|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
65.19.76.104
In Reply to: RE: Suggestion........ posted by jedrider on July 13, 2014 at 08:03:05
Exact audio copy allows comparing two WAV files for equality. This will work providing there are no offset errors. If there are offset errors (different number of empty samples at the start of the two files) then EAC won't necessarily give a complete picture of what is going on. In which case, a better approach is to use an audio editor to look at the individual samples and line them up. Then you can mix them together with one of the two out of polarity to get a difference file. This should be all zeros, but if there are errors you will be able to find them. I assume this can be done with free editors such as Audacity. I have been using Soundforge Pro 10 for a number of years and do these comparisons from time to time.
You should be aware that offset errors and changes to headers have been reported as causing audible differences. According to digital theory, if the samples are the same there should be no audible difference due to changes in headers, MP3 tags, file names, etc., but theory is nearly always trumped by practice. Changes in headers but not samples can produce audible differences, at least on some systems that are sensitive to things that, in an ideal world, they should not be sensitive to.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Follow Ups:
"Changes in headers but not samples can produce audible differences, at least on some systems that are sensitive to things that, in an ideal world, they should not be sensitive to."I have wondered about that before, but don't recall seeing anybody discuss it, so just assumed it wasn't important and just one more thing to not be concerned with.
Do you have any more discussion of this, offhand? Don't go to a lot of trouble as I will look for it myself, but if you recall anything (link, etc.)...thanks.
I do my audio-file assembling on a machine that's remote from the actual file player, so all the player sees is the .wav data *plus* the ethernet packaging headers. They're still headers though, in the sense of data that isn't representing audio. I don't think this is the type of thing you were referring to, as the data extraction scenario is different since the player app doesn't even "see" the header.
Edit: I am specifically speaking of my SB Touch setup here, not the BDP1/BDA1 setup which *could* be susceptible to the exact header thing you're speaking of.
Edits: 07/13/14
It seems plausible that what is going on in the computer will have less effect on the sound if the computer is physically and electrically very distant from any of the analog equipment. But then, a more complex playback chain will have more complex failure modes. I suggest somehow capturing the bits that are going to the DAC and verifying that they correspond to the bits on the CD. If you use SPDIF to the DAC you can connect the cable to a digital in of a separate computer and record the stream. There may be various "loop back" tricks available that don't require extra hardware, but these won't catch all the possible places where the bits could get corrupted. IMO it is a complete waste of time to do anything regarding subjective sound quality of computer audio until one has first verified that the correct bits are actually making it to the DAC.
There have been threads regarding the effect of FLAC to WAV conversion and impact on headers. There have also been threads discussing the effect of offset errors on digital sound files. There have also been threads on how two bit-identical files might sound different.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Unfortunately I don't have a computer with a general (audio) digital input, except USB. So I'd have to get some interface gadget and use USB somehow.
I didn't realize those topics you mentioned at the end there were the same thing as the header "problem". FWIW I don't notice a sonic diff between WAVs and FLACs, and the diff would have to be get significant before I'd be willing to give up the tagging flexibility/convenience of FLACs. Compared to the playback limitations of tagged WAVs, not worth it to me.
I really think that just a plain old hex editor with search mode will be sufficient for a basic test. Just search on a pattern given some offset and see if you can locate the similar pattern on the next bitstream at the same location. Trying to CAPTURE SPDIF info would be trickier IMO as the interface may always be transmitting SOMETHING such as some control signal to just wait for the sample to arrive.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: