Do Apple Lossless files really sound the same as AIFF?

[Mike Gillespie]

An audiophile friend compared a bunch of AIFF files with the same tracks copied to Apple Lossless and thought that the Lossless files didn't sound as good.

 

Has any one else had that reaction?

 

Is there any objective analysis (eg, a blind listening test) in the audio literature that shows any difference in listeners?

 

Obviously, I'd save a lot of space if I didn't have to use many terabytes for AIFF files.

 

Thanks.

[sandyk]

I only have experience with .flac vs. wav, but Lossless files in general, don't sound as good as the original version when played "on the fly." Convert them back to their original format before playing, and through a resolving system they will sound a little better.The Absolute Sound 220 and 221 did quite a bit of research in the area of .flac vs. wav etc.. although the usual suspects refuse to accept what they reported, or most people report about subjective findings in general, and that includes differences between USB cables too.

 

Alex

 

 

 

 

 

 

 

 

"If you can't hear the difference between an original CD and a copy of your CD, you might as well give up your career as a tester. The difference between a reconstituted FLAC and full size WAV is much less than that, but it does exist."-Cookie Marenco. cookiemarenco.com/

[wgscott]

I've never been able to detect a difference, and I can convert from alac to wav and aiff to wav and the two resulting wav files have the same md5sum, and I can convert between any two other lossless formats and then to wav and always recover the same md5sum. Thus the encoded information is identical. The only difference therefore would reside in the file container. A player like Audirvana removes the container before playback, so it should not matter what container the identical data is stored in.

 

A few others here do report an audible difference. I haven't been able to reproduce their experience.

 

I store all my lossless music as alac for convenience.

[Jb26]

I've tried A/Bing with a AIFF and ALAC, and couldn't hear a difference in my system.

 

I was prepared to AIFF anyway, but ALAC has better tagging options through mp3tag that AIFF doesn't.

[dmagnus1]

I was researching the new Simaudio Moon Mind streamer, which seems like a nice product, and ran across this in the FAQ section. Since I have ripped everything to AIFF, this gave me some pause:

 

Q: Will the MiND play AIFF files?

A: AIFF format is not straighforward and we don't recommend using it. There are different formats for AIFF files - compressed and uncompressed - but the file-name extension alone won't allow you to differentiate them. Simaudio is currently working to make the MiND compatible with the AIFF file format that iTunes creates (AIFF-C/sowt). It will be available in a future MiND App update during Q1 in 2013. Converting files from AIFF to ALAC is good way to create a new "cleaner" format. Converting a lossless file from one format to another will be transparent and should not create any sound difference at all since the MiND streamer will output exactly the same PCM data to the DAC as you heard. MiND will not support any other type of AIFF file.

 

I ripped everything to AIFF, reasoning that, 1) memory is cheap - the need for compression is minimized, and lossless only saves you about 50% anyway; 2) converting a compressed file to a native bitstream has to involve more processing, hence more opportunities for error; and 3) increased processing must generate more electrical noise, which can't be a good thing. Its splitting hairs, admittedly, but those were my reasons.

 

Now I'm confused. Does anyone know if there really is any advantage of ALAC over AIFF other than file size? I believe Chris recommended ripping to a noncompressed form for playback, and a lossless compressed form for backup - although I believe he was using WAV and FLAC. Maybe I better rethink this before ripping the rest? I'm probably about a quarter of the way through a couple thousand CDs...

[souptin]

Q: Will the MiND play AIFF files?

A: AIFF format is not straighforward and we don't recommend using it. There are different formats for AIFF files - compressed and uncompressed - but the file-name extension alone won't allow you to differentiate them. Simaudio is currently working to make the MiND compatible with the AIFF file format that iTunes creates (AIFF-C/sowt). It will be available in a future MiND App update during Q1 in 2013. Converting files from AIFF to ALAC is good way to create a new "cleaner" format. Converting a lossless file from one format to another will be transparent and should not create any sound difference at all since the MiND streamer will output exactly the same PCM data to the DAC as you heard. MiND will not support any other type of AIFF file.

