IPB

Welcome Guest ( Log In | Register )

3 Pages V   1 2 3 >  
Reply to this topicStart new topic
do you lose quality when lowering a wav file volume then raising it up
coolhotfun
post Apr 16 2013, 17:25
Post #1





Group: Members
Posts: 25
Joined: 4-July 11
Member No.: 92047



ok i'm editing some sound samples and they're starting to get annoying hurting my ears after editing stuff for hours.. so i went thru them all again later and got a lower level of volume, where you feel that its a bit too low. I'm wondering if i go to raise up the volume later in the future will i lose some quality? Wasnt sure if there's a threshold where if you lower the volume enough it loses that data. Ran a test where i lowered it by -40db and it was tiny, then raised it back up by 40 and compared it to the original and seemed identical. Saved a -20db file and a -40db file and they were both the same file size. So i'm guessing it's keeping all that data even though the volume is lower. But in the volume level tests i lowered it to the minimum, -inf db.. which seems to be a step past -59 so maybe -60, if i try and raise it up by +60 after, can't do it.. its mute like if it lost the data. 59 and back can work though. I'm wondering if some of those files i'm working on am i losing some part of the sound spectrum by it being a lower volume or can i raise it up back later and it would be the exact same file as if i kept it louder.

This post has been edited by coolhotfun: Apr 16 2013, 17:30
Go to the top of the page
+Quote Post
Propheticus
post Apr 16 2013, 17:30
Post #2





Group: Members
Posts: 219
Joined: 10-September 11
Member No.: 93615



why don't you just lower your playback level instead of altering the waves amplitude?
Go to the top of the page
+Quote Post
bennetng
post Apr 16 2013, 17:32
Post #3





Group: Members
Posts: 224
Joined: 22-December 05
Member No.: 26587



Save the files in floating point format will solve your problem.
Go to the top of the page
+Quote Post
db1989
post Apr 16 2013, 17:48
Post #4





Group: Super Moderator
Posts: 5275
Joined: 23-June 06
Member No.: 32180



QUOTE (coolhotfun @ Apr 16 2013, 17:25) *
so i went thru them all again later and got a lower level of volume, where you feel that its a bit too low. I'm wondering if i go to raise up the volume later in the future will i lose some quality?
I echo what Propheticus asked.

QUOTE
Ran a test where i lowered it by -40db and it was tiny, then raised it back up by 40 and compared it to the original and seemed identical. Saved a -20db file and a -40db file and they were both the same file size. So i'm guessing it's keeping all that data even though the volume is lower.
This is not how uncompressed audio works at all. By definition, the files will be of the same size regardless of what they contain. You would find the same result with one file being silence.

QUOTE
But in the volume level tests i lowered it to the minimum, -inf db.. which seems to be a step past -59 so maybe -60, if i try and raise it up by +60 after, can't do it.. its mute like if it lost the data. 59 and back can work though.
Naturally, for –infinity is equal to silence.

QUOTE
I'm wondering if some of those files i'm working on am i losing some part of the sound spectrum by it being a lower volume or can i raise it up back later and it would be the exact same file as if i kept it louder.
You are not “losing some part of the sound spectrum” in the sense of any particular range of frequencies, or the infamous ‘soundstage’, or anything like that. You are adding noise to a degree that varies directly with the number of bits truncated by your attenuation in proportion to the number of bits used in the relevant mathematics. This noise remains when the resulting file is re-amplified.

Sure, using floating-point could alleviate all/most such concerns. But again, do you not have a volume control?
Go to the top of the page
+Quote Post
Kohlrabi
post Apr 16 2013, 17:56
Post #5





Group: Super Moderator
Posts: 1046
Joined: 12-March 05
From: Kiel, Germany
Member No.: 20561



QUOTE (coolhotfun @ Apr 16 2013, 17:25) *
ok i'm editing some sound samples and they're starting to get annoying hurting my ears after editing stuff for hours.. so i went thru them all again later and got a lower level of volume, where you feel that its a bit too low. I'm wondering if i go to raise up the volume later in the future will i lose some quality?
If you lower the volume in a Digital Audio Workstation and save the file you will probably lose dynamic range, by simple removing the quiet part (lowest bits) of the audio. I said probably because if you save the resulting file using a higher bitdepth than the original file, you may not lose any dynamic range. Each bit of bitdepth is about 6dB of dynamic range. So to not lose any audio data after lowering the volume by say 32dB, you'd need to increase about bit depth by at least 6 bit. If your files are consistently too loud, you might want to expand the dynamic range of the signal with an expander, or record with more dynamic range.

