And you are right that lossless formats retain their bit depths when they are played.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Audiophile's Audio Format Showdown
- Thread starter Prodo123
- Start date
- Sort by reaction score
And you are right that lossless formats retain their bit depths when they are played.
Not that it's what this thread is about, but since we're on the topic I have a question about bit depths. DTS (as in plain old lossy DTS) supports 24 and 16 bit and two bitrates (768 and 1536kbps). Assuming that you encode a 24 bit source twice for both bit depths (and keep the sample rate), is there any improvement with the 24bit DTS track? Because the same bitrate is used, I don't see how it's any different.
Not that the codec matters, it's just the only lossy codec I could think of that supports 16 and 24 bit depths.
Not that the codec matters, it's just the only lossy codec I could think of that supports 16 and 24 bit depths.
MP3 is a proprietary format, patented by Technicolor in the US: http://en.wikipedia.org/wiki/MP3#Licensing_and_patent_issues
and a majority of MP3 codecs are owned by Fraunhofer. However, LAME is an open source codec for a proprietary file format.
All compressed lossless formats work the same way as ZIP, and ALAC, FLAC and WavPack all tend to get 40-60% compression.
For all ye extreme audiophiles, ALAC does not seem to handle 5.1 surround sound 24 bit 96000Hz audio well...(or my MacBook Pro for that matter)
Lossless is LOSSLESS. Got it. they are all bit for bit identical. The only reason to select on lossless format over another is because you have a player that accepts one format and not another.
Where's the option for AAC?
No, seriously. Although I don't consider myself the biggest hi-fi enthusiast on the planet, nor am I a cable guy, I do like achieving a great sound, but I'm a realist at heart. If I double blind tested multiple songs comparing their lossless vs 320kbps aac versions and couldn't find a difference, you bet I'd be using 320kbps aac.
Now time to eagerly await on getting called out for not having sufficient enough gear.
If you listen to just any random bit of music you may not be able to hear the difference between 320 AAC and lossless but if you listen to enough music you find that every now and then there is an "artifact" or transient defect in the compassed music. Another way to say this is that the difference in quality between AAC/MP3 and lossless is not constant over time. I hear these defects some times (on a few tracks) on certain drum hits and in some kinds of electronic music. AAC is good at encoding classical music and acoustic instruments. So A/B testing short segments will likely not turn up any difference. The way to test is to carefully listen and count "defects" over several hours and then see which has the most. MP3 will win hands down. AAC is much better and by definition lossless has zero.
That's not how it works. Your calculation assumes MP3 does 100% of its compression by reducing the dynamic range equally across all frequencies.
Honey, you've looked over a few points of consideration. Lossless is LOSSLESS, great. Now some formats are compressed and some are uncompressed. This makes a big difference in size. AIFF and WAV are our large formats, ALAC is smaller, and FLAC usually offers even better compression. Not only does size matter, but compressed formats can be more CPU taxing to decode. With a very powerful computer it usually does not matter. For slower systems, decoding a compressed lossless file can slow a system down and start affecting the sample rate. Some systems, like my iPod even skip when playing back large AIFF files because there is a lot of data to read.
Between a handful of lossless formats, some of them just don't have native support for certain players and devices. You pointed this out. But some formats have more features, such as storing tag information, streamability, etc.. While these formats are lossless, they do require time and space to reconvert if you are sharing them with others. ALAC is wonderful on Apple devices, but it's hard to find ALAC for sale or download anywhere. Some would just stick with FLAC to avoid the hassle of reconverting. I record my own music and a lot of live shows. I'd like to have them as ALAC but it requires me to make FLAC copies to post to archive.org or a torrent tracker.
This question is far more complicated than "lossless is LOSSLESS". I'd love to hear everyone's answers!
----------
If you read my other posts I make it clear that bit depth is a useless figure for discussing lossy compressed formats. I made a big point about PCM being an essentially graphical encode of the waveform, which corresponds bit depth to actual resolution. I said this wasn't the case with compression.
How slow does a computer need to be to have issues decompressing a FLAC/ALAC file? An iPod Nano/Sansa Fuze can do it without any issues.
I have no issues with ALAC files (individual songs) playing on my iPhone or iPod Classic. You said "large" AIFF files. Are you referring to a whole live show or something similar?
As you indicated, the preference has a lot to do with how you plan to use the files. For uploading to the web, FLAC is currently more popular and probably will be unless Apple decides to start selling ALAC on iTunes. However, if I download something, I always convert it to ALAC...I do not find it time consuming or much of a hassle. I just downloaded a program that can convert (you only need to do this one time) and them hit a couple buttons on a keyboard and the computer does the rest.