 

FWIW wav files are similar to aiff in that both are container formats, and can potentially contain something other than uncompressed PCM data. However I doubt you'll find many such files being offered as music downloads - figure it's more likely for them to be a result of someone clicking the wrong option in a conversion utility. Max, for example, is quite commonly recommended for file conversion, and it has an extensive / intimidating variety of encoder settings for both aiff and wav.

 

Anyway to get back to the point, it strikes me that Simaudio are making an elaborate excuse, but to be fair to them, they say they're working on an update. Call me a grumpy old cynic, but somehow I suspect the line about "AIFF format is not straightforward" will quietly disappear once the update is out

 

Converting a whole bunch of files from one lossless format to another is much quicker than re-ripping, and (given that you've been careful to check the conversion setup) carries no quality penalty no matter how many times you do it. So personally I don't see the need for you to change unless you're definitely going to buy the Simaudio (or some other streamer that has file format limitations).

[Richard Dale]

I've tried A/Bing with a AIFF and ALAC, and couldn't hear a difference in my system.

 

I was prepared to AIFF anyway, but ALAC has better tagging options through mp3tag that AIFF doesn't.

 

I'm interested in your comment about mp3tag tagging options - what does it allow you to do with ALAC that you can't do with AIFF?

 

I rip from CDs to AIFF with XLD into iTunes onto a 2Tb La Cie disk, and then I convert to Apple Lossless with XLD onto a 1Tb Toshiba portable USB drive. I have identical La Cie and Toshiba drives as backups and use rdiff-backup to keep the mirrored copies. I think disk space is cheap enough to make it worthwhile to maintain your collection in both AIFF and Apple Lossless. I prefer to buy downloaded music in FLAC format.

 

In my case if I was to compare the sound ALAC to AIFF, I would really be comparing the sound of a portable USB drive against a Firewire desktop drive more than the storage format I think. My collection will fit onto a 1Tb drive as ALAC, but not as AIFF. As far as I can tell nobody has said that Apple Lossless sounds better than AIFF - people either say they sound the same to them or they prefer AIFF.

[Paul R]

There's an old saying, "if you cannot dazzle them with your brilliance, baffle them with your bullshit" - and that is precisely what you see with the Q&A piece from Moon you quoted.

 

Let me translate that answer from Moon for you - "We did not take the time to find out the facts and make our product work right, so we are blaming that on a well known and long used file format while we fix our product."

 

Given all that, I do what you do, I RIP to AIFF, and convert to any format I want to use on a particular system. JRMC at one time had problems with playing back ALAC files, so I tend to provide JRMC with an AIFF library. iTunes based libraries, even using Amarra and Audirvana, tend to work very well with ALAC, so I use that format for those libraries. FLAC works great for some systems, but I tend to avoid it if the system will use AIFF or ALAC, just because both AIFF and ALAC are cross platform compatible, Apple, Windows, and Linux, and FLAC is not really supported in iTunes based libraries.

 

When it comes to sound, sometimes I *think* I can hear a difference between ALAC and AIFF, or ALAC and WAV, or FLAC and AIFF, but it is never something I can reliably replicate. AIFF and ALAC sound essentially the same to me, so I interchange them as I wish without worrying about it. Indeed, I RIPped a few disks last night, and accidentially ripped then as ALAC. I will eventually get around to converting them to AIFF for the archive library, but it isn't high on my priority list of things "to do."

 

Hope that helps a bit.

-Paul

 

 

 

I was researching the new Simaudio Moon Mind streamer, which seems like a nice product, and ran across this in the FAQ section. Since I have ripped everything to AIFF, this gave me some pause:

 

Q: Will the MiND play AIFF files?

A: AIFF format is not straighforward and we don't recommend using it. There are different formats for AIFF files - compressed and uncompressed - but the file-name extension alone won't allow you to differentiate them. Simaudio is currently working to make the MiND compatible with the AIFF file format that iTunes creates (AIFF-C/sowt). It will be available in a future MiND App update during Q1 in 2013. Converting files from AIFF to ALAC is good way to create a new "cleaner" format. Converting a lossless file from one format to another will be transparent and should not create any sound difference at all since the MiND streamer will output exactly the same PCM data to the DAC as you heard. MiND will not support any other type of AIFF file.

