Is it possible to tell which encoding is better by looking at this info? And if so, which one? (It's the same song recorded twice, obviously)

ENCODE 1

General
Format : MPEG Audio
File size : 3.65 MiB
Duration : 2mn 17s
Overall bit rate : 205 Kbps
Writing library : LAME3.97

Audio
Format : MPEG Audio
Format version : Version 1
Format profile : Layer 3
Duration : 2mn 17s
Bit rate mode : Variable
Bit rate : 205 Kbps
Minimum bit rate : 32.0 Kbps
Channel(s) : 2 channels
Sampling rate : 44.1 KHz
Resolution : 16 bits
Stream size : 3.36 MiB (92%)
Writing library : LAME3.97
Encoding settings : -m j -V 2 -q 3 -lowpass 18.6 --vbr-new -b 32


ENCODE 2

General
Format : MPEG Audio
File size : 4.43 MiB
Duration : 2mn 17s
Overall bit rate : 270 Kbps
Writing library : LAME3.98r

Audio
Format : MPEG Audio
Format version : Version 1
Format profile : Layer 3
Duration : 2mn 17s
Bit rate mode : Variable
Bit rate : 270 Kbps
Minimum bit rate : 32.0 Kbps
Channel(s) : 2 channels
Sampling rate : 44.1 KHz
Resolution : 16 bits
Stream size : 4.43 MiB (100%)
Writing library : LAME3.98r
Encoding settings : -m j -V 0 -q 0 -lowpass 19.5 --vbr-new -b 32

Thanks!!

http://www.hydrogenaudio.org/forums/index....showtopic=16295

Taking "better" to mean "likelier to sound indistinguishable from the original": the one with the higher bitrate, obviously. I'm sorry, but do you know anything about audio encoding at all?

(Also, it was encoded with a newer LAME version, so there's a chance it may benefit from updates made in the interim.)

Yeah you're right, I'm sorry. Its been a few years since I did any encoding and while sorting through all my stuff I got a bit confused. Especially with the presets. I was hoping to see --alt-preset extreme somewhere! :-) (It was also before my morning coffee so my brain wasn't running at full speed)

I remember at one time 'newer' versions of LAME were not necessarily better. I think at one point it was recommended to stick with 3.90.2 as opposed to the newer 3.91. And 3.92 was okay, but forget about 3.93 all together. So I was trained to question each release.

Anyway, Thanks for the obvious clarity.

OH! Now I remember the other issue I forgot to mention..... The second was encoded directly from a FLAC rip as opposed to the original CD. I assume since FLAC is LOSSLESS then that's okay, but can someone confirm that?

Thanks again. :-)

The second encode has an insanely high lowpass, though.
That said - judging by the settings and LAME versions I doubt you could ABX 'tween the two, nor against source. I'd take the smaller one.

The one with -V 0 is equivalent to --alt-preset extreme and the one with -V 2 is equivalent to --alt-preset standard.

Dang, just looked it up and that is the standard lowpass for V0. I guess I never saw that before as I never ABXd up that high.
(I still think it is insane)

Both encodes seem to be done with a frontend, since many of the options are just duplicates of what the setting already does ( -m j, -b 32..). As you say, there isn't a problem to encode directly from CD or from the lossless file generated from it.

So yes, it all ends up like being 3.97 --alt-preset standard vs 3.98 --alt-preset extreme

(Note that 3.98 uses a slightly higher bitrate with VBR. 3.97 with -V 0 probably would have been smaller, probably around 250).

Assuming that you cannot distinguish a 3.97 -V2 encode versus a 3.98.x -V0 encode, then I would say that the -V2 encode is better. (indistinguishable quality yet smaller file size is better)