1. 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.

    Dismiss Notice

24bit vs 16bit, the myth exploded!

Discussion in 'Sound Science' started by gregorio, Mar 19, 2009.
343 344 345 346 347 348 349 350 351 352
354 355 356 357
    Lol I want an answer this will admit to me science meets factual observations. the pic is not me but I do own it. In audio life there are those who claim to know and those who do know. Like baseball either can be correct someone is also wrong.
    I agree in theory it should not matter but yet it does in practicality. The major malfunction all or most here have is to admit what they hear and accept it. So while you guys feel I’m wrong you skip past reality into pure science. lastly one should never goof on someone’s possible mental acuity mishaps without knowing the person first hand.
    My avatar is exactly why I use it to fool the real fools.
  2. TheSonicTruth
    I hope by "fools"(your words not mine) you don't include this gentleman: https://ethanwiner.com/

    He will straighten you - and quite a few others on here - out, about a few many things...!
  3. sander99
    @ALRAINBOW: There definitely is a big communication problem. From what you write I do not understand what exactly did you do and what exactly was the result. Could you please answer the questions below.

    Now, do I understand correctly that:
    1. You toke a "real" high quality 96/24 file, let's call this file A?
    2. You created from this file, by downconverting, a 44.1/16 file, let's call this file C?
    3. You created from file C, by upconverting, a new 96/24 file, let's call this file B?
    4. You compared all these 3 files with each other with blind ABX testing?
    And what was the result?
    5. Did you hear a difference between file A and file C?
    6. Did you hear a difference between file A and file B?
    7. Did you hear a difference between file B and file C?

    [Edit: oh, I forgot: did you make sure the levels where exactly the same in all 3 files?]
    Last edited: Dec 3, 2019 at 10:09 AM
  4. gregorio
    So, apparently you're moving on from just making false assertions which contradict the science to even contradicting yourself now! For example:

    "It’s obvious that a file at 24:96 sounds different then a file at 16/44.1." and "mad for what’s asked I don’t hear any change on my end and I did this with three file ... But I can’t get why I would hear a change is the hi res file is down sampled and back up again." - So which is it? Does a 24/96 file obviously/easily sound different to a 16/44 OR were you in fact NOT able to hear a difference "with three files" you have tested and don't understand why you would???
    1. You're joking right? You've responded how many times to this thread and you've only just read the title post? That's nuts!

    2. And again! You said you could NOT hear any change/difference and now you're saying you can hear a difference; "a lower blacker noise floor"! So which is it?

    3. That's absolute nonsense, you definitely accepted your perception is correct and you've even insulted the rest of us for not having the "ears" or "setup" to hear "your perception"!!
    1. Huh? The science "meets" the test you've done (where you couldn't hear any difference) and it "meets" countless other controlled listening tests, the Meyer & Moran published study for example (which has been mentioned previously and let me guess, you haven't bothered to read that either)? The science already meets factual observation, if it didn't, then it wouldn't be the science! How can you not know this, do you know what science is?

    2. How does it matter in practice, you said you couldn't hear any change?

    3. You mean YOUR major malfunction! You said you couldn't hear any change, do you accept what you yourself said or not?

    4. What hypocrisy is this? You're "skipping past" both reality and science and even past what you yourself have said!!!

    As is so often the case, making false assertions here in the science forum and then defending them ad infinitum from a position of ignorance, without any reliable supporting evidence ALWAYS results in self-contradictions and/or making ever more ridiculous or hypocritical claims. The only thing this all demonstrates is their own ignorance, foolishness and inability to think critically. Surely YOU don't want to demonstrate you're ignorant, a fool and can't think logically/critically, do you?

    Last edited: Dec 3, 2019 at 10:29 AM
  5. SoundAndMotion
    @gregorio , take a chill pill!! I don't think you've understood @ALRAINBOW , and therefore you're just arguing with yourself. Do you really think berating him will get clearer responses. Or is your goal to simply frustrate him beyond his limit, so he'll just go away?

