#1
|
|||
|
|||
Wav > mp3
Lot's of discussion/debate surrounding these two aspects of digitally recorded sound. When I first read of it some years back I wasn't recording so could not relate.
I've been recording for a few years and now understand the sonic losses rendering to .mp3 from .wav files. When I'm in the DAW's .wav format and working a recording to a what I think sounds good, it changes once it goes to .mp3. There's certainly a loss of depth and sparkle between the two formats. Lately, once rendered to .mp3 and listening for those subsequent losses, I'll go back to the DAW and make an attempt to edit the .wav file to see if I can recapture them in the .mp3. I have not rendered two basic .mp3 comparison tracks yet to see if I'm making the right changes. Anyway, just thought I'd make mention of it. |
#2
|
|||
|
|||
Pay attention to the bitrate of the mp3 file.
Never mix down as low as 120kbps. You'll be able to hear the loss of frequencies. I use 320kbps and can't hear the difference at all.
__________________
All tracks on SoundCloud is recorded and mixed in Auria on the ipad using an Alesis IO Dock as sound interface. My tracks on SoundCloud (Lønhart Duo): YouTube page: https://www.youtube.com/channel/UCZT...HIR4tGnwhusf8w Taylor 324 Martin DX1AE Gretch G5420T Electromatic Aria Pro II Cardinal Series 250 |
#3
|
||||
|
||||
As TallDrink said the bit rate makes a noticeable difference .
Quick crude MP3 analogy : When converting digital audio to MP3 it is like taking very thin slices out of a whole pie, the result is some of the content is missing. The lower the bitrate number, the larger the size of the slices removed. Or removing pixels from a digital photo is often given as an analogy, the lower the bitrate, the more pixels are removed When I was taking an online course at Berklee College of Music we did a fairly extensive blind test to try to determine average listening threshold of detecting a difference in MP3 and WAV file formats . What the class test results showed was almost everybody was able to detect a difference with anything below 256 then about half the class tested being able to detect anything below 320. And two people out of class of 17 consistently determined the difference between 320 MP3 and WAV files . The difference in the personal results was speculated to be due to one or a combination of, three things 1. personal physical hearing differences 2. critical listening ability differences 3. home studio system reproduction quality differences . Long story short, if you record at 320 you will most likely not notice any degradation.
__________________
Enjoy the Journey.... Kev... KevWind at Soundcloud KevWind at YouYube https://www.youtube.com/playlist?lis...EZxkPKyieOTgRD System : Studio system Avid Carbon interface , PT Ultimate 2023.12 -Mid 2020 iMac 27" 3.8GHz 8-core i7 10th Gen ,, Ventura 13.2.1 Mobile MBP M1 Pro , PT Ultimate 2023.12 Ventura 12.2.1 Last edited by KevWind; 08-08-2017 at 06:53 AM. |
#4
|
|||
|
|||
Keep in mind that most average consumers today listen to 256k mp3 via iTunes Apple Music or other streaming media sources on portable devices. Many with very large libraries auto-convert to 128k to make as much storage room as possible anyway. It's unfortunate, but understandable that it's now about portability and convenience rather than quality. IMO 256k is actually rather good. You'll never tell the difference listening on a pair of iPhone headphones.
__________________
|
#5
|
|||
|
|||
The problem here is that the more recording and listening experience you get, the more stuff you hear. With enough experience, you'll absolutely hear the difference between a WAV and the highest-res mp3, even on earbuds. Between a 320k mp3 and a 256k one? Maybe not.
|
#6
|
||||
|
||||
I use 320 constant bit rate (Lame conversion). With that I can't say I hear a difference in the acoustic guitar between the mp3 and the wav (except just perhaps in some reverb tails).
If you are going to compare the wav to the mp3 do so with both side by side in the same DAW set up (the original wav and the mp3 imported back into the DAW). See if you hear a difference there.
__________________
Derek Coombs Youtube -> Website -> Music -> Tabs Guitars by Mark Blanchard, Albert&Mueller, Paul Woolson, Collings, Composite Acoustics, and Derek Coombs "Reality is that which when you stop believing in it, doesn't go away." Woods hands pick by eye and ear
Made to one with pride and love To be that we hold so dear A voice from heavens above |
#7
|
|||
|
|||
Thanks for the explanation / input, everyone. It'll shortcut my work. Now I have to put it to use.
|
#8
|
|||
|
|||
Theres just no way to not have a difference going from a .wav (uncompressed PCM file) to a lossy-compression format like MP3. Accept it as a reality...a casualty of convenience, if you will.