This post has been edited by Kohlrabi: Apr 16 2013, 17:57


--------------------
Ceterum censeo Masterdiskem esse delendam.
Go to the top of the page
+Quote Post
Arnold B. Kruege...
post Apr 16 2013, 18:56
Post #6





Group: Members
Posts: 3797
Joined: 29-October 08
From: USA, 48236
Member No.: 61311



QUOTE (coolhotfun @ Apr 16 2013, 12:25) *
ok i'm editing some sound samples and they're starting to get annoying hurting my ears after editing stuff for hours.. so i went thru them all again later and got a lower level of volume,


+1 to what Propheticus said - why not just turn down the volume control on your monitoring system?

QUOTE
I'm wondering if i go to raise up the volume later in the future will i lose some quality? Wasnt sure if there's a threshold where if you lower the volume enough it loses that data.


In any audio system there is a point where any further lowering of signal level will start pushing the signal below the noise floor. While you can hear music that is somewhat buried in noise, bury it deep enough and it is truly lost.

QUOTE
Ran a test where i lowered it by -40db and it was tiny, then raised it back up by 40 and compared it to the original and seemed identical. Saved a -20db file and a -40db file and they were both the same file size. So i'm guessing it's keeping all that data even though the volume is lower.


File size is not an accurate guide to sound quality.

QUOTE
But in the volume level tests i lowered it to the minimum, -inf db.. which seems to be a step past -59 so maybe -60, if i try and raise it up by +60 after, can't do it.. its mute like if it lost the data.


That would appear be a peculiarity in the techniology used to reduce the level. Apparently attenuation greater than a certain amount is interpreted as muting. A bit odd for an audio editor.

QUOTE
I'm wondering if some of those files i'm working on am i losing some part of the sound spectrum by it being a lower volume or can i raise it up back later and it would be the exact same file as if i kept it louder.


You may be confusing what you perceive with what actually happens technically speaking.

Because of the Fletcher Munson characteristic of your ears, reducing SPL at your ear decreases your perception of the loudness of sounds at both ends of the audio spectrum.

Any good volume control will attenuate all in-band frequencies the same.
Go to the top of the page
+Quote Post
greynol
post Apr 16 2013, 19:06
Post #7





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



QUOTE (Arnold B. Krueger @ Apr 16 2013, 10:56) *
Any good volume control will attenuate all in-band frequencies the same.

I'd much prefer one that is fitted to my equal loudness profile, personally.


--------------------
I should publish a list of forum idiots.
Go to the top of the page
+Quote Post
DVDdoug
post Apr 16 2013, 19:19
Post #8





Group: Members
Posts: 2588
Joined: 24-August 07
From: Silicon Valley
Member No.: 46454



You are loosing resolution (if you save in an integer format, such as 16-bit or 24-bit WAV). Even though you don't hear any quality loss, I'd say it's "bad practice" to do this as part of music/sound production. You wouldn't do that with your analog signal, would you? Reduce a line level signal by 20 or 40 dB, and then run it through a mic preamp to bring the level back up?

QUOTE
ok i'm editing some sound samples and they're starting to get annoying hurting my ears after editing stuff for hours..
Unless you have calibrated your system, there is no known relationship between the digital level (dBFS) and the acoustic sound volume level (dB SPL).

This article recommends an acoustic level of 85dB SPL:
QUOTE
Over 1000 convention attendees filled the theatre center section... “How many of you thought the sound was too loud?” About 4 hands were raised. “How many thought it was too soft?” No hands. “How many thought it was just right?” At least 996 audio engineers raised their hands.
This is an incredible testament to the effectiveness of the 85 dB at 0 VU standard...


You might want to get an SPL meter and adjust your (acoustic) monitor level. You don't have to monitor at 85dB full-time during all editing. But when you want to know what it "sounds like", it's good to have a standardized reference level.

This post has been edited by DVDdoug: Apr 16 2013, 19:32
Go to the top of the page
+Quote Post
saratoga
post Apr 16 2013, 19:38
Post #9





Group: Members
Posts: 4959
Joined: 2-September 02
Member No.: 3264



QUOTE (greynol @ Apr 16 2013, 13:06) *
QUOTE (Arnold B. Krueger @ Apr 16 2013, 10:56) *
Any good volume control will attenuate all in-band frequencies the same.