    My interpretation is that he says he hears a difference between hi-res and RBCD with files "as provided by ...?". When he downsamples and then upsamples, the difference disappears, which he thinks makes sense because he thinks since it has been upsampled it is the same as the hi-res, i.e. not different anymore. @ALRAINBOW : This is not true!!

    @ALRAINBOW If you have 2 files: Album5Track4_2496.wav and Album5Track4_1644.wav and you are certain you hear a difference (e.g. the first one clearly has a lower blacker noise floor), then I can only think of 3 possibilities:

    1- You really don't hear a difference, but believe you do. It's a type of illusion, just like an optical illusion.

    2- The 2 tracks are supposed to be exactly the same version, other than the first is high res., and the second is RBCD, but(!!) in fact the 2 are different (different masters, or someone did a flawed job of sample rate conversion, or different in some other way).

    3- The 2 tracks really are exactly the same except for the sample rate/bit depth, and you have a special gift of amazing hearing. This would be unusual, but very exciting.

    So, how to distinguish the 3 cases... (did I miss a possibility?)
    A good idea that has already been suggested (and I thought you tried) is use foobar2000 with the foo_ABX plugin. If you need help, check back a few posts, or just ask again.

    If you do the ABX, say 16 times, and the difference is not clearly detected anymore, then it seems to be possibility 1 above. Do you see it differently? If so, why?

    If you clearly keep hearing the difference with the ABX method, then the question is how to tell the difference between 2 and 3 above. One idea, that would require knowing how and taking the time, would be to analyze the 2 files to see that they really are the same versions.

    The simpler method is to do what sander99 wrote above (and others have mentioned in the last few pages): make your own 16/44 version to compare. This will eliminate option 2 (if you do it correctly). Do you know how? Do you have concerns about then upconverting the 16/44 file to a 24/96 file (sander99 's file B above)?
    Last edited: Dec 3, 2019 at 11:00 AM
  6. SoundAndMotion
    If you imagine that 24/96 is kind of like writing a number 98.765432, and writing the "same" number in 16/44 is 98.765 (i.e. downsampling)
    Then upsampling back to 24/96 is like 98.765000 . We do this in case your software/hardware treats 8 digits a little differently from 5 digits.
    Do you see that the upsampled is the same as the downsampled, except for the extra zeros? And they both have less resolution/precision than the original 24/96.
  7. bfreedma
    Whether intentional or somehow unintentional, there is clearly trolling going on.

    When posting history is reviewed, it becomes clear that this pattern has repeated itself over time. Doubtful that anything is going to change based on responding.
  8. SoundAndMotion
    That was an interesting skim... his English used to be better! Good tip.
  9. board
    I've also used some strong language in this debate, which someone asked me to tone down.
    I can include @gregorio here, I love your initial post and have linked to it several times, and I love all the knowledge you have on the subject, but what I'm slowly coming to realize is that for these types of discussions there are two groups of people (with somewhat fluid borders):

    1: People who are completely sure that what they believe in is the truth, and nothing or almost nothing can make them change their mind. This includes audiophiles who believe 24 bits sound better than 16 bits, and it also includes religous people, people who believe in dowsing, zodiacs, etc. and often build their entire lifestyle around this belief (going to church every Sunday/buying more products for their stereo systems, etc.).

    2: People who believe in science, proof, and experiments, and who are happy to read scientific papers, listen to scientists explain certain issues, and they are happy to participate in experiments, such as blind tests, and throughout life they change their mind about quite a bit of topics. They also often, foolishly, think that you can reach people from group 1 with rational arguments and debate.

    So the problem is when people from group 2 think they can convince people from group 1. They can't. We can't. We are from group 2, and Alrainbow is from group 1.
    We've been arguing with him for a while now. Before him it was Michael Fremer we argued with. Before that it was someone else.

    Unfortunately, some of us become a bit upset when the people from group 1 don't agree to do a simple experiment (blind test). Someone like Michael Fremer also refuses to do any kind of ABX tests ever again, because on that fateful day in 1991 Stanley Lipshitz cheated him out of his victory, and now Fremer is bitter (not that he wasn't before, but that's a different topic).
    So the discussion is going nowhere, and people from group 1 will never, ever change their mind about their beliefs, unless they were already close to the border to group 2 to begin with. This is something about personality, not that we haven't presented the right argument or method, etc.
    It should also be said that people who believe in science can actually belong to group 1.They have the attitude that we already know everything there is to know.
    Hifiearspeakers likes this.
  10. bigshot
    Scientific proof of the Dunning-Kruger effect.

