[mrksgrn]

As the title says, I have one CD that I ripped in EAC twice to extract 2 copies of the WAV CD image. In splitting and coverting these 2 WAV files, the resulting 2 FLAC files of the same track have small (<10kbps) bitrate differences.
 
For the record, EAC stated that the rip quality was 100.0% for both times.
 
Can anyone explain this bitrate difference and its implication, if any? I had assumed that a 100% quality rip entailed perfect CD quality with no variation. 
 
 

 

[El_Doug]

I can't speak to whether the 2 rips differed at all, though I'd highly doubt any difference that may exist would be significant at all (aka 1 bit in a lot)
 
However, the FLAC encode can be different.  Perhaps while you performed one compression, your antivirus software kicked in, or your processor was slightly warmer than normal.  With the CPU not working as efficiently, then at the same level of encoding, the compression could be SLIGHTLY less efficient at finding better patterns. 
 
Just a guess, but it seems logically consistent

 

[deadlylover]

If you use foobar, try using the bit comparator to see if there are truly any differences.

 

[mrksgrn]

Interesting. bit comparator says no differences found despite a 6kbps quality difference. I know this amount may be insignificant, but I find it absolutely strange that this translate to no difference at all. What does this say, then, about FLAC bitrate in general?
 
Quote:

If you use foobar, try using the bit comparator to see if there are truly any differences.

 

 

[Roseval]

If I understand correctly you have done
CDx > rip > WAV1 > Split > FLAC1
CDx > rip > WAV2 > Split > FLAC2
 
Are the WAVs bit identical?
What tool did you use to convert and split to FLAC?

 

[mrksgrn]

Ripped using EAC. 
 
Quote:

If I understand correctly you have done
CDx > rip > WAV1 > Split > FLAC1
CDx > rip > WAV2 > Split > FLAC2
 
Are the WAVs bit identical?
What tool did you use to convert and split to FLAC?

 

 

[Roseval]

As El_Doug suggested it might be related to CPU load.
All I know is that the compression level in FLAC means the amount of effort FLAC might spend to find the best possible compression.
If this is a limit in time, then it is possible that the efficiency of the compression is system load dependent.
 
An interesting experiment would be:

1. Convert a WAV to FLAC on a nearly inactive system
2. Convert the same WAV on a heavy loaded system, play a video, do a backup, some gaming, etc so lots of CPU and I/O

Wonder if this will make a difference.

 

[Stereodude]

Quote:

As El_Doug suggested it might be related to CPU load.
All I know is that the compression level in FLAC means the amount of effort FLAC might spend to find the best possible compression.
If this is a limit in time, then it is possible that the efficiency of the compression is system load dependent.
 
An interesting experiment would be:

1. Convert a WAV to FLAC on a nearly inactive system
2. Convert the same WAV on a heavy loaded system, play a video, do a backup, some gaming, etc so lots of CPU and I/O

Wonder if this will make a difference.

That's absurd.  It will just take longer to encode if the CPU is busy doing something else.  FLAC encoding is deterministic.  The bitrate won't change if the encoding settings are the same and the input was identical.
 
The OP should use the accuraterip feature of EAC + a test & copy to ensure the rips are correct

 

[sterling1]

I ripped Journey's "Don't Stop Believin'" to ALAC , once with iTunes and once with DBpoweramp. The rip encoded with iTunes is 900kbps. The rip encoded with DBpoweramp converter is 901kbps. Each of these rips sounds identical played with iTunes media player, or played on foobar. I don't know what the addition or absence of one kbps means.

 

[deadlylover]

Quote:

That's absurd.  It will just take longer to encode if the CPU is busy doing something else.  FLAC encoding is deterministic.  The bitrate won't change if the encoding settings are the same and the input was identical.

I agree.
 
My guess is that the encoder was having a bad day, which led to a less than optimal compression ratio.

And as from the bit comparison, this difference in compression had absolutely no effect on the sound quality.
 
Perhaps one was encoded in level 7 compared to level 8, or something like that.

 

[mrksgrn]

In my case, both were level 8 FLAC. But I will try the test again.
Quote:

I agree.
 
My guess is that the encoder was having a bad day, which led to a less than optimal compression ratio.

And as from the bit comparison, this difference in compression had absolutely no effect on the sound quality.
 
Perhaps one was encoded in level 7 compared to level 8, or something like that.

 

 

[sterling1]

Maybe it's about the silent groove, that's to say, spacing between selections.

 

[Todd R]

Quote:

Maybe it's about the silent groove, that's to say, spacing between selections.

That's the first thing that came to my mind when I read this. Are the track lengths starting and stopping at exactly the same times? 

 

[Stereodude]

Quote:

In my case, both were level 8 FLAC. But I will try the test again.

Was the tagging and album art identical?