I'd much prefer one that is fitted to my equal loudness profile, personally.


Would this be worth implementing? It could be implemented.
Go to the top of the page
+Quote Post
dhromed
post Apr 16 2013, 19:56
Post #10





Group: Members
Posts: 1314
Joined: 16-February 08
From: NL
Member No.: 51347



It could alleviate pop-fatigue. But it might be a smiley curve EQ could do the trick just as effectively.
Go to the top of the page
+Quote Post
greynol
post Apr 16 2013, 20:01
Post #11





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



It's effectiveness would depend on how well you choose curves to best fit the general population and provide tweaks/presets to better fit any given individual; compensation for those with hearing loss and other gross outliers notwithstanding.


--------------------
I should publish a list of forum idiots.
Go to the top of the page
+Quote Post
greynol
post Apr 16 2013, 20:02
Post #12





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



QUOTE (dhromed @ Apr 16 2013, 11:56) *
It could alleviate pop-fatigue. But it might be a smiley curve EQ could do the trick just as effectively.

Perhaps, but the point is that the amount of smile would be inversely proportional to the output level.

PS: I can split this if people are interested. I was throwing it out there as pie in the sky (emphasis placed on fitted to MY profile). FWIW, I like the variable loudness feature in Yamaha 2-ch receivers, though I generally use it more as a third EQ control, rather than a second (or primary) volume control as is recommended in the manual.

This post has been edited by greynol: Apr 16 2013, 20:07


--------------------
I should publish a list of forum idiots.
Go to the top of the page
+Quote Post
MikeFord
post Apr 16 2013, 20:18
Post #13





Group: Members
Posts: 132
Joined: 22-May 04
Member No.: 14241



QUOTE (saratoga @ Apr 16 2013, 11:38) *
QUOTE (greynol @ Apr 16 2013, 13:06) *
QUOTE (Arnold B. Krueger @ Apr 16 2013, 10:56) *
Any good volume control will attenuate all in-band frequencies the same.

I'd much prefer one that is fitted to my equal loudness profile, personally.


Would this be worth implementing? It could be implemented.

Equal loudness is playback level, so you would need some idea of the actual playback volume and frequency response. I wonder if such a frequency based volume wouldn't confuse our brains which are accustomed to frequency neutral volume. I could see it for simple listening, but working with sound and attempting to make judgments on balance etc, mixing, it might not work well.

Regards attenuating a sample, generally not a good idea, but maybe very dependent on the nature of the sampled sound. If you attenuate a pure tone it isn't clear to me anything is lost, maybe change in noise level. Attenuating signal effectively raises the final noise level.
Go to the top of the page
+Quote Post
2Bdecided
post Apr 16 2013, 20:29
Post #14


ReplayGain developer


Group: Developer
Posts: 5135
Joined: 5-November 01
From: Yorkshire, UK
Member No.: 409



QUOTE (greynol @ Apr 16 2013, 19:06) *
QUOTE (Arnold B. Krueger @ Apr 16 2013, 10:56) *
Any good volume control will attenuate all in-band frequencies the same.

I'd much prefer one that is fitted to my equal loudness profile, personally.

Isn't there already an fb2k or vst plugin for this? I know I have it installed, but don't use it. Not at my PC to check what it's called. To my ears (used to normal volume controls) it sounds quite unnatural, but that might be just a lack of correct calibration to actual listening level.

OP: -40dB and back at 16 bits is enough to add easily audible noise or distortion to quiet sounds. It leaves only the dynamic range (background hiss) of cassettes. If you have only loud sounds you will not hear the problem.

Cheers,
David.

This post has been edited by 2Bdecided: Apr 16 2013, 20:34
Go to the top of the page
+Quote Post
greynol
post Apr 16 2013, 20:31
Post #15





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



QUOTE (MikeFord @ Apr 16 2013, 12:18) *
Equal loudness is playback level

No, equal loudness is equal loudness. While it is a function of playback level it is quite different from playback level.

@David's question that was slipped in during my response:
I have no idea. Calibration is definitely key, both to SPL and to the user's response.

This post has been edited by greynol: Apr 16 2013, 20:56


--------------------
I should publish a list of forum idiots.
Go to the top of the page
+Quote Post
bernhold
post Apr 16 2013, 20:34
Post #16





Group: Members
Posts: 41
Joined: 22-March 13
Member No.: 107330



QUOTE (bennetng @ Apr 16 2013, 18:32) *
Save the files in floating point format will solve your problem.


