Michael
Did you read the report from that English musician that I sent you ?
I know it proves nothing, but he also commented that when he took the earlier DL files to work, he was annoyed that his engineers were unable to hear the differences that he could hear. He later reported that on a recent tour to Germany, much to his surprise, the engineers there had no problem hearing differences. One of those engineers even gave him a small portable Class A headphone amplifier, as he now much prefers Class A Amplifiers, when using the AKG K701 headphones.
Alex
Did you read the report from that English musician that I sent you ?
I know it proves nothing, but he also commented that when he took the earlier DL files to work, he was annoyed that his engineers were unable to hear the differences that he could hear. He later reported that on a recent tour to Germany, much to his surprise, the engineers there had no problem hearing differences. One of those engineers even gave him a small portable Class A headphone amplifier, as he now much prefers Class A Amplifiers, when using the AKG K701 headphones.
Alex
Oh, yes - I almost forgot.
I did multiple checksums on the files using different methods and different softwares. Always the same result, identical. When I opened one file in a hex editor and changed just one bit, no checksum match. So I am confident that the checksum works.
And I ran the 2 files thru the excellent Audio DiffMaker software. It found a -300dB null between the 2 files. The difference file contained nothing.
So IF there is a difference in these files, it does not lie anywhere that normal software can read it. No software I used found the tiniest bit of difference. Therefore I find it hard to believe that a "mere mortal" software player like Foobar, Winamp, Media Player, iTunes, etc would read any difference. Why/how could they?
Does anyone have a theory about how a music player could read differences in audio files that no other software can? Anyone?
I did multiple checksums on the files using different methods and different softwares. Always the same result, identical. When I opened one file in a hex editor and changed just one bit, no checksum match. So I am confident that the checksum works.
And I ran the 2 files thru the excellent Audio DiffMaker software. It found a -300dB null between the 2 files. The difference file contained nothing.
So IF there is a difference in these files, it does not lie anywhere that normal software can read it. No software I used found the tiniest bit of difference. Therefore I find it hard to believe that a "mere mortal" software player like Foobar, Winamp, Media Player, iTunes, etc would read any difference. Why/how could they?
Does anyone have a theory about how a music player could read differences in audio files that no other software can? Anyone?
Perhaps you might want to do just that, play a copy of the same file from flash drive as compared to hard disk drive playback.
It's easy enough to hypothesize that a hard disk drive pulls different amounts of current when seeking and reading vs. "idling" with the disk just spinning, and this could affect the power rails which also power the D/A converter, and this could conceivably affect the sound.
A flash memory device, running on lower power, would not have as great an impact on power rails, even though it surely gets read much the same way, in short bursts that get buffered into RAM and then clocked into the D/A converter.
I hear a difference , we know it is there, err I'm saying the world is not flat .. so tell us why a ripped and burned CD sound different..
Human error.
No. Seriously. Unless you've some well crafted, double-blind AB methodology on which to base your claims, "sounds different" is hogwash.
Instead of sitting on your rear end, why don't you just try it for yourself. The appropriate software is free and it doesn't take more than 20-30 minutes to read and burn a copy that you can compare for yourself. The differences in many cases aren't too subtle and there are many that have done it with satisfactory results. If it works for you great, if it doesn't then you have some experience to go on. Either way you're ahead of where you were before.
I've explained why you can expect a difference in previous posts in this thread, so I'm not going to keep going over (and over and...) what I've stated before.
Best Regards,
TerryO
Do you think there is anything else in an audio file that the serie of 0 and 1s ??
-- If you don't then it is just not possible. Every beginner ingeneer in computer knows that it is easy since dozens of years to transfer huge amount of data without a single bit loss over any device you like.
I transfer routinely (every day) gigabytes of data from Hawaii to France without a single bit error in years.
-- If you think there is more than the serie of 0 and 1s, please explain what.
I did, of course. I even let the hard drive spin down for flash drive playback. No real difference. But I'm willing to try that one again. In normal operation my music server pulls from a remote HDD then buffers into RAM, so it is effectively like reading off a flash drive.
So still no answers or theories about where the extra or different information may be in the files. Anyone?
You might as well give up on this one, jrn. I've already politely suggested to the guy that he conducts ABX tests and posts the results for all to see. The result of this suggestion was being laughed at and effectively told that he knows there's an audible difference so the tests would be pointless.
It's a real shame that so many claiming to hear a difference won't join in and play fairly, as there's a lot to be learned from such testing that could benefit not only themselves, but the whole diyAudio community. Placebo is an enormously powerful effect, and I can't help but feel that it's blinding many around here to the truth.
I would offer to write a guide on using Foobar2000 with the ABX comparator tool if I thought it'd be any help, but I have a strong feeling that I'd only be wasting my time. 🙁
Please do write it! It might become a sticky or even an article here on the forum.
That would be great.
That would be great.
Indeed, there are a lot of members, including me, who take a rational and logical approach to getting good sound (trusting our ears, not just any stupid claim someone makes on the Internet) who would greatly appreciate that guide!