 

I ripped everything to AIFF, reasoning that, 1) memory is cheap - the need for compression is minimized, and lossless only saves you about 50% anyway; 2) converting a compressed file to a native bitstream has to involve more processing, hence more opportunities for error; and 3) increased processing must generate more electrical noise, which can't be a good thing. Its splitting hairs, admittedly, but those were my reasons.

 

Now I'm confused. Does anyone know if there really is any advantage of ALAC over AIFF other than file size? I believe Chris recommended ripping to a noncompressed form for playback, and a lossless compressed form for backup - although I believe he was using WAV and FLAC. Maybe I better rethink this before ripping the rest? I'm probably about a quarter of the way through a couple thousand CDs...

[Superdad]

Okay, I'll throw in my two cents here:

 

I have a highly revealing system/room and little things are audible big time (tiny tweaks of A+ filter settings; a couple of millimeters angle of speakers, CD masterings, etc.).

 

Back when I was stuck using a 1.5GHz G4 Mac mini w/1GB RAM and Leopard, it was very easy to tell the difference between AIFF rips and Apple Lossless, but only if you listened to one version then the other. Soundstage was flatter, female voices did not have the same air, emotion, or reality, and my friend used to comment how all the ALAC files in our library had a "sameness" to their presentation. We even demonstrated this to a few select people in our (Hovland Co.) room at CES '08. I know I did the comparison--using a female vocal--for John Atkinson, Robert Harley, and if I recall correctly, CA's own Chris Connaker.

 

I always had a nagging suspicion (and others here suggested as much) that the differences heard were due to the real-time decompression processing taxing the G4.

 

So last year, when I finally got a new Mac mini (2010 2.4GHz Core 2 Duo w/8GB RAM, slimmed Mountain Lion, A+, etc.), I did the comparison again. Well, darn, as much as I hate to admit it (I went into it biased to hear a difference) I now can NOT tell any appreciable difference between an AIFF track and the same one converted to ALAC.

Again, I am sensitive to tiny differences in my system (lifting the 5V off the USB cable at the computer end, or even smaller, quitting Finder), so I was pretty surprised, given what we consistently heard back when our computers were under-powered and under-RAMed.

 

All that said, I still rip everything to AIFF. And, using iTunes/A+, I pre-convert FLACs to AIFF before adding them to my library.

Hope this info is helpful to the OP.

 

-ALEX

[Musicophile]

 

So last year, when I finally got a new Mac mini (2010 2.4GHz Core 2 Duo w/8GB RAM, slimmed Mountain Lion, A+, etc.), I did the comparison again. Well, darn, as much as I hate to admit it (I went into it biased to hear a difference) I now can NOT tell any appreciable difference between an AIFF track and the same one converted to ALAC.

That is not really surprising given that A+ preloads (and thus preconverts) the file prior to playback. This has always been my opinion. I somehow get why the conversion on the fly of an ALAC file may have a sound impact, especially on a an older computer with less horse power.

 

I never got how you're supposed to have any difference if you're using any of the players that play from memory, which should be the case for 98% of the users on this site.

[Superdad]

That is not really surprising given that A+ preloads (and thus preconverts) the file prior to playback. This has always been my opinion. I somehow get why the conversion on the fly of an ALAC file may have a sound impact, especially on a an older computer with less horse power.

 

I never got how you're supposed to have any difference if you're using any of the players that play from memory, which should be the case for 98% of the users on this site.

 

Good point. I will repeat the test this weekend without A+ (don't see a way in A+ to just disable memory preloading and the slider for allocation does not go down to zero).

[Mike Gillespie]

Good point. I will repeat the test this weekend without A+ (don't see a way in A+ to just disable memory preloading and the slider for allocation does not go down to zero).