Could you elaborate on that? Can you actually save a WAV file as floating point or as integer? I never heard of that before, sounds interesting. To test it, I opened a WAV file in Audacity (audio editor), and it shows "32-bit float" in file info. However, when I normalize the audio to a very low volume and then turn it back up again, a lot of noise is added to the file. So it seems to be a "lossy" process even if it's saved in floating point format.

By the way, isn't floating point discarding accuracy as well? What about values like 5,3333333~?

This post has been edited by bernhold: Apr 16 2013, 20:38
Go to the top of the page
+Quote Post
bennetng
post Apr 16 2013, 20:50
Post #17





Group: Members
Posts: 224
Joined: 22-December 05
Member No.: 26587



QUOTE (bernhold @ Apr 17 2013, 03:34) *
QUOTE (bennetng @ Apr 16 2013, 18:32) *
Save the files in floating point format will solve your problem.


Could you elaborate on that?


I think this thread answered your question pretty good.
http://www.hydrogenaudio.org/forums/index....st&p=434237

And I think you did not correctly configured Audacity to use floating point format. I uploaded a video to show it.
http://youtu.be/E_D6d83nVUE

This post has been edited by bennetng: Apr 16 2013, 21:12
Go to the top of the page
+Quote Post
bernhold
post Apr 16 2013, 21:15
Post #18





Group: Members
Posts: 41
Joined: 22-March 13
Member No.: 107330



Thank you. Links useful.
Go to the top of the page
+Quote Post
coolhotfun
post Apr 16 2013, 22:27
Post #19





Group: Members
Posts: 25
Joined: 4-July 11
Member No.: 92047



QUOTE (Propheticus @ Apr 16 2013, 11:30) *
why don't you just lower your playback level instead of altering the waves amplitude?


then i won't hear the hiss for noise removal
Go to the top of the page
+Quote Post
Propheticus
post Apr 16 2013, 22:58
Post #20





Group: Members
Posts: 219
Joined: 10-September 11
Member No.: 93615



Still don't get it. Won't lowering the waveform by xxdB instead of changing playback volume xxdB result in the same problem? Either way the hiss will be softer... By attenuating your wave and then turning up the playback volume you could possibly even introduce/amplify noise produced by your own equipment.

This post has been edited by Propheticus: Apr 16 2013, 23:00
Go to the top of the page
+Quote Post
coolhotfun
post Apr 16 2013, 23:17
Post #21





Group: Members
Posts: 25
Joined: 4-July 11
Member No.: 92047



QUOTE (Propheticus @ Apr 16 2013, 16:58) *
Still don't get it. Won't lowering the waveform by xxdB instead of changing playback volume xxdB result in the same problem? Either way the hiss will be softer... By attenuating your wave and then turning up the playback volume you could possibly even introduce/amplify noise produced by your own equipment.


I'm not doing any of this on headphones.. its thru an amp and then these big tower speakers that have 3 woofers on them. If i lower my amp volume i dont hear the hiss as well. So even though the wav volume level is now lower i can still hear the hiss compared to if i turned down the amp instead. So gotta have it at this volume level.

I've got 1800 sound effect samples i'm going thru.. done 900. On the weekend thought of re-checking the first files i had done and way too loud so had to go thru the 900 again turning down all the volume levels. After that checked the start ones again and they're all close.

A couple months ago when i was making the 1800 there was a volume knob change. Those earlier files when i go thru them now, i hear more hiss that i didnt hear before and gotta re-do noise reduction on those older files. Not gonna mess with my volume knob at this point. lol.


ok now back to what i'm wondering.. if anybody knows how wav files work.. like those tests i ran where i took a normal sounding wav and lowered it by -40db then raised it again.. sounded the same. You'd think doing that would add some kind of hiss or something. But it didnt.. It seems like in a wav file anything to do with volume is just a number so even though you dont hear it, its still contained in the file. But apparently somebody up there mentioned that file size doesnt have to do with that so even if it was a -inf it'd still be the same size but the original material isnt there anymore. Thats what i'm wondering.. wheres the point where that data gets lost when i do -inf compared to -59db.. its getting lost somewhere. So wheres the point where stuff gets thrown out and if i go to raise the volume later are there things that'll be lost? You guys got all that? and dont use any big words like attenuation or floating point format cause i dont know what that means. I didnt go to college for all this, like you guys ya know.