    I'm interested to see how low the game of limbo needs to go to get down to the right level.
    Last edited: Dec 3, 2019 at 1:23 PM
  11. Daiyama
  12. bigshot
    Download their 24/96 file and knock it down to 16/44.1 yourself. Then bounce it back up to 24/96 and level match.

    I wouldn't trust an audiophile record label to prepare the lower resolution file properly.
    Daiyama likes this.
  13. old tech
    Tell me more... I recall he piked out on James Randi challenge on cables but haven't heard of this one.
  14. miksu8
    If you quote someone please don't change the original message. I just asked how often 24bit music sounds better because it's produced better. It's reasonable question, since if someone has a better quality version of a track, it's probably sold as 24 bit version for marketing reasons.

    Your answer didn't make any sense to me, sorry.
  15. gregorio
    1. I think I have understood him, as my interpretation is similar to yours (see point #2) but even if I haven't, it doesn't make a substantial difference because ...
    1a. No, I don't think berating him will get clearer responses, we're past the point of clearer responses. Even if his responses become a little clearer as far as his use of language is concerned, they'll become less clear as far as a factual basis for his arguments and rational thought/critical thinking are concerned. And, neither is my primary goal to frustrate him so he'll just go away. My aim was to highlight the self-contradiction, hypocrisy and sheer ridiculousness of his responses. This might get him to evaluate his future posts more carefully (my primary goal) but probably not. He'll probably just ignore it completely, brush it off as me twisting his words or invent some other false excuse and just carry on regardless. There's also the possibility that others reading this thread might think that underneath the poor use of language, Alrainbow could have a rational/valid point. Highlighting Alrainbow's self-contradiction, irrationality (etc.) might make it easier for them to realise that he doesn't (my secondary goal). And lastly, as he descends ever deeper into his logical black-hole, he'll sooner or later reach a point where even he realises how ridiculous he's being or at least, that his arguments are futile. At which point he'll either leave the thread due to embarrassment or frustration or, there's a very small chance he might re-evaluate. Whichever it is though, his (deliberate or inadvertent) trolling/thread cr@pping will cease and the (third) "goal" of my post was to reach that point sooner rather than later!

    2. That's my interpretation too but that just brings us back to my (and others') very first response to him, that he's falsely conflating different masters with different container formats. So, we're now going round in a circle of his making!
    2a. No, he doesn't think they're the same, he specifically stated: "The 24/96 file is not the same as 16/44.1 but it’s not as much as downloaded files in two formats" [emphasis mine] - Clearly he's talking about the file he has converted rather than different masters and clearly he has somehow analysed the difference. Also clearly: If he thinks that upsampling the 16/44 will restore everything that downsampling removed, then firstly that's already been refuted and secondly, he can easily find out for himself by making exactly the same comparative analysis between his downsampled 16/44 and it's upsampled version as he did between his 16/44 version and the "downloaded files". Whichever way you look at his assertions, they are self-contradictory/nonsensical!

    I'm not sure that is the problem. I'm obviously a group 2 person and I DON'T think I can convince people from group 1 or rather, I think there is only a tiny chance I can convince them. The problem is sort of the reverse, group 1 people come here, to a group 2 (science/fact based) forum and try to convince us with IRRATIONAL arguments and debate. That approach can never work here, so why do they try it? It indicates an inability to differentiate the rational from the irrational, ignorance not only of the science specific to audio but of what science is and why it exists, arrogance and rudeness in trying such an approach in an actual science/fact based forum, hypocrisy in that they'll happily employ science when it suits them and, when applied to areas beyond/outside of audiophilia, increasingly represents an existential threat to our entire species (and many others)!

    Last edited: Dec 4, 2019 at 5:32 AM
343 344 345 346 347 348 349 350 351 352
354 355 356 357

Share This Page