What I have found is that I can maximize the final quality of the mp3 with a few tweaks. First: I find the higher the bit & sample rate of the original .wav file the better the final results. I typically start with 32-bit 96k files. Be super aware that MP3 compression will reveal itself a lot in the mids. I tend to scoop a little at 300Hz to lighten the low mids. This usually yields a better sounding mp3. Help out the conversion by using high-pass & low-pass filters just "inside" the Nyquist limit. The less you let the mp3 conversion work in it's extreme ranges the better off you are. In a practical sense that means I high-pass somewhere around 25Hz and low-pass somewhere around 18k. Even at that if I pull up the original .wav file and the converted .mp3 in SpectraFoo, you can see that there are significant differences in the frequency response. So, it's not just voodoo, I really am hearing a difference. The one thing I can never quite fix is how the mp3 loses space and dimension. The soundstage collapses and the "air" around instruments disappears. As I said, it's just a casualty of convenience. I accept it as a reality. This is why the publishers I work with only accept .wav files for final delivery. HTH
__________________
-Steve 1927 Martin 00-21 1986 Fender Strat 1987 Ibanez RG560 1988 Fender Fretless J Bass 1991 Washburn HB-35s 1995 Taylor 812ce 1996 Taylor 510c (custom) 1996 Taylor 422-R (Limited Edition) 1997 Taylor 810-WMB (Limited Edition) 1998 Taylor 912c (Custom) 2019 Fender Tele |
#9
|
|||
|
|||
shouldn't that be "high cut ... low cut" or am i confused?
|
#10
|
||||
|
||||
Basically the same thing (different terminology) so not confused, just understand that a high pass is a low cut (and vice versa) . A high pass lets all the frequencies above the selection point "pass" through. Low pass lets everything below the selection "pass" through
__________________
Enjoy the Journey.... Kev... KevWind at Soundcloud KevWind at YouYube https://www.youtube.com/playlist?lis...EZxkPKyieOTgRD System : Studio system Avid Carbon interface , PT Ultimate 2023.12 -Mid 2020 iMac 27" 3.8GHz 8-core i7 10th Gen ,, Ventura 13.2.1 Mobile MBP M1 Pro , PT Ultimate 2023.12 Ventura 12.2.1 |
#11
|
||||
|
||||
Quote:
The bold is a very good an explanation of the difference, with the caveat that the better the reproduction listening system and the more developed th critical listening skill, the more detectable the degradation in (3D-ish) effect. One other thing I have noticed is sometimes the balance of frequencies and or the relationship and sound of any applied FX can change slightly also. So I usually recheck on both my studio system and my laptop any converted file befor publishing to the internet.
__________________
Enjoy the Journey.... Kev... KevWind at Soundcloud KevWind at YouYube https://www.youtube.com/playlist?lis...EZxkPKyieOTgRD System : Studio system Avid Carbon interface , PT Ultimate 2023.12 -Mid 2020 iMac 27" 3.8GHz 8-core i7 10th Gen ,, Ventura 13.2.1 Mobile MBP M1 Pro , PT Ultimate 2023.12 Ventura 12.2.1 Last edited by KevWind; 08-09-2017 at 06:28 AM. |
#12
|
|||
|
|||
The standard and accepted audio jargon uses "pass," not "cut." Yes, it's confusing. But this got decided a long time ago, probably by a committee, and we're stuck with it.
So "high-pass filter" means that any frequencies higher than the filter's cutoff point are allowed to pass through unscathed. Last edited by Brent Hahn; 08-09-2017 at 08:50 AM. |
#13
|
|||
|
|||
hi-cut/low-cut is a shelving filter
a bandpass filter is different, what a bandpass filter does is allow all frequencies higher than (a high pass filter) or lower than (a low pass filter) to go through. Think of a shelving filter as a hill while the bandpass is a wall HTH
__________________
-Steve 1927 Martin 00-21 1986 Fender Strat 1987 Ibanez RG560 1988 Fender Fretless J Bass 1991 Washburn HB-35s 1995 Taylor 812ce 1996 Taylor 510c (custom) 1996 Taylor 422-R (Limited Edition) 1997 Taylor 810-WMB (Limited Edition) 1998 Taylor 912c (Custom) 2019 Fender Tele |
#14
|
|||
|
|||
thank you all for the replies. i keep learning ...
i asked about it because i thought DupleMeter is cutting off the frequencies below and above certain values, not letting them pass. so the naming was somehow confusing for me. |
#15
|
|||
|
|||
Quote:
__________________
-Steve 1927 Martin 00-21 1986 Fender Strat 1987 Ibanez RG560 1988 Fender Fretless J Bass 1991 Washburn HB-35s 1995 Taylor 812ce 1996 Taylor 510c (custom) 1996 Taylor 422-R (Limited Edition) 1997 Taylor 810-WMB (Limited Edition) 1998 Taylor 912c (Custom) 2019 Fender Tele |