This post has been edited by coolhotfun: Apr 16 2013, 23:19
Go to the top of the page
+Quote Post
coolhotfun
post Apr 17 2013, 01:10
Post #22





Group: Members
Posts: 25
Joined: 4-July 11
Member No.: 92047



oh, another thing i'm wondering about is when i'm lowering the volume of a file lets say its still too loud and i run it again to lower the volume another time after already doing it, is that bad? like do i have to lower it all down in one shot? Lets say you do -10db all at once, is that the same as doing -1db ten times? or would that one have more hiss or something. Cause sometimes as i go thru the files i'm working on a second time i'll go oh that's gotta go down 1 or 2db. Is this causing more noise or something? I guess i'm thinking about the old days with cassette where its so bad doing things a second time.
Go to the top of the page
+Quote Post
MikeFord
post Apr 17 2013, 01:26
Post #23





Group: Members
Posts: 132
Joined: 22-May 04
Member No.: 14241



If you want to listen for hiss, turn the overall volume down and boost the treble.

Having read all that you have posted I still don't know what your goal is, but I don't think its happening.
Go to the top of the page
+Quote Post
saratoga
post Apr 17 2013, 02:02
Post #24





Group: Members
Posts: 4959
Joined: 2-September 02
Member No.: 3264



QUOTE (coolhotfun @ Apr 16 2013, 17:17) *
A couple months ago when i was making the 1800 there was a volume knob change. Those earlier files when i go thru them now, i hear more hiss that i didnt hear before and gotta re-do noise reduction on those older files. Not gonna mess with my volume knob at this point. lol.


Can you explain this in more detail? I can't understand what it is that you're doing.

QUOTE (coolhotfun @ Apr 16 2013, 17:17) *
ok now back to what i'm wondering.. if anybody knows how wav files work.. like those tests i ran where i took a normal sounding wav and lowered it by -40db then raised it again.. sounded the same. You'd think doing that would add some kind of hiss or something. But it didnt..


It depends on what you did. If you did it to a 16 bit WAV file, saved it as WAV, and then read it back in, then yes it reduced the SNR of the file by 40dB (although if the SNR was already low it might not have mattered).

If you just lowered and raised it in your editor without saving it, the software hopefully used floating point precision and therefore introduced no additional noise.

QUOTE (coolhotfun @ Apr 16 2013, 17:17) *
It seems like in a wav file anything to do with volume is just a number so even though you dont hear it, its still contained in the file. But apparently somebody up there mentioned that file size doesnt have to do with that so even if it was a -inf it'd still be the same size but the original material isnt there anymore. Thats what i'm wondering.. wheres the point where that data gets lost when i do -inf compared to -59db.. its getting lost somewhere. So wheres the point where stuff gets thrown out and if i go to raise the volume later are there things that'll be lost? You guys got all that? and dont use any big words like attenuation or floating point format cause i dont know what that means. I didnt go to college for all this, like you guys ya know.


A 6 dB volume reduction throws away 1 bit. For 16 bit WAV you have 16 bits. Throw them all out and nothing is left. So -96dB reduction == -inf.

QUOTE (coolhotfun @ Apr 16 2013, 17:17) *
Cause sometimes as i go thru the files i'm working on a second time i'll go oh that's gotta go down 1 or 2db. Is this causing more noise or something?


Yes it is, although not a huge amount. If you're going to do multiple adjustments, save them to 24 bit WAV files so that you have more dynamic range available.
Go to the top of the page
+Quote Post
Glenn Gundlach
post Apr 17 2013, 05:55
Post #25





Group: Members
Posts: 369
Joined: 19-April 08
From: LA
Member No.: 52914



QUOTE (DVDdoug @ Apr 16 2013, 10:19) *
You are loosing resolution (if you save in an integer format, such as 16-bit or 24-bit WAV). Even though you don't hear any quality loss, I'd say it's "bad practice" to do this as part of music/sound production. You wouldn't do that with your analog signal, would you? Reduce a line level signal by 20 or 40 dB, and then run it through a mic preamp to bring the level back up?

[<snip>


Actually there is commercial equipment that does EXACTLY that. The Sony BetaCam SP decks from the '80s have a switch for each audio channel that adds input attenuation of a line level signal to the mic preamp. It is NOT a variable gain amp. They are not alone in this practice.

Something to remember - in an integer system every time you drop 6dB you're effectively discarding a bit and you will not 'get it back' when you boost the gain.

Go to the top of the page
+Quote Post

3 Pages V   1 2 3 >
Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 17th September 2014 - 15:38