I like to have discussions like these, but right now I am rather annoyed with a few people here.
The claim was made that two identical files which are bit-for-bit the same can sound different. OK, I and others disagree. Discussion begins.
But now a few people start to use arguments like: "Well there are differences between CD production method A and CD production method B." or "why do burned CD's sound different?" Or: "Why does playback hardware/software X sound different from Y if the files are the same?"
These are all irrelevant from the discussion at hand. So please try to focus and stay on target.
Explain this: What exactly gets stored besides the PCM data that will account for the proposed audible differences between two identical files?
If the playback chain is the same, and the files are identical, but you hear a difference, what causes that? What information gets recorded on the SSD or HDD or DLT or whatever besides the bits and bytes?
It cannot be PSU noise during ripping, it cannot be timing/jitter during ripping, these things are not something that ends up in the file on the medium.
So explain, without deviating into bluespec or burned CD's or linear PSU's for USB flash drives without explaining what exactly causes the difference when storing the information, and how exactly that difference gets picked up again when the information is read out again.
Or are you suggesting a bit can have more than two states perhaps?
The claim was made that two identical files which are bit-for-bit the same can sound different. OK, I and others disagree. Discussion begins.
But now a few people start to use arguments like: "Well there are differences between CD production method A and CD production method B." or "why do burned CD's sound different?" Or: "Why does playback hardware/software X sound different from Y if the files are the same?"
These are all irrelevant from the discussion at hand. So please try to focus and stay on target.
Explain this: What exactly gets stored besides the PCM data that will account for the proposed audible differences between two identical files?
If the playback chain is the same, and the files are identical, but you hear a difference, what causes that? What information gets recorded on the SSD or HDD or DLT or whatever besides the bits and bytes?
It cannot be PSU noise during ripping, it cannot be timing/jitter during ripping, these things are not something that ends up in the file on the medium.
So explain, without deviating into bluespec or burned CD's or linear PSU's for USB flash drives without explaining what exactly causes the difference when storing the information, and how exactly that difference gets picked up again when the information is read out again.
Or are you suggesting a bit can have more than two states perhaps?
Or are you suggesting a bit can have more than two states perhaps?
I would say if the differences are heard on the same hardware, then the hardware needs to be revisited (e.g. jitter, power supply etc).
Like for example a "typical" PC power supply - I had my moments of surprise testing a circuit powered by one of these, monitoring the +/- 12V supply lines on the scope when I accidentally turned my halogen desk lamp off
It's actually an equation with unknown number of unknowns.
That's not the kind of difference these guys mean. They mean that one track will be consistently and repeatably audibly superior to the other.
w
Alex, you did send some bitwise identical files to me, I chose not to listen rather than 'refused', which makes it sound like I put up strong resistance
The reason I chose not to listen is quite simple - if I heard a difference then I'd put it down to some uncontrolled variable. For example, was my upstairs neighbour making a cellphone call at the time? So then I'd have to listen again and again at various times of the day and night in an attempt to average out all possible uncontrolled variables.Assuming I still heard a difference which was repeatable and correlated only with the two otherwise identical files I'd then be at a loss to explain its origin. I would not even know where to start in formulating a hypothesis as to why. So I'd learn nothing and continue on in my own merry way none the wiser. If you'd like to persuade people that you hear a difference then the way to do that would be to describe in fairly meticulous detail the experimental setups where you do hear the difference. Also describe the nature of the difference that you hear and offer a potential candidate as a hypothesis for why. With that as a basis, some might be inclined to try the experiment for themselves. That might give some hope of progress, otherwise we'll just be stuck in the same old 'he said, she said' routine which I prefer to stay well clear of.
Yup, here we go again.
Nope, not going to happen.
You can't expect those arguing in favor of non-logic to argue logically. That would be inconsistent.
The claim that he can hear a difference is easy to test. Just rename one of the files and email it back to him. If he can tell you which one it is, would that convince you? I'd be impressed.
Maybe you and/or Pano and/or whoever else has the files could repeat this a few times for statistical significance.
As I see it, the three most likely outcomes are:
a) He can consistently tell which file is which.
b) Results are random.
c) Nobody who claims to be able to hear the difference is willing to take part.
Alex didn't actually send me the files, he sent links to them. They're a bit big to email.
You don't quite seem to understand me. I am convinced already he hears a difference, it could though be the very well known placebo effect. I'm not yet convinced I'd hear any difference.
Yeah, maybe I could, I'd need persuading its a worthwhile use of my time though. You're welcome to try
Several Rock Grotto members did just that.
