Jump to content
IGNORED

AIFF vs Apple Lossless redux


rom661

Recommended Posts

I know that this is a topic that has been discussed before on the Forum but I have recently done something that gives me a different perspective. It tends to validate my earlier experiences as well as the explanation that was offered, but does so to a greater degree.

 

I recently purchased a new car. I went with the entry level stereo, which is truly horrible (A VW Jetta with the turbo and DSG transmission option - nice car for a modest investment but that's for another forum). One advantage the car had is that the stereo, modest/bad though it is, is a dual din height and is not tied into all the other electronics in the car. In other words, a complete replacement is feasible. I know this is not a car stereo forum either but bear with me.

 

Alpine makes a head unit that allows USB connection with an iPod and provides full touch screen control of your music. My last car would only allow blind control of six playlists. More significant though is how the music signal is handled. It is also on the USB output and provides a digital bitstream to the head unit. The D/A conversion and analog output stages are provided by the Alpine unit, not by the iPod. Starting to sound more familiar now. I was really excited about the potential for at least decent sound in a car. The rest of the system is JL speakers, amp, and sub.

 

I had done some earlier experiments with AIFF vs Apple Lossless on a very revealing two channel system running on a Mac Mini into a Wavelength Cosecant USB DAC. I concluded that there were subtle improvements with AIFF so I stuck with that. Others expressed their view that there couldn't be an audible difference and I have to emphasize that it was subtle. Because of that experience I decided to store my music on the iPod in the car in Lossless, figuring that the much lower fidelity of that system would make any potential difference completely irrelevant and I would save the space.

 

I was not happy with the sound. At first I thought I needed to tweak the system, which I did, and it needed to break in, which it did, but I finally put the CD of one of my ripped CD's in the drive of the Alpine; much better. I was kind of resigned to the apparent fact that the iPod source just didn't have the potential to sound decent. On a whim I duplicated one of the CD's already on the iPod but ripped this one in AIFF. There is a MAJOR difference, much greater than what I heard in the comparisons in my home system. The CD was still better but it was much closer and good enough to be quite enjoyable.

 

Gordon Rankin of Wavelength suggested that the better your computer, i.e. faster, more memory, etc., the less difference you will hear between AIFF and Lossless. Purely a guess on my part but I suspect that is what is happening here. Obviously the difference between a Mac and an iPod is quite large.

 

I don't know if this is helpful to anyone either in terms of explaining the differences heard between the storage formats or helping someone try to get optimum performance out of their iPod when using it strictly as a drive for your music. It certainly turned my little project around.

 

Please note that I don't pretend to understand the "whys" of it beyond passing along Gordon's thoughts. I do know what I observed and it is not at all subtle.

 

Best wishes to all.

 

Audio Research DAC8, Mac mini w/8g ram, SSD, Amarra full version, Audio Research REF 5SE Preamp, Sutherland Phd, Ayre V-5, Vandersteen 5A\'s, Audioquest Wild and Redwood cabling, VPI Classic 3 w/Dynavector XX2MkII

Link to comment

Hi rom661,

 

Assuming that nothing went wrong during the ripping process, the playback bitstreams from AIFF or ALAC should be identical. Lossless encoding is by definition, lossless, so will be bit-for-bit identical to the original.

 

So this only really leaves jitter, and I can see that this could be worse using ALAC on the iPod. In decoding an ALAC file, the processor in the iPod is having to work a lot harder than for AIFF, and it will therefore draw more current. This in turn will cause supply voltage fluctuation, and in turn clock jitter. So if the Alpine unit doesn't do any jitter reduction (highly unlikely), its DAC will do a worse job with ALAC.

 

With AIFF, the iPod doesn't have to do much, other than write the L and R samples out to the USB port, so it is only lightly loaded.

 

I can't think of any other cause.

 

Chris.

 

Link to comment

I should have mentioned, Chris (CS) that I did a comparison of the duplicated piece of music on a high end system in the store and heard a very small difference, so the ripping was not an issue.

 

As far as explanations, you may be right. I am good at observation and comparison but am not qualified to offer explanations in the digital world. My technical expertise is good where speaker design and acoustics are concerned and fall off quickly after that. Thanks for your input.

 

Yeah, Chris, I'm good at painting a target on my back.

 

Rick

 

Audio Research DAC8, Mac mini w/8g ram, SSD, Amarra full version, Audio Research REF 5SE Preamp, Sutherland Phd, Ayre V-5, Vandersteen 5A\'s, Audioquest Wild and Redwood cabling, VPI Classic 3 w/Dynavector XX2MkII

Link to comment

CS,

 

Not correct it is not the jitter.

 

Yes both the ALAC files are bit equal to the AIFF and in my test we actually converted one from another and verified they were equal. But it seems like there is a processing issue with ALAC that is not required with AIFF as it is pure PCM, only with the tag information included, something that makes it win out over WAV format.

 

I have not figured it out but in test some of these USB iPod or what is called "USB to Go" iPod products I found that the results from ALAC were not nearly as good as AIFF. I would comment on numbers but I have gotten a few calls from presidents about that action.

 

~~~~~~~~

 

But I wonder if really Alpine is doing the "USB to Go" iPod thing or if it's like my Mini setup were they use the line in and control it via the serial interface control of the iPod. I mean mine sounds pretty good with the old standard 10GB hard drive unit I have with my updated Focal speakers and such with ALAC files on it.

 

Thanks

Gordon

 

Link to comment

Gordon

 

I wondered the same thing and contacted Alpine. They assured me that their piece was doing the conversion. In fact they said it upconverts to 24/96. I know you don't consider that a positive but it would seem that it is using the Alpine DAC.

 

Rick

 

Audio Research DAC8, Mac mini w/8g ram, SSD, Amarra full version, Audio Research REF 5SE Preamp, Sutherland Phd, Ayre V-5, Vandersteen 5A\'s, Audioquest Wild and Redwood cabling, VPI Classic 3 w/Dynavector XX2MkII

Link to comment

Gordon, Rick,

 

What is the form of connection between the iPod and the Alpine unit ?

 

If it is just USB (ie. digital connection) the problem can then only be incorrect decoding of the ALAC file (by the iPod) or sample timing (ie. jitter in the Alpine's DAC and/or USB interface). There isn't anything else to get wrong !

 

If however, the Alpine unit is actually taking the analogue output from the iPod, then you are at the mercy of the DAC in the iPod, which won't have very good jitter performance. Again, it could be that the iPod isn't correctly decoding the ALAC file, but I doubt that. Apple will almost certainly be using the same algorithm as they use in iTunes.

 

Chris.

 

Link to comment

Rick,

 

Are you able to capture the file onto your Mac or PC from the iPod via the USB interface ?

 

Then it would be possible to determine if the iPod is doing the ALAC decoding correctly. (It seems you are confident that the stored file is OK).

 

Chris.

 

Link to comment

I am confident that the stored file is OK. I really don't have the time or inclination to pursue this beyond what it took to get me decent sound. I know; a classic case of punch and run. I just wanted to share an observation with others and possibly help shed some light on a "contentious", to quote the other Chris, subject. I don't know if Gordon's explanation is correct but based on my very positive experience with his products I am inclined to consider him credible.

 

Happy listening.

 

Rick

 

Audio Research DAC8, Mac mini w/8g ram, SSD, Amarra full version, Audio Research REF 5SE Preamp, Sutherland Phd, Ayre V-5, Vandersteen 5A\'s, Audioquest Wild and Redwood cabling, VPI Classic 3 w/Dynavector XX2MkII

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×
×
  • Create New...