Best file format for Mac & converter software

[wgscott]

You can check the claim that the wav file produced by max is not corrupted by changing it back to flac (with max or xld) and then to wav again using xld. If that wav file differs from the other wav file only touched by max, then one of these is corrupting the audio. Otherwise, this requires a shrink to figure out.

 

[goldsdad]

"try this

You can check the claim..."

 

In case that's for me, yes, I absolutely agree that the audio data will be identical in the WAVs. I am confident that, to my ears, they will sound identical, too.

 

The point I was making was that you have to be careful how you compare the files. MD5 will say files do not match simply because Max puts the time (and other data) in the end of a WAV.

 

ST5 should say the audio contents of the files are identical. Also, inverting the polarity of one file's audio and mixing it with the other's untouched audio should produce a null, again proving that they contain identical audio.

 

 

[Audio_ELF]

I've decided on a new maxim...

 

"You may think that (they soon different)... I could not possibly comment!"

 

Thanks Francis

 

Eloise

 

[wgscott]

Both converters produce the identical content, as goldsdad suggested.

 

The version giving the most realistic sound was the WAV copy from 'Max', which was more real than the WAV copy from 'XLD' which was better than the original FLAC copy.

 

So Max produces a padded wav file that, due only to the padding and time and other data, sounds better than the one produced by XLD?

 

Sorry, but I have to go smelt some more aluminium ore for my hat production.

 

[elcorso]

"There is something rotten in Denmark" (William Shakespeare - "Hamlet")

 

"There are things that seem things, and returned the other way around are things again" (Anonymus Phylosopher & Comedian - Translated from Spanish)

 

"...That said, I don't discount anyone for believing what they want about audio and measurements. We can all enjoy music no matter how we think." (Chris Connaker - CA - Sun, 05/10/2009 - 20:44)

 

I know, @wgscott, this is hard to swallow and even worst to digest: The sound differences are minimum, but listenable...

 

But right now I want to keep away from this discussion, it could very heavy (or maybe crazy?) for some members, and I don't want to be called, again and again, a nonsense or imaginary guy anymore. Not today, I don't now tomorrow!

 

Regards & happy listening,

 

Roch

 

[Blu]

I only have a few moments to add to my previous comment about WAV being the best format and 'Max' being the best converter, but I will elaborate when I have more time.

 

In the settings use only one encoder in 'Max', as I found it produces a better sound than a higher number.

 

EDIT: See signature for software settings.

 

Thank you.

 

[Blu]

Rene,

 

I have included the settings I use for Max in my signature. For Format choose WAV (Microsoft) and the 16 bit top setting.

 

I hope it works for you as it has for me.

 

Blu

 

[Blu]

I don't use checksums to test the quality of the conversions, I use a simpler method called hearing. Which ever version sounds closest to real instruments or voices and can cause me to get up and dance or sing along, is the best version.

 

My goal is to have my sound system produce the most realistic representation of a musical performance in my listening room, using resources available and through trial and error arrive at my goal.

 

I shall listen with an open mind, as I'm aware that we are still discovering properties of ripping, conversion and playback software we cannot yet quantify.

 

An example being the different sounds produced by the various players using the same source material. 'Pure Music', 'Decibel', 'Fidelia' & 'Audirvana', all sound different. To me which ever converter or player causes the least change to original well recorded material is the one that is going to sound the best.

 

I have found that the methods I have outlined so far are helping me achieve my goal. No doubt over time as more knowledge is gained, adjustments will need to be made.

 

Thank-you, keep enjoying the music.

Blu

 

[Audio_ELF]

[T]here are known knowns; there are things we know we know.

We also know there are known unknowns; that is to say we know there are some things we do not know.

But there are also unknown unknowns – the ones we don't know we don't know.

 

The question is: do we know two files with the same checksum are the same, or are there unknown unknowns coming to play...

 

And if you think that's confusing... Well join the club.

 

Current computing theory says two files with identical checksum must be the same (within limits of checksum where two completely different files can have same checksum but you'd notice that!) and therefore using the same player the resultant output must be the same. [note: this doesn't preclude possibility that the same file stored on different drive may sound different at playback]

 

THIS IS FACT NOT OPINION.

 

If the resultant sound is different. Then there must be some other unknown factor coming into play!

 

Eloise

 

[Audio_ELF]

Hearing vs checksum for "testing" is comparing a subjective testing method (hearing) to an objective one (checksum).