Jitter is not something I want to think about. It's barely measurable and it's not noticeable, but you know, this thread is about audiophiles who are the most foolish group of enthusiasts I can think of. Any insignificant -- but still real -- strain on a CPU is enough to trigger the jitter debate. I just wanted to point out to ChrisA that there is more than one point of interest when comparing lossless formats.
Sometimes. All of my live recordings are 24 bit, some are at high sample rates. I still notice some delay when advancing to the next 16/44 track in a cue. I use a 4th generation iPod Touch. I was hoping with the solid state storage it would be faster than the iPod Classic. Sometimes it isn't.
I really wish iTunes would sell lossless media. I am sure everyone here wishes that. I won't buy any music on iTunes because it is lossy. It's a problem now that artists aren't even bothering with CD releases. It's led me to piracy occasionally.
I still don't know if I will find an advantage to ALAC over AIFF right now. I have storage issues with AIFF, but it still plays native on all my devices and it allows tagging. I have an external library that has no space limitations, but my CD library is around 80 GB and my iPod is 64 GB. I don't care for taking everything with me. I would like to store a partial library on my MBP with SSD. I have 100 GB free, but I need most of it empty for using Photoshop. I've been considering a 32 GB or 64 GB SD card in my card slot. I want something light and unobtrusive. I use my Touch as an external library away from home sometimes, but it still isn't good for moving around a lot.
My personal favorite is AIFF at 48kHz, 24bit, and ~6000kbps... you practically need a Mac Pro just to play the file in iTunes
48kHz? Puny. 192kHz. 24-bit 7.1 surround AIFF is where it's at. Not even a Mac Pro can play it.
My personal favorite is AIFF at 48kHz, 24bit, and ~6000kbps... you practically need a Mac Pro just to play the file in iTunes
Sorry, is this a joke or something?
192khz is actually a downgrade in quality over 96khz because of less efficient filters resulting in higher distortion, ringing and phase I've read. So you're right in that aspect.
Oh trust me, I definitely don't just limit myself to short segments of A/B testing.
Then you should know that AAC is a horrible codec...especially for classical music, since it makes distinguishing between instruments extremely hard. If you're going to use a lossy codec, use a VBR codec. In 80-90% of the cases, LAME V0 sounds much better than either AAC 320Kbps, LAME CBR 320Kbps or LAME ABR 320Kbps. Takes up much less space, too.
I just re-encoded my entire library from ALAC to AIFF. Trust me, it's a freakin' huge leap from AAC.
And for the compression vs uncompressed debate, YES there is a difference between ALAC and AIFF. Big, noticeable difference in 70% of my songs. Also takes up 40% more space, but that's alright (at least for now while I have enough space lol).
There must be something wrong with your player if you are actually hearing a difference. It's also so easy to prove.
1. Open DAW (I'm using Logic) and import an AIFF version and an ALAC version of a song. (Logic doesn't do any conversion when importing, I don't know about other DAW's)
2. Invert the phase on any one of the tracks.
3. Listen to the perfect silence.
The silence means the ALAC file is decoding perfectly.
You're just like the group of people I can't stand reading on head-fi, the group who overly exaggerates the difference between a decent bitrate compressed song and an uncompressed song, almost to the point of implying it's a night and day difference. It's ironic considering that most of the most knowledgeable people on that forum with the most experienced background acknowledge that there's very little difference with uncompressed media.
I've taken distortion tests and abx tests online with good results, and I have adequate gear, so it's not like I'm trying to play you here about the difference being very subtle, nor do I see a reason for you to try and blow the differences out of proportion if only to sound more sophisticated.
Of course there's differences between 320kbps AAC and a lossless, but they're very subtle, and usually only seen in a slight dynamic range difference and more natural decay on treble percussion. Please come back down to earth.
I do not exaggerate one when I say those things. ALAC is known to be duller than AIFF, although the file may contain the exact same data bit for bit.
And have you even tried V0? It's much more practical than 320Kbps AAC.
I call snake-oil. Bit for bit means bit for bit.
Does 1 = 1 or not?
Your drinking too much Koolaid, man.
I believe Prodo123 is referring to the decoding process, what's contained within the file is only half the story.
Bit for bit is bit for bit, but compression methods are different.
I don't exactly know why, but AIFF sounds better than FLAC and FLAC sounds better than ALAC. I can theorize that compression methods are making an impact but it's improbable. WAV seems to sound just as well as AIFF. One thing is for sure: uncompressed codecs tend to sound better than compressed from my experiences.
