- This topic has 12 replies, 4 voices, and was last updated 16 years, 4 months ago by
.
-
Posts
-
Hi,
from the FAQ / Forum I assumed that Breakaway sort of ‘hijacks’ the volume control from all sound generating applications on the PC (that is, I don’t really hear any difference in volume if I try to change it in Winamp for example… I need to change it in Breakaway).
This is fine. A bit unconventional, but OK.
However, after using my licensed copies on my 2 PCs, I realized that it is not always so! It entirely depends on what is playing in WinAmp! I do have a lot of MP3 items that when played render the Winamp volume slider completely ineffective. However, I also have a few MP3 files that when played will always make the Winamp slider work in addition to Breakaways volume control… 😯
The first question is of course whether this is something out of Breakaway, relating to the initial dynamic range of the sound, or do I utterly misunderstand something? It seems that this phenomenon happens with rather low dynamic sound files — like MP3s generated from very old sound archives, more often than not mono and rather low on volume, too. I also noticed that Breakaway’s scope shows almost no ‘input’ dynamic range for such files, however if I turn Winamp’s volume all the way up, then Breakaway does produce extensive sound enhancement on these, too.
Secondly, if it is Breakaway, is there any obvious rule that can be learned here? Like "Breakaway will not be triggered by low volume / dynamic input and all sound generating apps will behave as if there was no Breakaway installed" ?
I still need to experiment with other sound generating apps like MPlayer Classic, etc. to see if it is only Winamp & Breakaway or if it is truly the type of sound that is playing, irrelevant of the actual source app, but it all seems worth noting.
Thanks!
Hi Pekkaar,
Concerning varying audio levels….
I cannot answer for Breakaway – however – if you want consistant audio sound levels to come out – it greatly helps any system if you put known sound levels in. If your audio levels vary wildly – the system will struggle to cope. It might take you ages to do – but it is best for your overall sound to go to each of your mp3 files and to "normalise" the file to a known volume level. I personally use Cool Edit Pro to do this (There are many editing programmes that do the same thing) on my PC it takes about 15 seconds for a 3 minute cut.
Many of my older mono mp3 files were so low that you could barely see them on the waveform – I would imagine that had I left them like that – they would have fallen below the threshold at which Breakaway best kicks in and takes over. Leifs system works miracles – but you need to do a bit of work to get the best from it.
Mike 8)
to change volume on mp3s reliably, and losslessly (including ability to undo the change):
http://mp3gain.sourceforge.net/as far as the proper way to use breakaway with winamp, or any other software that provides a volume control… the software’s volume control should ALWAYS be set to maximum output.
you must use breakaway’s volume setting to control the output volume of the soundcard, after your soundcard’s physical master output (not wave out) has been properly matched with whatever system it’s running in to… be it an amp, or small crappy speakers, or headphones.
this way Breakaway’s volume knob at 100% should be as loud as you will ever want to turn up the sound, and then turn down Breakaway’s volume setting to where-ever you happen to want to listen at the moment.
Thanks knightrider, JesseG!
I did apply quite a bit of gain (+10dB, it is now 95dB & clipping a bit) to this particular MP3 with MP3gain, to not much avail.
I am following the Breakaway recommendations to turn all sound generating apps’ volume level to max, and am using Breakaway’s ‘knob’ to drive the volume all the time. I assume the outcome of this configuration is that changing the volume on the sound apps does not make any difference as Breakaway continuously adapts the sound it produces to the changing (decreasing volume) input.
This works OK for 99% of my files, but not all! Some of these files (one of which I added a significant gain now to no avail) will enable WinAmp’s volume slider to make a difference… it is more of a question than a problem report, really. I assume that with the gain added we can put the ‘low volume = no processing’ idea aside?
So what am I observing here? Is there any analysis done on the input stream the outcome of which could defeat the whole processing? Why am I seeing this only on some files?
Many thanks,
PeterHi Peter,
Peter – do you want to send me a copy of the offending MP3 file – I’ll take a look at it (listen) and try and tell you why it sounds cruddy.. 😐
Hi Peter,
I took a look at the description of the MP3Gain system – it got me wondering – does that system just flag the file in the tag information – giving an instruction to turn the volume up? Or – does it actually edit the file itself? I suspect that it flags via the tag. That might be the problem – if something in your audio chain fails to read that instruction – the cut will playback at the original low level…..????
This is just a guess from me – but it might pan out.
Try doing as I suggest – send me a file which is giving you grief – I will "tweek" it and return it to you to see if the problem goes away….
Mike
Hi Peter!
I too would love a copy of that MP3 file – I have no idea how volume control could work for one MP3 file when it’s disabled for all others.