3 files were uploaded using Filemail. The 3 participants (Sydney, Brisbane and U.K.) could hear clear differences and identify which track was which, when downloaded directly from the Filemail links, although 2 were harder to identify. However, the resident EE sceptic from The Netherlands then moved the .wav files around on his PC during renaming, and sent the files back again.
It appears that the files may have suffered some degradation while being moved around between HDDs during renaming.
This PC had no alterations to it that would normally be made for a PC used for multimedia or Audio.
None of the 3 participants were then able to say with any certainty which file was which originally, as none of them sounded as "clean" as the originals.
The .wav file that I regarded as the better sounding of those originally uploaded was however picked by all 3 to be the closest sounding to the original best choice.
SandyK
Do you think the degradation of sound quality was due to the renaming of the files, or the fact that they were temporarily stored on a PC that's not normally used for audio?
I suspect the former as the Filemail servers apparently had no adverse effect and I doubt they're specifically set up for multimedia or audio.
I suspect the former as the Filemail servers apparently had no adverse effect and I doubt they're specifically set up for multimedia or audio.
I suspect that the degradation of sound quality had nothing much to do with the files themselves and pretty much everything to do with the participants' conscious and subconscious ideas of the files. But that's just me, YMMV.
Godfrey
I have found that .wav files appear to suffer a minor degradation when moved around between spinning HDDs.
Yet, I do not notice a similar degradation with commercial 24/96 .flac files when moved around on different media, including burning to a DVD then decoding again. There appears to be some difference in the structure of .flac files.
" Or simply don't bother posting again."
Wakibaki
I feel certain that I speak for the many members who post subjective reports ,
when I say that we refuse to be intimidated into silence by the likes of arrogant EEs like yourself.
SandyK
Last edited:
You won't be intimidated, SandyK, you will be shamed.
The big problem with what you have done, SandyK, you and Terry O in particular, is that when your claims come tumbling down as this one is, the rest of your credibility goes with it. When you come out with your next claim, people will just go:- 'Oh, yeah, those were the guys who could hear differences in checksummed files...'
Remember Peter and the Wolf?
Prokofiev.
The boy who cried wolf?
w
The big problem with what you have done, SandyK, you and Terry O in particular, is that when your claims come tumbling down as this one is, the rest of your credibility goes with it. When you come out with your next claim, people will just go:- 'Oh, yeah, those were the guys who could hear differences in checksummed files...'
Remember Peter and the Wolf?
Prokofiev.
The boy who cried wolf?
w
Well, read through this thread and seei if my subjective findings were imagined. BTW, the thread was started by AKSA, who was aware of these findings, and had used an additional diode in the normally tied C and B of the current mirror transistor with improved results in one of his commercial amplifiers.
http://www.diyaudio.com/forums/solid-state/133018-current-mirror-discussion-15.html
If the file name was part of the played data (it isn't) it would make a difference. What is a "PC that's not normally used for audio"? Does a computer 'know' what the data is for? Why? How?
I'll grant you a checksum CAN be in error but the larger the checksum, the more difficult to make a mistake. A successful full file comparison of all bytes would have to sound identical - at least on the same hardware. I don't dispute that the hardware can sound different but I have a hard time with files. I.E. two identical files with only different names will sound identical. I consider that as reliable as gravity holding me on the Earth.
G²
Flattering as it is to be the recipient of such a concentration of attention on your part, abraxilito, it really is OT. It does, however, relieve me of the necessity of quoting all my qualifications in quite so much detail.
I'm a programmer and mixed-signals designer, anybody who doubts this only needs to search through my posts. I can design a SDR.
If you haven't seen any of my other screenshots, here's a screenshot of a satellite link budget calculator I wrote a few years ago.
13 years actually, doesn't time fly.
I regularly test my hearing. ~15k one ear, 16 the other. No gaps.
I've been a musician for over 40 years.
I pointed out quite a long way back in this thread that I can tune a stringed instrument. I have trained hearing.
Now, I'm not going to put up my hearing against even a randomly chosen 18-year-old and expect to come out a winner. The physiological effects of age are well documented.
I do notice that none of the claimants here can boast a similar achievement, however, or if they can, they didn't think it worth mentioning.
CD players are so good that most people can't tell the difference between them, and if there are detectable differences, then they're trivial in terms of enjoying the music.
I'm not saying that all CD players have an identical sound, but, what I am saying is: in my experience the differences are so small they're not worth getting exercised over. Unless you get a faulty one. In which case you just get another one.
