Primary tabs

Same File? Does "A" Sounds Better? Or Does "B" sounds Better?

Same File? Does "A" Sounds Better? Or Does "B" sounds Better?
This is a test. Had it been a real emergency two files would have been posted here. Wait! They have been posted here!

These two files, "A" and "B", contain the same short snippet of the song "If I Had My Way" by the Reverend Gary Davis, sung by Peter, Paul & Mary on their 1962 eponymous debut album (Warner Brothers WS1449). It's a fantastic sounding recording by engineer Bill Schwartau that puts each singer, closely miked in his or her own space. Mary appears in the "phantom center channel".

These two files are either identical (same file repeated) or two different files with one variable changed for the second recording. Don't let the slightly different lengths throw you off. Each was edited to a different length but they still could be the identical recording. Or not!

So please download the 96/24 AIF files and listen carefully. Then vote "same" or if you hear differences, vote for the one that sounds better and feel free to explain your votes in the "comments" section.

Good luck! And remember: the future of high quality audio is now in your hands.

Here are the two files:

File "A"

File "B"

chris8519's picture

I looked at the Spectrograms before listening. The files do indeed appear to be different; there are different peaks and 'pops' from the vinyl.

Over all, it looks like Sample A has higher, quicker transient peaks. I'm sure it's a case of self-fulfilling prophecy, but I liked Sample A better when I listened. They are both very similar.

At least 'on paper', I liked A better. Maybe I'm a fool?


edit: damn, I love these listening tests!

Michael Fremer's picture

I'm not going to wait long before explaining what this is all about. But thanks for participating!

Jim Tavegia's picture

I thought B had lower opening surface noise and over all my favorite.  Other than that I thought they sound oh so similar. Odd that the accompaning vocal opposite the soloist is much higher in level, almost annoyingly so. 


terrybbagit's picture

I heard exactly the same things Jim did

Bob HiFi's picture

The first one did sound a bit harder over all. The vocals were more "gravely"?

The second one was smoother and a bit quieter alright but that might be an output thing.

Overall I can imagine the second being less fatiguing over a long period of listening but depending on the system I can see both being able to sound ok.

edit : I listened through my phone and a decent set of headphones

gMRfk6LMHn's picture

Having listened I find 'File A' a bit strident and Mary's voice is drowned out a bit, 'File B' is easier on my ears and Mary's voice is not as drowned out. I'll go with File B.

James, Dublin, Ireland.

figaro's picture

B seemed to have more space around the singers

Kurt's picture

I'm listeming through my Grados plugged into my Mac, and I don't know if that's a good enough playback system to hear any differece, but i felt like the guitar in A had just the right sound. In B, the guitar sounded dull. Overall B seemed duller. 

BTW, according to Roy Book Binder, Rev. Gary Davis called his house on Long Island "The house that Peter, Paul, and Mary built" because he paid for it with the royalties from their cover of his song. 

jimel84's picture

I prefer the sound of the " A" track better. To me it sounds more realistic. There seems to be more "air" around the instuments. I get the feeling that track "B" may have been cleaned up with software and in doing so it lost some realism along the way. Just my .02 cents worth.

markbrauer's picture

The files sound very much the same.

For comparison I also cued up the same cut ripped from CD and up-converted to 96/24 FLAC. It was WAY BETTER than either of the samples. The instruments and voices were fuller and more dynamic. The ambience was alive. The music is more exciting.

I even experimented with giving an advantage to the LP rips by playing them at a relatively higher volume. The digital version still won.

I had a similar experience with your earlier cartridge "shootout". While there were certainly bigger differences between those tracks, the same cut served up by Spotify (I believe my setup plays Spotify "bit perfect") improved the sound in almost every way. Although I did not buy it, experience tells me the high-def version would have been better yet.

These "tests" are fun. Keep up the good work.


Michael Fremer's picture

I find it somewhat astonishing that anything from Spotify, which is still low resolution even if "bit perfect" outperforms vinyl ripped at 96/24 but even CD in direct comparison to vinyl rarely sounds as good. I have no idea on what you were listening to make these judgements but at analogcorner we are open to all experiences and opinions.

markbrauer's picture


My system consists of a Windows 7 PC running Logitech Media Server streaming over WiFi to a Squeezebox Touch using it's internal DAC to feed a Purity Audio K.I.C.A.S. desktop headphone amp driving Sennheiser HD650 phones. I'm not up on all the tech but I'm thinking the server sends the Spotify Ogg Vorbis stream directly to the DAC. There is no Windows audio stuff in the data path.


I know the 650s do have a tendency to warm things up but the amp was reviewed as tending toward brightness and the two seem to work well together.


On this system Spotify generally sounds quite good. Ripped CDs of the same material always sound better. Hi-res files always better yet, and by a large margin. While not state-of-the-art by any means, the equipment has enough resolving power to easily discern the differences and appreciate the improvements. 


For example; a while back, I discovered the recent remaster of Clapton's Unplugged on Spotify. Comparing it to my old CD rip the differences are not subtle. The soundstage on the lo-res Spotify stream is more open with better instrument placement. The footstomping is much more distinct and the bass it produces is more impressive. The instruments ring out in a more lifelike way. This from a 320mbps stream. I did go out and get the remaster on CD and it is, as expected, even better than the Spotify version.


In the case of your Peter Paul & Mary samples, to me at least, Spotify just plains sounds better. The LP rips both have a muffled quality that masks the music. The digital stream is more open, dynamic, and fun to listen to. 


I repeated the test on my PC using the Spotify desktop app and J-River for the local files. Audio-Technica ATH-A700 headphones were connected to the audio out jack on the on-board sound card. I only use the PC to manage my library, so no audio optimization has been done. Overall sound quality on all tracks was much poorer but the Spotify stream and CD rip still head an edge over the LP rips.


Also... I would especially like to thank you for reminding me to revisit PP&M. This was the very first album I ever purchased with my own money (almost 50 years ago!) and the start of a lifelong music habit. I had not sat down and really listened to it in a long time. That's now rectified - and it's as delicious as I remember it. May have to replace my CD rip with a hi-res version.

atomlow's picture

Sounded much better to me. If it's the same file then I must like the letter B in the alphabet.

dhyman's picture

A is tipped up every so slightly.  I can see A's appeal. Over time, it would cause more listener fatique as it's voicing is a bit less natural than reality. B is long haul winner i'd want to marry.  Maybe date A for a night.



dhyman's picture

A sounds like a Lyra

B sounds like a Transfiguration

: )

JohnnyCanuck's picture

Keeping in mind that I'm playing the files back through the DAC in my Chromebook so this is just a guess but I think that B is softened somewhat by what appears to be some form of noise reduction.  A is a little noisier but I'm hearing a slightly more low level detail.

Chromebook feeding V-CAN headphone amp and AKG-702 headphones.

samman's picture

A does not have the "weight" of B. Everything is smoother with B, in my opinion.

samman's picture

Clear up the mystery. Explain A and B

Michael Fremer's picture

Maybe tomorrow.... or the next day (Friday)

AnalogJ's picture

Okay, so my desktop doesn't have the highest fidelity audio chip nor the highest fidelity speakers (they're Monsoon planar speakers with a subwoofer).

That being said, 'B' sounds a bit rolled off and damped to my ears. There is a bit more top end on 'A' (and for what other possible reasons), rendering a bit more clarity and distinction between the vocals. 'A' also seems to, for the want of a better term, swing more. The dynamic syncopation seems to jump more with 'A'. I prefer 'A'.

Miner42's picture

Is this

Michael Fremer's picture


jjgr's picture

B sounds "tinnier" or thinner, with less depth, less of a sense of real people singing and seemed to have slightly more surface noice - overall a more homogenous sound (not in a good way).  A sounded more real to me.  Thanks for posting this.  Cheers!  

jmset315's picture

A sounded clear and rather 'sterile' - like digital sould

B sounded full and more natural to me

Rick Tomaszewicz's picture

As promised, I tried to listen through the gear to the performance.  

"A" sounded compressed, sibilant and with plenty of surface noise.  A kind of "artificial" detail.  IOW, a hi-fi sound which sort of put everything on the same level - initially exciting, then boring and finally irritating.

"B" sounded cleaner and more relaxed, but at the same time had greater dynamic range and moment to moment differentiation among singers and instruments - more like a live performance.  It kept surprising me as I listened.

Smafdy Assmilk's picture

They sound very similar, if not exactly identical. They are two different transfers using the same cartridge and table.

*If* there is any difference, it would have to be something like a cable swap or waving a magic wand over the record before the next play. You know, something audiophiles can really dig without doing actual harm to the music.

Smafdy Assmilk's picture

BTW, this is a pretty bad recording, both sound quality and mix. Why is it so muted at the beginning and why are the musical balances so off?

StonedBeatles1's picture

I think they're identical and if not very close that my ears cannot tell any difference whatsoever.  In any event I think we all need to get lives!  :)

Analogplanet is fun..

Smafdy Assmilk's picture


Smafdy Assmilk's picture


*If* there is a diiference, Mr. Fremer will say he used a power conditioner or put his cables on a block of wood or something like that and everyone will say they knew they could hear it.


For now, look at the results: 40% and 39%. Statisitcally, that's guessing. If there *is* a difference, it's sure not noticeable.