alright well then you're on my #$^%tlist too ! just kidding......no i'm not willing to do a short burst ab test for the reasons i laid out previously. If you'd read them and followed it you would know i'd need months to do it otherwise it's meaningless. It's like psychologists who can't study human intention because it's totally incomprehensible and so they reduce it to stimulus and response actions from dogs and unintelligent people and they think they're studying motivation of a fully aware human.
alright well then you're on my #$^%tlist too ! just kidding......no i'm not willing to do a short burst ab test for the reasons i laid out previously. If you'd read them and followed it you would know i'd need months to do it otherwise it's meaningless. It's like psychologists who can't study human intention because it's totally incomprehensible and so they reduce it to stimulus and response actions from dogs and unintelligent people and they think they're studying motivation of a fully aware human.
This just doesn't make sense. Human auditory memory is vanishingly brief. This is a proven fact. Long-term listening has no relevance to distinguishing between lossless formats. If you can hear a difference on a particular part of a song or an entire song, you can play that song in the various formats and tell which one is which, right next to each other. Nothing else matters in the context of proving this. It's not an A/B test; it's an ABX test. You know beforehand that, for example, A is WAV and B is FLAC. A and B are played one after the other. Then an unknown variable, X, which is either A or B, is played at random after that. You choose whether you think X is A or B. If you pass (with 9/10 or 15/20, for example), then it proves that you really can distinguish between those two samples. If you cannot do that, then you cannot prove that you can distinguish between them.
we are on different wavelengths here alchemist. I'm not talking about memory, i'd agree with you on that point.....please reread some of my past posts if you have an interest in my take.....short story, i don't believe we hear with our ears and it's a much deeper process than most think (especially those here in the sound science sub section).....and I already admitted proving it to others is impossible, these kinds of things have to be first hand, similar to spiritual experiences. If you had one you'd not be able to convince or explain it to another. and that's the way I listen to music. I don't just sit there like a log and 'take it in'. I'm actively involved. It's a long story and one you'd not believe so it's moot point. Pretty soon someone will suggest that a robot can hear two a/b samples and that will once and for all prove if he can detect a difference. It always devolves into robots. check out stew's understanding of this debate for why reductionism may not hold water, then consider how this applies to music, sound and headphones and youlll have my response to your question
https://www.youtube.com/watch?v=YpUVot-4GPM
we are on different wavelengths here alchemist. I'm not talking about memory, i'd agree with you on that point.....please reread some of my past posts if you have an interest in my take.....short story, i don't believe we hear with our ears and it's a much deeper process than most think (especially those here in the sound science sub section)
I read everything you said...and none of it has anything to do with successfully documenting your ability to distinguish between lossless formats. How could it possibly? Casual listening and stating your preference for one player over another proves nothing.
There isn't anything metaphysical about this. It's just identical computer bits being converted into analog waveforms...
because I told you it's impossible to document......I added to the last post so check it out and let me know......and even I understand that they are not only being converted to analog, be decompressed first in the case of flac. that is the possible area of difference. Otherwise, we agree they are identical. the question is, is that one difference meaningful or not? you and others are convince it's not, and you may be right, but are you 100% certain the unpacking is of no significance....I can't proove it either as i am technically illiterate, and the only way i could prove my subjective experience is over time, months of listening to blind flac and WAV. i've done this experiement, just can't reproduce it for you to 'see' and again, I may be delusional i admitted as much
You're right about one thing: the test you are proposing is impossible to document. An ABX test (which is more or less the only accepted methodology for this type of thing) is easy to document, on the other hand. Not sure what your goal is for posting in Sound Science about how WAV sounds better to you if you aren't willing to back it up.
I have no purpose other than to 'weigh in' my delusional first hand experience and support the poor OP as he is being attacked from all fronts like he's the antiXrsit. in any case, check out that stewart vid, some good stuff there! ..........
bigshot also feels all amps should sound the same or else they are faulty and should be sent back the the manufacturer? I'd ask him, what should that 'same sound' be? how does he know when they sound the same? imo, bigshot is so off base with his primary assumptions about sound it's no wonder he has lost interest and think's i'm here in need of entertaining.
is this true> Unfortunately for our purposes here, the ABX comparator creates a temp .wav file for each sample, and that is what you are hearing, since it assumes no difference in real time decoding.
I understand it convert all audio to 32 bit float PCM files first and then randomize them. Great, you have 2 WAV files you can’t identify so indeed a blind test. Not so great: if the hypothesis is that differences in sound quality between WAV and FALC are due to the computational effort required during conversion to raw PCM, this ABX test won’t help as you are comparing bit identical raw PCM to raw PCM
some good info and comments here
http://www.audiostream.com/content/further-adventures-flac
I like the way you operate alchemist. you have conducted yourself like a right gentlemen and it's been a pleasure chatting. now where's that evidence for alchemy!
touka koukan! ever then you can't make something out of nothing.
