Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Overall which is the better Encode to keep? (Read 3675 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Overall which is the better Encode to keep?

While attempting  to cleanup & organize a friends mp3 collection I have come across quite a few duplicates. 'senility '
Using Encspot  as 1 source there were a few LAME versions as well as Xing, Blade (eeek!) & FhG.

With hopes of trying to keep this simple and under a 1,000 scenarios. Lets say the source use was the same for all,
no 'tweak' settings, same bitrate (for the most part) if cbr,any & all vbr using LAME and the music genre being mixed.

Am I correct in thinking:
Toss any Blade, LAME > Xing, FhG > Xing, FhG (high cbr) > LAME (cbr 128), FhG (high cbr) <tossup> LAME (high cbr),
LAME (vbr) > FhG, LAME (higher ver) > LAME (lower ver).

Though I remember reading but seemingly can't find which version NOT to use (at the time) as was problematic with vbr.
I gather thats one area where my 'higher the' version theory is wrong. IIRC files were mainly version 3.87 and higher,
though a few 3.83 or 3.80's might have been in there.

Now after being (mis)educated    by myself , he uses lossless or the preset  settings.

Regards...

Overall which is the better Encode to keep?

Reply #1 –
foobar2000 being able to gaplessly play most LAME >=3.90 mp3s would be enough for me to choose them over any other encode, except in the "FhG (high cbr) vs LAME (cbr 128)" where I would resort to ABC/HR tests to check if the FhG's are noticeably better.

Just a thought.

Overall which is the better Encode to keep?

Reply #2 –
"LAME" alone isn't a seal of quality. There are so many settings and compiles that produce bad quality files, so if they aren't preset encodes I'd be a little wary.

Overall which is the better Encode to keep?

Reply #3 –
Hmm... No real definitive answer. I was afraid of that.
Well of the material he still has the source for (CD),  I've decided to re-rip w/3.90 or latest using presets. 
For the things the source is still in playable but questionable condition (it's days are numbered)
like vinyl,reel-2-reel,4 track tape & the like, figured lossless be the best.
Which also leads me to a semi-related topic but I'll see if I can figure it out (best way to encode the 4 track stuff).

All which means a lot of re-ripping and leaving quite a lot of dupes  until can
do some ABxing on the questionable ones (older LAME versions, high rate FhG's).

Didn't a version of LAME have a bit bloat issue with vbr? Seems like a few files I looked
at which were either 3.92 or 3.93 were quite large for what they contained.
In looking at the ones with LAME headers, though far from an expert, didnt think anything was set extreme.
There also were a few files with what seemed lower (than avg) quality values in the 50's (58).
What might be up with those things?

Overall which is the better Encode to keep?

Reply #4 –
Quote
There also were a few files with what seemed lower (than avg) quality values in the 50's (58).
What might be up with those things?
[a href="index.php?act=findpost&pid=265634"][{POST_SNAPBACK}][/a]

If the files are CBR, 55 means q=5, 58 means q=2 and 60 means q=0 was set for encoding the files. Therefore, "58" in the quality index is nothing to worry about. For VBR, the values go from 99 (or was it 100?) (=V0), 88 (=V1), 77 (=V2) etc.  IIRC, --alt-preset standard in lame 3.90.3 set the quality index to "78".
Proverb for Paranoids: "If they can get you asking the wrong questions, they don't have to worry about answers."
-T. Pynchon (Gravity's Rainbow)
 

Overall which is the better Encode to keep?

Reply #5 –
Thanks much...
I had recalled mention of the quality rating not having much merit but kept thinking vbr for some reason and 58 to be very low.
Then it clicked when reading your post & you were right on the money. CBR w/q=2 sounds like the typical default for most rippers giving them a 58.

Searching again I see 3.93 was the problem child. Checking with ENCSPOT it appears it reports 3.93 whether 3.93.1 or not but
I gather the files with the higher than normal average bit rate be a good indication they are 3.93.
Since from all the posts I read on 3.93, nobody seemed to be clear on the problem, just be sure to use 3.93.1  because 3.93 with presets is broke.