 

Scientifically if the objective method gives different results to the subjective method, it's the subjective one that should be questioned NOT the objective one.

 

Sherlock Holmes said it best: once you have eliminated the impossible (that the files are different: we've proved they're not by checksum) whatever remains, however improbable, must be the truth.

 

Eloise

 

[Renerator]

Thanks for the response. Like a dummy, I looked closely at your signature AFTER I asked the question. I have made the adjustments as you say and all is good!

 

Thanks again.

 

R

 

[elcorso]

Eloise,

 

This could be a never ending discussion... since I don't have your ears, and you don't have my ears! Then it will be impossible to make an statement regarding wich "identical" music file sounds the best, or different.

 

And you are right, in computer terms they are identical. But, who knows? Technology is improving every year, every month and minute, and somebody could prove, some day, that "identical" files are different.

 

But under today scientific test your are right.

 

My only complain is why not enjoy the music by itself, and let the test suffering for recording engineers. I understand they make this test when they listen that something is wrong after listening. Or because the recording was so great to their ears they want to see in graphs the answer.

 

At the end, are our ears who enjoy the music, not our eyes, nor perfect mathematics formulas!

 

Happy listening,

 

Roch

 

 

[wgscott]

You'll enjoy this: Two identical aiff files, with identical checksums, that are different in size:

 

zsh-% ls -l temp*/*.aiff

-rw-r--r-- 1 wgscott staff 26K Mar 6 10:10 temp/StyleChangeMultiple.aiff

-rw-r--r-- 1 wgscott staff 26K Mar 6 10:10 temp2/StyleChangeMultiple.aiff

 

zsh-% md5sum temp*/*.aiff

80aacb65e092b91b743df3fc85081641 temp/StyleChangeMultiple.aiff

80aacb65e092b91b743df3fc85081641 temp2/StyleChangeMultiple.aiff

 

 

zsh-% du -h -d 1 temp*

28K temp

16K temp2

 

 

I made the second one from the first one using hfs filesystem compression, i.e.,

 

ditto --hfsCompress temp temp2

 

However, the differences are an illusion. (Sorry if I got your hopes up.)

 

[nigel]

And you are right, in computer terms they are identical. But, who knows? Technology is improving every year, every month and minute, and somebody could prove, some day, that "identical" files are different.

 

Please stop using your ignorance of computers as an argument that everyone else must be equally ignorant.

 

Some of us actually design and build computer systems for a living (and have done for over 20 years). If we were as ignorant as you, none of your computers would ever work at all (or even exist).

 

[Audio_ELF]

That's very good advise there Roch... Enjoying the music is the real purpose of this hobby/obsession of ours. I'm happy for anyone to believe using EAC vs XLD; a Teac drive vs inbuilt Mac drive; or sprinkling fairy dust all over their computer and ripping only when there's a full moon will influence the sound quality of their playback: but there is no objective evidence this is true so telling people (especially newbies) these so called facts is just confusing for them.

 

As Nigel said, if identical files could be different then a computer wouldn't work. Every time you copied from a CD to install a programme there would be a risk that the programme wouldn't work for starters.

 

Eloise

 

[elcorso]

@wgscott,

 

Somebody said:

 

"Everything in the universe is movement and thereby illusion. When we learn who we are we will become Magician's and play with the illusions."

 

The complete poem:

 

Reality and Illusion (Tom Kitt)

 

There is no middle ground with reality

Things are either real, or not!

If nothing is real

Then all is illusion.

If all is illusion

Then all is magical.

If all is magical

Then everything is possible.

If everything is possible

Then we have no limitations.

If we have no limitations

Then we are capable of becoming real.

 

Eloise regarding: "As Nigel said, if identical files could be different then a computer wouldn't work" Was me who stated that, no @Nigel, but please remember I don't know nothing about computers (@Nigel statement)...

But I will change this statement because last night I was playing a music track, and then, in the middle at the music, was ten seconds of "pink noise", then music again. I was so bored because this disturbed my joyce of listening, that compared this file to the original (who sounds OK) and they was "identical". And the app and my Mac wasn't disturbed by this situation, they went working together...

 

And Eloise, you are right, my stupid (@Nigel compliment) statements could be confusing, since they are not objetive. Then I'll stop my stupid (@Nigel compliment) observations regarding this matter, and maybe let my Mac make the next comment.

 

And yes I'm an ignorant (@Nigel compliment) in computer matters, thanks God I don't "think" like a computer, even if I had being 40 years on them, because my Mac work for me, not me for the Mac.