How about you post it here as an attachment for everyone who wants to look at it? 🙂
///Leif
[quote author=”knightrider”]I took a look at the description of the MP3Gain system – it got me wondering – does that system just flag the file in the tag information – giving an instruction to turn the volume up? Or – does it actually edit the file itself? I suspect that it flags via the tag. That might be the problem – if something in your audio chain fails to read that instruction – the cut will playback at the original low level…..????[/quote]
mp3Gain will do a few things. First off, it’s using the ReplayGain algorithm to read the track, to decide it’s actual "loudness".
Then it will figure out the difference in relation to the reference db level you’re asking mp3Gain to make all of the mp3s you’re going to be processing.
After having that all figured out, when you click the button to apply the gain settings… It’s doing three things.
1. it changes a scale value built into all mp3s, which does NOT re-encode the file, but only changes how loud it will decode. and this is the function that works in ALL mp3 players of any kind. this is really where the magic happens. 🙂
2. it’s setting/changing ReplayGain values in an APE tag in the mp3 files, which is in place partly so that ReplayGain-capable players don’t try to figure out the gain and then change the gain, and also so they don’t bother to write the ReplayGain values, since they are already there. This also allows the player to use it’s own reference playback loudness, without any analysis.
3. it’s setting the mp3Gain correction values in an APE tag in the mp3 files. This will allow the mp3Gain modifications to be undon in the actual mp3 file’s scaling as well as reversing the changes in the ReplayGain APE tags, so that ReplayGain-capable players will continue to properly adjust the gain to whatever the reference loudness is set in that player.
Pretty neat eh? Actually, mp3gain will sound more transparent than normal ReplayGain, the more extreme the gain changes in ReplayGain are… because ReplayGain-capable players are altering the gain after the mp3 has been decoded back into PCM, but mp3gain is altering the gain *BEFORE* decoding into PCM, so you don’t lose a single bit of accuracy. 🙂
As is usually the case – answers from JesseG have me in awe – this is no exception! 8)
Leif, knightrider,
appreciate your efforts!
I was about to send you one file that seem to produce the issue every single time… however it is a 30+ MB vinyl compilation from podOmatic. I intended to cut ~1 min out of the beginning, but interestingly enough this short piece does not show the issue.
So, how do you think I can send you the 30MB file? I am not sure I can do that from my gmail account. Any free file pusher service you fancy?
Peter
Hi Peter,
I use http://www.mediafire.com/ to swap large files – 100% free and reliable.
Upload your file – then copy the URL (There is a button specially provided to do just that) and post it here on the forum – then we can all go grab a copy 🙂
Mike
Thanks for the links! I uploaded this vinyl collection here:
http://www.mediafire.com/?jlgj0223c19Nevertheless I spent some additional time investigating the issue and collected a few other MP3s just to see what could be triggering it. I found some files from a completely different genre (Linkin Park) but it shows that the ‘thing’ is not relevant to old, vinyl based, scratchy, etc. sounds. I assume the post-processing on these LP tracks are also way different. Just copied one of these tracks to here as well FYI:
http://www.mediafire.com/?monx2dm9nxhI don’t seem to find any commonality in the MP3 parameters of these two files… could this be the way these MP3 were ripped?
Regards,
PeterHi Peter,
I have looked at, and listened to, both pieces of audio. Nothing obvious hits me that would explain your problem!
The jazz big band track was ripped at 128kbps/44,100Hz (mono) – which is a popular format.
The audio was constantly peaking at 0dB – which is fine – and the dynamic range was fine – it was quiet when it was meant to be.The Linkin Park track was ripped at 96kbps/32,000Hz (stereo) – which is a low quality choice.
This was constantly peaking over 0dB – and the sampling rate is not one I would suggest you use if you want to get the max out of the sound.Here is a screen shot of how the entire Linkin Park song looks – showing all those peaks
Over on the right – you see my output meters – top of the scale are two red squares – these are clip indicators – when they go red – the audio is way too loud 🙂Both used CBR (Constant Bit Rate) which I would encourage you to use over VBR (Variable Bit Rate) because I have known of problems with computerised playout systems incorrectly reading the durations on VBR recordings….
As far as your problem goes – I can find nothing on your MP3 files that leads to any helpful suggestions!
On the other hand – I can suggest that if you are ripping tracks yourself – that you consider using a higher quality sample rate – it eats more disc space – but Gbs are not expensive these days – and the quality of a cut ripped at 320kbps/44,100Hz stereo is waaaaaay better than either of the cuts you supplied. Perhaps Linkin Park is not the best way to show you this – as they introduce so much "grunge" into their mix – but – I have uploaded a copy of Given Up – ripped as I describe above. Give it a listen and see what you think – also use it to test if the same problem occurs.
Get it here: http://www.mediafire.com/?meguoydzgqb
Another question – are you using any other sound "treatment" in your set up?
Let me know how my version of the song pans out.
BTW – I really don’t much like Linkin Park – not my style – but try playing it without – and then with Breakaway – at least Breakaway makes it sound almost acceptable 😆
Mike
- You must be logged in to reply to this topic.