If only that were true. Errors are exceedingly common. Computers use error-checking at many stages and correct errors all the time. The result is even though the output can be bit-perfect to the input almost all of the time, there were errors being made in the meantime.
RAM has errors; they are mapped out to the point where they can't be mapped out any more. Then data corruption happens and you replace the RAM. Same with HDs, same with plenty of stages. How many people are aware that UV light and certain radiation from space slowly destroys RAM and Flash Memory over time? The chip makers always spec it; it's not some phantom, unknown phenomena. Early RAM chips (and some server-class memory today) have parity chips (the "9th chip") which basically just did constant checksumming to find errors ... they most certainly do happen often.
Flash drives rotate the data over the entire chip because each read and write increases the likelihood that that area of the chip will fail. There was a significant improvement in reliability of flash memory when the manufacturers, beginning about 10 years ago, began writing routines to manage the number of rewrites on a given area. This is why today they are reliable enough to even be considered as a replacement for magnetic hard drives. They had to re-think how and where data was stored to make this happen ... simply re-writing the same area, as HD's had done since forever, dramatically increased the failure rate to the point where magnetic storage was far more reliable. It took a change in how drives stored data to make flash memory a viable alternative to magnetic drives.
RedBook CDs actually have robust error correction (for example, data is written to multiple areas of the disk multiple times) and providing the redundant data can be read in time, no error on output. If data always made it through the various electronic paths and processes unscathed, Intel would not need a pipeline flush routine in it's processors. There are a host of examples; essentially it's everywhere in both hardware and software.
The result we want ... bit for bit input and output, does happen virtually (but not perfectly) all the time. But, it's wrong to think the bits moving around in the various digital devices we use everyday are error free. This simply is not the case in reality. It's rather a testament to the robustness of the error-correction routines and methods used in all things digital that the output matches the input, or the answer is the expected answer, despite the errors.
Last edited:
No, they don't. Fault-tolerant computers are quite a rarity, and not to be found on or under the average desktop. They are not too hard to build, just not economical.
Memories have parity bits. Yes, they can flag errors (up to a certain extent). But there is zero scope for correction once the error has occured.
Ah, the argument goes on. I think both sides are too hung up on the 1's and 0's issue. So two CD's with identical checksums sound different? The issue should be more like as they play at 1x what do the two data streams look like? (timing and accuracy), at least for a start. Unfortunately the subjective camp has no interest in engineering questions or answers just equipment "rolling" and the endless "dramatic" differences in uncontrolled listening. I also see the two camps still can't sit in the same room. Imagine SandyK and SY listening to "Kind of Blue" through a USB stick vs SD card...
So basically you agree with me, other than you want to point out that errors do occur from time to time, but the statistics of error correction, which is built into the system, ensure that they reach the user with sufficient rarity that the entire world's financial accounting is done on computers.
Do you really imagine that I don't know that?
Werner's point is valid too. It's all just nit-picking and people trying to impress one another with how much they know. So what?
Scott, you've missed the point. The issue is whether there's any difference between identical files depending on whether they were ripped to a silicon memory as opposed to a rotating hard disk.
More to the point, where's the offer by SandyK to distinguish between files with identical content under controlled conditions?
That's what this is all about.
Somebody says they can do this.
OK, do it. Otherwise what IS the point of this thread?
w
Actually Waki there are TWO points.
1) That identical checksum files can sound different depending on the medium they were ripped to.
2) That playback of identical files can sound different if replayed from different moving media. (CD, CD-ROM, HDD, DVD, Blu-Ray)
#2 I can believe. #1 makes no sense to me.
1) That identical checksum files can sound different depending on the medium they were ripped to.
2) That playback of identical files can sound different if replayed from different moving media. (CD, CD-ROM, HDD, DVD, Blu-Ray)
#2 I can believe. #1 makes no sense to me.
From what I've seen lots of A/B's have been mentioned, different brands of blank CD (frequently), HD, flash card, USB stick. The question I ask is on the same machine what does the datastream going to the DAC look like? This would take into account all the claims of making the hardware "work" harder or differently (IMHO mostly bogus these days).
Data integrity in the static sense is a non-issue.
And as I already said, you also want to look at the "power stream" to the DAC.
Nobody would dispute that.
1)It's extremely unlikely that anyone has a system that would permit the comparison of different moving media where the signal path is identical. The signal path could contribute differences irrespective of the bitstream coming off the media.
2)Even I don't insist that even identical players sound identical. I just insist that the differences are trivial.
The problem is that some people believe in your point #1. The fact that they believe this creates a real problem for ALL of us. People write in with reports of stuff that I know to be impossible. They want to ARGUE about it, and they even want to shut people up because they won't accept the impossible.
There has to be some kind of conclusion to be arrived at. Otherwise the dam will break, and we'll have to say goodbye to evidence based electronics. Which will not suit these individuals, even though they have not the wit to realise it. Who really wants a forum where they can come on and write whatever they choose regardless of evidence? Nobody will come here if the replies degenerate to the point of just being random.
w
- Status
- Not open for further replies.