 

@Nigel profits that there is no "computer filter" for offensive words in this forum, but it should be Chris!

 

Happy listening to music (not to bit and bytes)!,

 

Roch

 

[wgscott]

Hesham:

 

If you haven't given up, I hope the following provides something of an answer for you:

 

The three Apple-optimized formats are

 

AAC -- generally lossy compressed, (analogous to mp3)

AIFF -- how Apple sees what is on a CD. Lossless, uncompressed. (analogous to wav)

ALAC -- Apple's Lossless Audio Codec. Lossless, compressed. (analogous to FLAC)

 

iTunes will also recognize your WAV files, but not FLAC. I don't know about the other formats. Third-party player software options will generally handle more, almost always FLAC.

 

My iTunes library consists of ALAC, AAC, and mp3 tracks. I don't hear a difference between any of the lossless formats (AIFF, wav, flac, ALAC) and am quite skeptical of claims to the contrary. If I rip a CD I care about, I rip it to ALAC. If I download FLAC or wav or AIFF, I convert to ALAC using XLD, and I save the originals. If I rip a CD for which lossy compression is something I can live with, I rip it to AAC.

 

If you believe all lossless files are fundamentally the same (since it can be verified you can convert from wav to any of them and back to wav and recover the same file with the same checksum or md5sum), then I would choose ALAC, since it has been optimized by Apple for Apple equipment. If you want universality and platform-independence, FLAC is the best bet since it is an open-source standard. You can always regenerate the wav or AIFF file from these losslessly compressed formats, should the need ever arise.

 

 

XLD is free and has the advantage of being able to recursively go through all your audio files in subdirectories automatically, output ALAC (or whatever you choose) to a mirrored hierarchy of directories, and will also automatically load everything into iTunes, if you select that option.

 

 

[wgscott]

In about 45 minutes, the tsunami originating from the tragic earthquake that struck Japan 12 hours ago is predicted to arrive. The waves were about 6' when they arrived in Hawaii. Some of our local residents have put all of their belongings in their cars and have driven up to the 2000 ft. summit of the highway that goes to San Jose.

 

My guess is these are the same types of people who firmly believe we really don't understand enough about physics to predict what will happen. Perhaps amongst these refugees are a disproportionate number, relative to the local population, who believe files with identical checksums can sound different.

 

I'm about to take the ocean highway to drive my kid to junior high. If my "flat-earth faith" in physics, and science in general, has turned out to be wrong, this will be my last post, and it has been nice knowing you all.

 

My guess, however, is you won't be rid of me that easily...

 

 

[Paul R]

AAC is of course, a "lossy" format, so converting them to AIFF is not really going to buy you a lot. It will not restore the missing bits, though some converters will try to interpolate them.

 

If the AAC files are ripped from CD's, then your best bet is to re-rip them as AIFF files. The improvement will be - eh - dramatic. If they are from iTunes, then converting them will change the way they sound, but the change may or may not be an improvement. It will be different.

 

While I can hear slight differences between AIFF and WAV, AIFF is the format of choice for me.

 

-Paul

 

 

[4est]

I can tell a WAV file from the other room. Literally, the other room. I find it to often be a good test for frequency balance to move into another room. In this case the "ease" I hear in a WAV file can be heard even when I switch between them in shared screens from another room. One of these days I am going to have to figure out how to batch convert all of my AIFF and keep the metadata if that is possible.

 

[blaine78]

If you had access to a windwos machine, dBpoweramp will convert, delete old file, keep metadata from original file and add it to the wav file.

I will only use WAV now, really does sound better than all lossless including AIFF, this is on a windows machine

 

[willmanus]

So a while back I rashly ripped all my CDs to Apple Lossless. I too have heard the benefit (albeit slight) of WAVs, but my question is this: do I a) re-rip everything to WAV or b) convert all my Apple Lossless to WAV.

 

If i understand it correctly all the information is in the Apple Lossless it's just that it's in a compressed format. So converting to WAV should be fine right???

 

Has anyone tested this? I will do my own listening but I wanted some of you experiences first.

 

 

 

[wgscott]

If you take a wav file (for example), convert it to Apple Lossless, and then convert it back to wav, the md5sum is the same as the original wav file, which means that it is bitwise identical.

 

I have done this, and have posted the results here (on other threads on the subject). So have many others.

 

Before you go to the trouble, you should do a double-blind test to make sure you really can identify the difference. You have to get it right 5 or 6 times in a row for it to be statistically compelling.