@thelostMIDrange
you make the assumption that long listenings are better to tell differences. it's false, pure and simple. others have told you so, some have pointed out to some links to experiments showing it unequivocally. you believing in your own opinion isn't making it true. and obviously your conclusions based on your wrong experiment methods can only be dismissed as irrelevant.
it has been studied a lot and it's actually easy(but mighty long) for anybody to conduct some tests on their own to see the very logical and obvious conclusion. each step of the memory is faulty and as soon as you go past the first ... 3to 5seconds, the memory moves toward storage in the long term memory and is further distorted by keeping only the most obvious information and making them bigger so that they're easier to remember. there is no way we would actually store all of what we heard, that would just be too much data at any given moment(+all the other senses, +whatever we are thinking about). don't let yourself fool you. you remember something alright, but it's not what you heard. it's a caricature of it.
so no wonder that when you listen to something else a long time after, even if it's actually the same thing and you just don't know it, you will find differences. and they're not differences in the music, they're differences from the way you memorized the data.
because you're not really interested in how stuff are done, you just go with what seems intuitively right to you. sadly intuition as a scientific method is real crap and will make us all get the wrong idea.
making long lasting sessions can in theory have some interests, like trying to find out if something is fatiguing, but even then you will always have too many external factors for the conclusion to actually be valid. did you sleep well? were you in a noisy place all day? did you drink? did your wife complain for 1hour because of something you did 17years ago? unless you remove all that, long listenings are useless because you have no way to only judge sound.
try to read a little on the subject of memory, and then about audio memory, you certainly will find many people telling you that they have better result with long listenings on the forum. I was like that too before I actually went on to do some tests. it's the intuitive conclusion we all come to, we get more and more obvious differences so we believe it's a method that is more revealing. but it's not, it's just so faulty we end up with differences that we made up ourselves, and that why we notice them so well.
but again if you look into it, you will also find out that 100% of actual research suggest short(less than 3S) samples with unnoticeable and as instantaneous as possible switching.
your test with the 2 DAPs is faulty too. you would need to be ignorant of the tracks that are on each DAP else it means nothing. and the color could sadly be enough to have a preference(it was enough for my mother to pick a car model only because the other one didn't have that color, so it surely can push toward a format ^_^), or maybe one will have a little more battery and that will subjectively make you prefer it, or maybe one kind of track takes 0.2second more before it starts when you press play, and while you might not notice consciously, that will annoy you.
it's a bad test and I'm not even talking about the fact that you will always be tempted to pick up the one closest to you and that would likely be the last one you used. just like I use the underwear that are on the top of the pile, not because I prefer them, but because they are the easiest to grab.
and last but not least, did you measure both DAPs to check if they measured the same? the fact is that the DAP doesn't even reach the resolution of the format. so how could it show a difference? the only way would be that the DAP itself is creating some loud noise or distortion while extracting the FLAC. that maybe could be audible, but it then would be the DAP's fault, not the format's.
all that to say that a bad test is a bad test and the result irrelevant.
I actually do something a little like you do, and put the album in 2 formats into the same folder and play it in shuffle mode. then sometimes I will feel like the sound is better or worst and, and I will note it on something I keep for that purpose and check what the file was. I do that for a looooong times with different albums when I get a new DAP, and when I have a great deal of notes I look at the results. so that works with a lot less biases than your way but it's still far from an objective test TBH. still I only once got a clear win and it was for mp3 vs flac where flac sounded better on a fiio X3. I never got to even get a clear win on lossless vs mp3 320, so I let you guess how it went for wave vs flac ^_^. and don't get me wrong, each time I would really think I heard something good enough or bad enough to be pretty confident in myself and note it. so I thought there were differences almost every times. the results just showed otherwise all but one time and with mp3 only(I learned later that the X3 was at the time wrongly extracting the mp3 as someone showed with measurements).
now an objective method: you can take a wave file, play it and record the output in a loop with your soundcard, as long as it measures better than the file itself it's all good, so you might need to limit the test to 16/44. then take a flac of it, and do the same. now you can analyze the differences. with ABX or with some software, both are ok this time.
but flac isn't called bit perfect just for fun. if you have differences they will come from the device, not from the format. that much is a certainty.
I disagree but it's all good... I feel long sessions is the only way to go. I just don't feel short sessions are meaningful. I do agree it's hard to eliminate all the variables but we can try and do our best. What do you think about any of those links or comments from that other forum? anything of value there? I can't get biggie to come back and contribute and I need your guys knowledge since they do some tech talk. Seems like some good stuff to discuss no?
I'm rereading your post since there' alot of info there......I'll get back if I can pick up on anything to discuss. ....but in the meantime, any feelings on those links? are those guys off base?
just to be clear, when i did my 'testing' i was using two sansa players . the same kind same model etc. just diff color. I agree the color could be a problem but you think they have different processing methods or speeds? I'm not quite following that idea but I guess it's possible, I don't know anything about it or how it works, I just assume two sansa DAPs are the same....,maybe not
You are correct. Short samples with direct line level matched A/B comparison are MUCH more accurate than long period listening. Human ears have a very short memory... In the range of a few seconds at most. Go beyond that and anything remotely close sounds the same. Or bias creeps in. Either way it isn't correct.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