Both AIFF and ALAC do not have the distortion that AAC and LAME have. I've noticed that the crispness is the big difference between the two codecs; presumably from the decoding process. I really converted to AIFF expecting no difference in sound quality, but I was quite surprised when most of my songs sounded different. I really hadn't the time to do a scrutinizing analysis of AIFF and ALAC until now (due to lack of storage, solved by 3TB drive). Logic shows no difference between the waveforms of the two, but they sound different on most computers and devices; I can't quite explain why.
Last edited:
I'm currently ripping all my CDs into ALAC, and have a question after reading this thread.
if I convert an ALAC track to AIFF directly (say using the right click option in iTunes) is the new AIFF track exactly the same as if I ripped the track directly from the CD to AIFF?
also is the same true if I were to convert from ALAC to FLAC/WAV ect?
if I convert an ALAC track to AIFF directly (say using the right click option in iTunes) is the new AIFF track exactly the same as if I ripped the track directly from the CD to AIFF?
also is the same true if I were to convert from ALAC to FLAC/WAV ect?
Hearing any difference between an ALAC version and an AIFF version of a song can only mean one of two things. Either something is broken and not is doing it's job properly or you are imagining things. Just do the simple test I suggested above and see for yourself. If you put two perfectly identical waveforms side by side and invert the phase of one of them, they cancel each other out and you get pure silence. Do it in Logic, one AIFF and one ALAC, you will only get silence every time, proving that they sound exactly the same.
A bit new to digital audio!! Used to really enjoy good old analog..but..portability!!
I will always have a place in my heart for anolog audio, vinyl and such with tubes glowing
(such depth/warmth with, to my ears anyway, a PRESENCE that I'm striving to match with digital). With the right dig equipment -and no, it need not be over the top expensive!- I'm finding that it's possible to achieve a convincing and pleasurable audio adventure, and this leads me to my question/advice seeking.
Currently, I have a FLAC app running on my iPad, and hence the majority of my music collection is in FLAC. I do believe it's certainly better than .mp3/.aac running @ 320kbs, perhaps not by a stunning margin, but still superior. On the XPS 630 desktop w the ASUS Xonar ST, the difference in quite noticeable.
My reason for switching to .AIFF over ALAC from FLAC would be as follows: .AIFF compatibility with iTunes, and .AIFF's supposed better (but possibly negligible) performance over compressed ALAC. I suppose, why not give your sound every practical (ah, that pesky definition of practical!) advantage in terms of quality. Does this sound like it makes sense? I'm a bit of a newbie, I admit it also any advice on maximizing sound quality from the iPad2? Thank you all very much for all of your time and good advice!
I will always have a place in my heart for anolog audio, vinyl and such with tubes glowing
(such depth/warmth with, to my ears anyway, a PRESENCE that I'm striving to match with digital). With the right dig equipment -and no, it need not be over the top expensive!- I'm finding that it's possible to achieve a convincing and pleasurable audio adventure, and this leads me to my question/advice seeking.Currently, I have a FLAC app running on my iPad, and hence the majority of my music collection is in FLAC. I do believe it's certainly better than .mp3/.aac running @ 320kbs, perhaps not by a stunning margin, but still superior. On the XPS 630 desktop w the ASUS Xonar ST, the difference in quite noticeable.
My reason for switching to .AIFF over ALAC from FLAC would be as follows: .AIFF compatibility with iTunes, and .AIFF's supposed better (but possibly negligible) performance over compressed ALAC. I suppose, why not give your sound every practical (ah, that pesky definition of practical!) advantage in terms of quality. Does this sound like it makes sense? I'm a bit of a newbie, I admit it
also any advice on maximizing sound quality from the iPad2? Thank you all very much for all of your time and good advice!
Last edited:
I will always have a place in my heart for anolog audio, vinyl and such with tubes glowing
Currently, I have a FLAC app running on my iPad, and hence the majority of my music collection is in FLAC. I do believe it's certainly better than .mp3/.aac running @ 320kbs, perhaps not by a stunning margin, but still superior. On the XPS 630 desktop w the ASUS Xonar ST, the difference in quite noticeable.
My reason for switching to .AIFF over ALAC from FLAC would be as follows: .AIFF compatibility with iTunes, and .AIFF's supposed better (but possibly negligible) performance over compressed ALAC. I suppose, why not give your sound every practical (ah, that pesky definition of practical!) advantage in terms of quality. Does this sound like it makes sense? I'm a bit of a newbie, I admit it
As a guitar player, I love the sound of a good tube amp
but so far I have not been able to fully replicate such warmth on digital signals. Mostly because of better-than-average DACs that most computers ship with.http://www.apogeedigital.com/products/duet2.php
This will improve sound quality by a lot.