the quality settings.  So I must have a missmatch of encoder to front end.


That's strange. It works perfectly well here. On both AACenc and Fastenc.

QUOTE
I am turned off at the restricted encoder issues with aac already.  Drink from a hose that can be turned off at will, and find your self addicted to something you must pay for regularly


I fail too see the send and/or meaning in your analogy. (Or is that a metaphor?)
degarb
>QUOTE

I am turned off at the restricted encoder issues with aac already. Drink from a hose that can be turned off at will, and find your self addicted to something you must pay for regularly



I fail too see the send and/or meaning in your analogy. (Or is that a metaphor?)
>End quote
------

That is why you prefer aac over ogg. If you and industry at large get addicted to a patent held aac. Who is to say they don't raise patent fees?

Perhaps, you know they cannot?

Anyway, I do think the writing is on the wall. AAC is here to take over in the next year. And her fame will blot out most knowledge of ogg.
rjamorim
QUOTE (degarb @ Oct 16 2002 - 07:25 PM)
That is why you prefer aac over ogg.

Well, the reasons I prefer AAC are these:

-High quality
-Extremely flexible
-ISO standard
-Fast decoding using FAAD2
-AAC patents aren't valid here in Brazil
-It's the only "next generation" (if you consider AAC, OGG and MPC) format to have hardware support
-DVD-A compatible
-...

Maybe, if patents were valid here in Brazil, I would be using something else. Don't know. Remember RareWares started as a distribution point for AAC binaries. What means I have been sympathetic to AAC for a long time (started with AACenc 1.2, which still sits at my HD). And I have been frequenting AudioCoding.com forum for much longer than HA.

Of course, license prices can go higher if AAC becomes widespread. But MP3 is already widespread - and the licensing prices at Thomson's site only decreased since they started charging. I wouldn't be too afraid of that happening then.

Regards;

Roberto.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Invision Power Board © 2001-2009 Invision Power Services, Inc.