1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.

    Dismiss Notice

FLAC vs. 320 Mp3

Discussion in 'Sound Science' started by icedup, Sep 7, 2011.
6 7 8 9 10 11 12 13 14 15
17 18 19 20 21 22 23 24 25 26
  1. Brooko Contributor
    I think you need to go back to your original post - where you said you could tell the difference between 24-96 flac and lossy.  What we've established since is that you were likely comparing different masters.
    As to why you should re-encode ...... you shouldn't unless it's for portable use.  The great thing about re-encoding if you are putting music on a portable device is that once you've discovered your own limits, you then no longer have to make a painstaking choice between lossless (at huge size) and lossy.  If you can't tell the difference - then go comfortably with the smaller size.
    For the record, I still buy tracks from HDTracks (among other sites).  As long as the price is right, and the master is acceptable to me, then I regard them as a great source.  For home listening, I listen to the original flac (in whatever resolution I purchased at).  I never buy greater than 24/96 though, and for my iPhone - I always re-encode at 256aac.
  2. bigshot
    The thing about the Beatles catalog is that they always started with the original session mixdowns. They can't change a lot when they do that. The primary difference between the original CD releases and the new stereo re-releases is that on the most recent re-release, they slightly compressed the dynamics. Other than that, they are pretty much identical and fine. The exception is Rubber Soul, which was remixed for stereo and Let It Be (Naked) which was completely remixed. For Rubber Soul, the mono version is better and for Let It Be, the bootleg Peter Sellers acetate is the best. In general, I prefer the mono box up to and including Sgt Pepper and stereo from Sgt Pepper on.
  3. Iamnothim
    If you want to attribute the difference in "what I hear" to different masters, fine.  I never knew that different masters were used for iTunes recordings vs HD Tracks recordings.  I'll take your word for it and use that as a good explanation why "I prefer" the sound of my lossless recordings.   Boutique production of digital music.
    It explains a lot.
    iTunes is like McDonalds... billions served... versus flac and your local steakhouse.  You don't have time for perfection, recording to recording, when you are turning out 10's of thousands of recording.  Speed and Volume rule.    iTunes is good for music equipment purchased from aisle 9 at Fry's.
    If it didn't matter, or was undetectable, why spend the dough on great equipment?  Why would venders publish frequency specs of 5Hz to 50kHz and mfg products with those capabilities?  Why would a sound engineer use the stuff?  No one can hear the difference right?  It's 50kHz, all ya need is 10kHz.  Based upon "people" taking a blind A/B test.
    I agree with your last paragraph.  It makes perfect sense.  That's what you should listen to at home. iPhone, absolutely 256aac !
     I'm only talking about home audiophile listening.  Nice speakers, nice headphones.  That's why we are here.  The members here are ...  a guess...  0.01% of digital music listeners.  As in way, way, to the right of the curve. I rarely listen to portable music.  When I do it's on stock buds.  I do like the new Apple5 buds.  That's Rarely.
    I see no value whatsoever in going through the gymnastics necessary to set up an adhoc  lab that is capable of  balancing source levels etc. to personally compare one format against another.  Why?  Perhaps I'll conclude that I still like "lossy".   It has no practical use in my decision process.
    I would, however, be interested in reading the results of testing performed by a qualified entity. Where is the research?
    I have learned through this discourse.  I did state that I thought the sound quality I detected was due to the format.  In the end I just like  how it sounds.

  4. Iamnothim
    Wow !
    Great stuff. Thanks.
    Now I have to read up so I can understand all, strike that, some of it.
  5. bigshot
    By the way, I have a kickass speaker system that I've built over 40 years. It's pretty darn good. It's all served from a Mac Mini loaded with 256 AAC VBR files. I keep my CDs in the garage. That's how good AAC sounds.
  6. Iamnothim
    I never for a moment thought you would have a mundane system.  After all you're in the 0.01%
  7. atistatic
    Well IMHO, with my equipment and piano experience the unique 2 difference can i hear in 320 and FLAC is one, the force of push the piano keyboards of the pianist player and piano vibrato and trill connection. the vibrato and push force you can hear that clearly when compare especially with Lang Lang.
  8. Brooko Contributor

    You realise you're in the sound-science section right? Use a single source file, transcode the flac one to 320, use Foobar2000's abx comparator plugin (volume matched using replay gain), compare the 2 files for 15 iterations, and post the log.
  9. bigshot
    All I hear when I play Lang Lang is an awful piano player.
  10. Iamnothim

    Show me the science.
  11. xnor
    What do you want to know? How to calculate ABX test results? Just look up a binomial cdf table.
    If that doesn't satisfy you:
    - Srednicki (1988) "A Bayesian Analysis of A-B Listening Tests" , J. AES. Vol. 36 No. 3
    - Burstein (1989) "Transformed Binomial Confidence Limits for Listening Tests",  J. AES. Vol. 37 No. 5
    ... there are many more papers. [​IMG]
    Brooko likes this.
  12. sonitus mirus
    Normally I've been ripping a CD with Exact Audio Copy to flac and ripping it again to Lame V0 320 kbps and then abx'ing the 2 files.  I can't tell a difference between the files with any regularity.  I've never used replay gain at all in my testing.  Does this mean to check the box for replay gain from the abx component in Foobar, or is there something more that I need to do? 
    I'd like to be able to test some HD tracks at 24-bit 96kHz converted to mp3's at 16-bit 44.1 kHz 320kbps.  This would really make my day to be able to test this and prove that I can't tell any difference between the 2 files, but I want to make sure I am creating the files correctly using the same volume level.
    Can you hand walk me through the scenario to get this done?  Really I think I just need clarification on the part about using Replay Gain.  I've watched videos and have read a ton of forum comments and even read what little documentation is available, but I'm only ever presented with overall instructions without any details.  I wouldn't be surprised if this is why many people don't post their own logs, because the instructions are a bit vague for many of us.  
  13. xnor
    Get foobar2000, choose full on installation, add your 24/96 file, right click - convert - ...
    Choose output: Add new
    encoder: lame.exe (download it from rarewares.org)
    extension: mp3
    parameters: -S --noreplaygain -b 320 - %d
    format is: lossy
    highest bps supported: 32
    Choose processing: Add Resampler (SoX)
    (download it form here and install it using File - Preference - Components - Install ...)
    Configure the resampler for 44.1 kHz
    Convert the file.
    Now select the 24/96 file and mp3 and right click - ReplayGain - Scan per track gain.
    sonitus mirus and Brooko like this.
  14. sonitus mirus
    Thanks, your instructions will make it much easier for me to have confidence that I am performing the test accurately.
    Will try this first thing tonight when I get home and post the logs of my results.
  15. Iamnothim
6 7 8 9 10 11 12 13 14 15
17 18 19 20 21 22 23 24 25 26

Share This Page