Now, of course, this is just my opinion, but I am an expert musician, I have trained ears, recently tested, and I understand the technology backwards. I've taught music to numerous people. Against that we're setting the opinions of a bunch of people who thus far have shown that they neither understand the technology nor do they know the meaning of evidence. It certainly doesn't inspire confidence in their judgement when they claim to hear differences in circumstances where differences just can't exist.
There's so much stuff posted on here that proclaims the advantages of one tweak or another, I think it's important that people trying to weigh up the evidence should have something to put in the other pan of the scales.
w
I'm a programmer and mixed-signals designer, anybody who doubts this only needs to search through my posts. I can design a SDR.
If you haven't seen any of my other screenshots, here's a screenshot of a satellite link budget calculator I wrote a few years ago.
13 years actually, doesn't time fly.
I regularly test my hearing. ~15k one ear, 16 the other. No gaps.
I've been a musician for over 40 years.
I pointed out quite a long way back in this thread that I can tune a stringed instrument. I have trained hearing.
Now, I'm not going to put up my hearing against even a randomly chosen 18-year-old and expect to come out a winner. The physiological effects of age are well documented.
I do notice that none of the claimants here can boast a similar achievement, however, or if they can, they didn't think it worth mentioning.
CD players are so good that most people can't tell the difference between them, and if there are detectable differences, then they're trivial in terms of enjoying the music.
I'm not saying that all CD players have an identical sound, but, what I am saying is: in my experience the differences are so small they're not worth getting exercised over. Unless you get a faulty one. In which case you just get another one.
Now, of course, this is just my opinion, but I am an expert musician, I have trained ears, recently tested, and I understand the technology backwards. I've taught music to numerous people. Against that we're setting the opinions of a bunch of people who thus far have shown that they neither understand the technology nor do they know the meaning of evidence. It certainly doesn't inspire confidence in their judgement when they claim to hear differences in circumstances where differences just can't exist.
There's so much stuff posted on here that proclaims the advantages of one tweak or another, I think it's important that people trying to weigh up the evidence should have something to put in the other pan of the scales.
w
Most of this goes off topic so I'll just confine my responses to those where there's some semblance of relevance to the current discussion.
Most people can't I agree. However this is not a board for most people, its 'by fanatics, for fanatics' to quote the header at the top. So whilst your claim is not disputed, its irrelevant. To me, its no trivial thing that my DAC sounds better than my $15 DVD player, that's the reason for tweaking.
It would be churlish to dispute your experience. However everyone's experience is a fairly unique thing, individual. You've posted up your equipment in the past, to my way of thinking you're not really concerned about sound quality and your equipment list reflects that. You're on a board where the differences do matter to many, I suggest you learn to live with that.
Ironic then that you put up no evidence to support such a claim, no?
Yep, what's sauce for the goose is also relish for the gander. To quote Gandhi 'Be the change you wish to see'.
Most people can't I agree. However this is not a board for most people, its 'by fanatics, for fanatics' to quote the header at the top. So whilst your claim is not disputed, its irrelevant. To me, its no trivial thing that my DAC sounds better than my $15 DVD player, that's the reason for tweaking.
It would be churlish to dispute your experience. However everyone's experience is a fairly unique thing, individual. You've posted up your equipment in the past, to my way of thinking you're not really concerned about sound quality and your equipment list reflects that. You're on a board where the differences do matter to many, I suggest you learn to live with that.
Ironic then that you put up no evidence to support such a claim, no?
Yep, what's sauce for the goose is also relish for the gander. To quote Gandhi 'Be the change you wish to see'.
Yes, this does need to be split off. I'll work on it tomorrow or ask another mod to.
FWIW, I did listen to the 2 different files. I "thought" that I heard a difference, but could not reliably tell which was which. I also put the same file in repeat and "thought" I heard differences. And I renamed a copy of the file and also "thought" I heard differences, but could not reliably identify one or the other. BTW, the two downloaded files had the same checksum. I found this after listening.
So my guess is that my imagination was doing most of the work, here.
Note: I have not seen TerryO claim that identical files of the same medium sound different. So far that I have read, his claims of difference all relate to playback off different media.
FWIW, I did listen to the 2 different files. I "thought" that I heard a difference, but could not reliably tell which was which. I also put the same file in repeat and "thought" I heard differences. And I renamed a copy of the file and also "thought" I heard differences, but could not reliably identify one or the other. BTW, the two downloaded files had the same checksum. I found this after listening.
So my guess is that my imagination was doing most of the work, here.
Note: I have not seen TerryO claim that identical files of the same medium sound different. So far that I have read, his claims of difference all relate to playback off different media.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.