 

Anyone, What is A+? Sorry, I just have never heard of it. Thanks.

[Musicophile]

Anyone, What is A+? Sorry, I just have never heard of it. Thanks.

Audirvana+, a software player for Mac.

[Jud]

Good point. I will repeat the test this weekend without A+ (don't see a way in A+ to just disable memory preloading and the slider for allocation does not go down to zero).

 

Maybe your system's too good. ;-)

 

A little odd, but the experience where I most noticed the difference between ALAC and AIFF with A+ was when I was traveling with my MacBook Pro and AQ Dragonfly, and hooked up the Dragonfly to our rental car radio with a $17 Monster miniplug-to-miniplug cable. I had A+ on shuffle. The first few songs sounded really nice (a little surprisingly). Then a favorite Ryan Bingham song came up, and it didn't sound quite as good, more like I would have expected from a rental car radio. I hit a key to bring back the display on the MBP, to try to see what was responsible for the change (occasionally iTunes has brought over tracks as lossy AAC when I've moved them to the MBP for travel, and I thought this might be such an instance). Checking that track and the past few that had been played, the difference was that the Bingham track was ALAC and the others were AIFF. Doubly odd, because not only was this with a much less resolving system, but it was also kind of an unintentional blind test, and I'd never noticed that much difference in non-blinded listening at home.

[sone]

Comparing the quality of various codecs is a test that anyone with a CD burner and a decent transport can perform for themselves. I'd recommend to listen to the various options and choose for yourself.

 

Personally, I chose the tradeoff of using AAC at 256kbps, which produces files that are one-third to one-half the size of ALAC lossless compression. I arrived at this decision by performing my own listening tests on CDs I created by transcoding from WAV->AAC->WAV at various bitrates and confirming that 256 kbps is indeed excellent. On several very nice sounding systems, my fellow enthusiasts and I had difficulty distinguishing the re-transcoded 256 kbps AAC tracks from their originals, though this test is admittedly a little different from decoding AAC on-the-fly.

 

Here's how to master a CD that contains all the various options. Listen to the CD on a good transport, and compare to the playback software on your computer, and go with the tradeoff that you're happiest with.

 

Use cdparanoia to rip a wav file, then run these Terminal commands:

 

[size=1][font=courier new]afconvert --file m4af --data aac --bitrate 256000 --quality 127 --src-quality 127 --verbose --strategy 2 track_orig.wav track_256kbps.m4a
afconvert --file WAVE --data LEI16@44100 --verbose track_256kbps.m4a track_256kbps.wav 
afconvert --file m4af --data aac --bitrate 128000 --quality 127 --src-quality 127 --verbose --strategy 2 track_orig.wav track_128kbps.m4a
afconvert --file WAVE --data LEI16@44100 --verbose track_128kbps.m4a track_128kbps.wav 
[/font][/size][size=1][font=courier new]afconvert --file m4af --data aac --bitrate 64000 --quality 127 --src-quality 127 --verbose --strategy 2 track_orig.wav track_64kbps.m4a
afconvert --file WAVE --data LEI16@44100 --verbose track_64kbps.m4a track_128kbps.wav 
[/font][/size][size=1][font=courier new]afconvert --file m4af --data alac --quality 127 --src-quality 127 --verbose track_orig.wav track_alac.m4a
afconvert --file WAVE --data LEI16@44100 --verbose track_alac.m4a track_alac.wav [/font][/size]

 

This gives the wav track_orig.wav, files track_256kbps.wav, track_128kbps.wav, track_64kbps.wav, and track_alac.wav (note -- 'diff track_orig.wav track_alac.wav' should yield no difference), which you can burn DAO to a CD and do your own listening tests using a good transport and DAC. It's also easy to listen to the residuals using a Python script that differences the samples in these WAV files from the original:

 

wavdiff.py

[font=courier new][size=1]#!/usr/bin/env python                                                                           

import wave,struct                                                                              

# open the original, codec, and output difference files                                         
orig = wave.open('track_orig.wav','r')                                                          
codec = wave.open('track_256kbps.wav','r')                                                      
diff = wave.open('track_diff256.wav','w')                                                       
diff.setparams(orig.getparams())                                                                

