Off topicmoved from here http://www.diyaudio.com/forums/pc-based/212145-new-mac-minis-internal-dac.html 
So, do you claim that you can hear differences between lossless AAC, AIFF, FLAC ?
So, do you claim that you can hear differences between lossless AAC, AIFF, FLAC ?
Do I claim that I can hear differences? Maybe, maybe not. When I do AB comparisons my ears tell me that I perceive subtle differences. I'm not making any strong claims here. It's quite likely that if I did double blind tests I wouldn't hear a difference. This is the case with SO much DIY audio. You are faced with endless numbers of choice situations. You evaluate A and B. Could be a resistor, a tube, a piece of software, anything. If your ears tell you A sounds better than B on that particular day at that time in those particular circumstances, you would make the real-life decision to go with A. If you can't tell the difference you'd go with what's cheaper of more convenient.
In this case it's more convenient not to go with AIFF, however, my ears were telling me that there was a possible advantage with AIFF. So that's what I did. It costs me nothing - I have a large external hard disc. So I was happy with that decision and I'd do it again. As for other people I have no idea. I've read posts that prefer AIFF, and posts that hear no difference, so there we are.
andy
In this case it's more convenient not to go with AIFF, however, my ears were telling me that there was a possible advantage with AIFF. So that's what I did. It costs me nothing - I have a large external hard disc. So I was happy with that decision and I'd do it again. As for other people I have no idea. I've read posts that prefer AIFF, and posts that hear no difference, so there we are.
andy
Lossless formats are called lossless because when data packed in them is unpacked, the unpacked data is the same (bit-identical) to the data that has been packed.
If your ears hear the same set of bits differently from session to session, something else than the format should be checked.
If you think that DAC plays something else than the bits fed into it,. then what is it ?
If you think that lossless formats alter bits to fed into DAC, can you present any proof ? If you can, first file a bug against the lossless codec - because by definition of 'lossless' there can't be differences.
If you think I miss something in formats <-> DAC reality, what is it ?
It's not a matter of theory - theory is constantly in flux. You might want to ask yourself, if lossless and AIFF are identical, then why does AIFF exist at all?
What I'm referring to is your assumption that YOUR definition of theory determines what EVERYBODY ELSE experiences. If the result of this is that you go on to label those who contradict you as mad or deaf, then you show a pitiful lack of curiosity about the world you live in. I've already said I may be right and I may be wrong, but I have done what DIY builders do in the real world in choice situations, which is to make AB decisions. What more do you want me to say? That I worship your superior intelligence? What exactly do you want out of this conversation?
What I'm referring to is your assumption that YOUR definition of theory determines what EVERYBODY ELSE experiences. If the result of this is that you go on to label those who contradict you as mad or deaf, then you show a pitiful lack of curiosity about the world you live in. I've already said I may be right and I may be wrong, but I have done what DIY builders do in the real world in choice situations, which is to make AB decisions. What more do you want me to say? That I worship your superior intelligence? What exactly do you want out of this conversation?
<<<if lossless and AIFF are identical, then why does AIFF exist at al?>>>
- a wrong question showing wrong understanding of reality.
AIFF is an uncompressed format. Lossless compression assumes finding and eliminating redundancy (if any). If there is no redundancy (i.e. if you record true noise, there will be no compression.
So, AIFF still exists because it was historically invented before ausio lossless compression and because it is easier to implement. I.e. a programmer has less job to do.
<<<What I'm referring to is your assumption that YOUR definition of theory determines what EVERYBODY ELSE experiences>>> - nonsense. The definition is not mine and you won't be able to present ABX testing results proving your point.
<<<What more do you want me to say? That I worship your superior intelligence? >>> - provided that the former doesn't contradict the latter.
My definition of G-d: the entity responsible for concrete realization of random processes.
You failed to answer my very simple questions:
1) does lossless compression change the bit pattern fed into DAC ? (the answer is NO);
2) what else other than the bit pattern the DAC plays WRT to data format ? (the answer is NOTHING).
You appear to misunderstand very fundamental things about how things (do not) work.
For example, if you think your AIFF file is stored on disk as is, you are wrong.
The data is stored in sectors, and since reading may fail, ECC codes ( see Error detection and correction - Wikipedia, the free encyclopedia ) are introduced and then "stripped off" of your data. I.e. data on disk doesn't at all look as AIFF data, and on different disks different error correction code can be used.
So now I suggest to start up a new audiophile religion that will describe perceptual differences as a function of ECC codes used on various disks. I expect from now on to read heated discussions how audio played from various normally functioning disks perceptually differs.
Don't forget to address the issue that nowadays semiconductor "disks" (SDDs) are becoming more and more popular.
By the way, there is a thing called "ramdisk": RAM disk - Wikipedia, the free encyclopedia .And RAM can be implemented with ECC or without ECC. If you omit in your religious discussions perceptual differences of sound plaid from normally working ECC RAM and non-ECC RAM, as well as if you omit discussions of ramdisk <-> (hdd, sdd), I'll consider you to be biased and partial.
Don't also forget that RAMs can be static and dynamic, and there is a variety of the latter.
How about the fact that RAM cells can be of different size ?
How about the differences between the same bit patterns coming from an optical drive and from an HDD ?
Well, my view of reality here is that we are on an Internet forum having a conversation. I still have no idea what your motivation is and what you want out of this discussion. Do you want me to admit something? Do you want me to change my listening habits? What exactly is motivating you here?
My motivation is to remind people that storing music in lossless (rather than uncompressed) format saves disk (storage to be exact) space and thus either allows to save money because smaller amount of storage is needed or to store more music for the money already spent.
No music quality is lost in such a scenario because of the definition of 'lossless'.
AAC & AIFF yes. FLAC doesn't play natuvely in iTunes so Max changes them to AIFF for me.
AAC is marginely better than MP3 and you know what they say "friends don't let friends listen to MP3s"
dave
But the work of unpacking the file causes the CPU to work harder & perturbs the power supply. Not an issue if you use memory play in PM, because it makes you wait while it loads the uncompressed file into RAM before playing it.
Memory play sounds better -- again because loading it off the HD perturbs the power supply.
dave
The old bits is bits argument. That is a prereque for good playback, but then you have to get the timing right.
dave
and you can't prove, using ABX, that they sound the same -- because ABX is not capable of proving 2 DUT are the same, only that they are different. ABX is a weak test (in the statistical & the real-world sense)
The timimg. One of the few perceptual studies that has been independently verified is that perceptive people are able to detect differences on the order of 5 µsec. That is a pretty fine line.
Very few people understand this thouroughly and even they are learning stuff.And from your argurments you seem to be missing some important fundementals.
In the end, if it makes Andy,or me, or someone else, feel more confident and lets them relax, use more of the artistic side of the brain, allowing a greater emotional connection with the music, then it is good, because in the end, it is about the music.
dave
Nonsense.
Music is played not from file, but from memory buffer. I.e. first the memory buffer is filled (be it from uncompressed or compressed file, and be it in lossless or lossy format - in the latter case first plaim PCM is created - with all the losses), and then the memory buffer is fed into DAC.
So, there is no difference in timing - the whole buffer is flushed into device.
Timing is not related to compressed/plain format.
Say, an hour ago the radio was broadcasting a nice jazz recording made in 1956. 56 years delay, and still sound good.
Every time the buffer is refreshed the power supply is perturbed. You have to buffer an entire track to get around that.
dave
Sure it is. The act of decompressing the file just before play perturbs the power supply more than just reading the uncompressed file.
dave
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.