SHRT_MAX = 2**15-1                                                                              
SHRT_MIN = -2**15                                                                               

# process over a blocksize of bs samples                                                        
nsamps = orig.getnframes()                                                                      
bs = 1024*1024                                                                                  
bs = min(bs,nsamps)                                                                             
while True:                                                                                     
   osamps = struct.unpack("<%dh" % (bs*orig.getnchannels()),orig.readframes(bs))               
   csamps = struct.unpack("<%dh" % (bs*orig.getnchannels()),codec.readframes(bs))              
   samps = tuple(map(lambda x, y: x - y, csamps, osamps))                                      
   samps = tuple(map(lambda x: min(SHRT_MAX,max(SHRT_MIN,x)), samps))                          
   diff.writeframes(struct.pack("<%dh" % len(samps),*samps))                                   
   nsamps -= bs                                                                                
   bs = min(bs,nsamps)                                                                         
   if (bs <= 0): break                                                                         

orig.close()                                                                                    
codec.close()                                                                                   
diff.close()[/size][/font]

 

This code produces a WAV file track_diff256.wav of the sample-wise difference between track_256kbps.wav and track_orig.wav.

 

It wouldn't surprise me if I or others could distinguish between AAC at 256 kbps and the original bits, but it would surprise me if I could do so without a lot of classification error. I spent a little time on this and decided that AAC sounded excellent, and I probably wouldn't be able to reliably distinguish the difference, so I went with AAC for the computer. It would surprise me a great deal if someone could reliably distinguish the difference between ALAC and WAV.

[cvision123]

Basically AIFF and ALAC store the same information except that the original PCM data is compressed in the later format. With AIFF you don't have to decompress the data but the player needs to access the hard drive more to retrieve data. On the other hand the player has to decompress the data when using ALAC but reads less as the file size is about 40% smaller.

 

It totally makes sense when people reports that they hear differences. The reasons could be:

- On an old/not strong enough PC, the decompression process may stress the CPU and lead to some delay thus audible differences.

- When using a hard drive that has large latency between read commands also affect data retrieval process

 

However, on a strong enough computer such as the one with core i5, i7 and new hdd I doubt that audible differences exist since the computer can stably provide the same PCM stream to the DAC.

[Sik_Lescinovid]

Basically AIFF and ALAC store the same information except that the original PCM data is compressed in the later format. With AIFF you don't have to decompress the data but the player needs to access the hard drive more to retrieve data. On the other hand the player has to decompress the data when using ALAC but reads less as the file size is about 40% smaller.

 

It totally makes sense when people reports that they hear differences. The reasons could be:

- On an old/not strong enough PC, the decompression process may stress the CPU and lead to some delay thus audible differences.

- When using a hard drive that has large latency between read commands also affect data retrieval process

 

However, on a strong enough computer such as the one with core i5, i7 and new hdd I doubt that audible differences exist since the computer can stably provide the same PCM stream to the DAC.

 

Actually the CPU power can't have much to do with this since a small ALIX board running a measly 500MHz Geode CPU can easily handle on the fly decompression. If anything it might be the hard drive as you say, though even then I'm on the skeptical side of the camp. Regardless, it makes sense to do what sone did: find what works best for yourself.

[Boris75]

An audiophile friend compared a bunch of AIFF files with the same tracks copied to Apple Lossless and thought that the Lossless files didn't sound as good.

 

Has any one else had that reaction?

 

Is there any objective analysis (eg, a blind listening test) in the audio literature that shows any difference in listeners?

 

Obviously, I'd save a lot of space if I didn't have to use many terabytes for AIFF files.

 

Thanks.

 

 

With enough RAM, there is no difference.

[paul30d]

I am not trying to be smart, but files don't have a sound. They just sit there on a disk. So you cannot have a discussion on whether AIFF files sound the same as ALAC. You can convert from one format to the other thousands of times and still have the same data when you finish. Each file type contains exactly the same basic data.

 

So the question really becomes "do players play AIFF and ALAC the same way". Even though it seems logical that all players would use the same basic transcoding libraries (probably from Apple) I guess it is possible that some players handle these files in a different way and therefore could sound different when playing one format compared to the other.

 

Personally I hear no difference. But I am open to the possibility that some players do things differently with different file types. If that's the case, then the sound quality depends on the player, not on the file type.

[Musicophile]

With enough RAM, there is no difference.

Now I'm curious, why would you amount of RAM make a difference?

[Boris75]

Now I'm curious, why would you amount of RAM make a difference?

 

My hunch is that a computer with too little RAM will have to do a lot of data swapping between its RAM and hard drive when decoding the FLAC into PCM, and that this may create hiccups, or delays, in the transmission of the PCM signal to the output. Memory play can avoid this sort of issues, but also requires significant RAM.

[Musicophile]

My hunch is that a computer with too little RAM will have to do a lot of data swapping between its RAM and hard drive when decoding the FLAC into PCM, and that this may create hiccups, or delays, in the transmission of the PCM signal to the output. Memory play can avoid this sort of issues, but also requires significant RAM.

But then we are talking serious hi-res given that a full uncompressed CD is only about 700MB, correct? I don't suppose these days you'll find many computers with less than 2GB of RAM.

[elcorso]

I am not trying to be smart, but files don't have a sound. They just sit there on a disk. So you cannot have a discussion on whether AIFF files sound the same as ALAC. You can convert from one format to the other thousands of times and still have the same data when you finish. Each file type contains exactly the same basic data.

 

So the question really becomes "do players play AIFF and ALAC the same way". Even though it seems logical that all players would use the same basic transcoding libraries (probably from Apple) I guess it is possible that some players handle these files in a different way and therefore could sound different when playing one format compared to the other.

 

Personally I hear no difference. But I am open to the possibility that some players do things differently with different file types. If that's the case, then the sound quality depends on the player, not on the file type.

 

On the players and on the listeners. Personally I do prefer uncompressed.

 

Contrary to the common joke I do prefer the full cow than a piece of it.

 

Roch

[sone]

ALAC is compressed, but lossless. It's the exact same bits being sent to the DAC as the uncompressed file.

 

If anyone doubts this, or astonishingly believes that a computer with Gigabytes of memory cannot reliably decode 10s of Megabytes of audio content, perform this simple test: redirect your output audio stream to Soundflower, then using mplayer or Audacity record the ALAC and WAV and/or AIFF outputs to two files, then do a bytewise comparison of these files. They will be the same.

 

If anyone can identify a numeric difference in the output stream between ALAC and WAV/AIFF, please post it and settle the issue.

 

Otherwise, there simply is neither evidence nor any reason why lossless compressed should differ from uncompressed. The burden of the proof is on anyone claiming that a difference exists.

[elcorso]

ALAC is compressed, but lossless. It's the exact same bits being sent to the DAC as the uncompressed file.

 

If anyone doubts this, or astonishingly believes that a computer with Gigabytes of memory cannot reliably decode 10s of Megabytes of audio content, perform this simple test: redirect your output audio stream to Soundflower, then using mplayer or Audacity record the ALAC and WAV and/or AIFF outputs to two files, then do a bytewise comparison of these files. They will be the same.

 

If anyone can identify a numeric difference in the output stream between ALAC and WAV/AIFF, please post it and settle the issue.

 

Otherwise, there simply is neither evidence nor any reason why lossless compressed should differ from uncompressed. The burden of the proof is on anyone claiming that a difference exists.

 

From a mathematic or graphic point of view could be the same, but my ears/brain system tells another history.

 

Some people with trained ears catch this in less than 10 seconds, some others no.

 

Since I'm not interested in proof nothing will not discuss it anymore.

 

Even nasty politics and false propaganda force you to believe on the unbelievable and to buy things you don't need and can hurt you.

 

Roch