Suggest ~160kbps VBR setting. Options

[guidryp]

Originally posted by JohnV


Still I think if you test even more tracks, you'll find a track that bloats clearly with --lowpass 16 but doesn't at -Y. 
If you don't care about over 16kHz then for best results you should use both. If you wan't to retain even somekind of over 16kHz freq response, you should use only -Y which prevents file size bloating.

I guess I could include -Y as well to be safe.

I would like the over 16KHz, but only if it is free, since thus far I haven't been able to detect a difference in music. It may even be a safety margin for me.

I couldn't consistently pick out the 15KHz mustang track from the original. I felt there were difference, but I couldn't tell reliably.

Peter

[Dibrom]

Originally posted by scoper
Reading the suggestions of Dibrom and others regarding the use of -Y along with aps brings me to a question.

Many of the mp3's I encode are from FM radio sources, and FM of course cuts off around 15khz. That being the case, would Dibrom's suggested setting of --aps -V3 -b96 -Y be optimal for encoding FM sourced WAVs?

Scott

Sure... though if it were me, I'd just use --alt-preset standard -Y, without the -V3 or -b96...

[pacohaas]

I ripped some complete albums and did some encoding, here's the results, interpret as you may:

Albums:
1. Adam Sandler-What the Hell Happened to Me
2. Cinema Beer Nuts (Hopeless Records Compilation)
3. Green Day-International Superhits
4. Lars Frederiksen and the Bastards
5. Punk Bites (Fearless Records Compilation)
6. Punk O Rama 5 (Epitaph Records Compilation)
7. The Vandals-Look What I Almost Stepped In

Bitrates: (kbps)
--alt-preset standard
1. 185
2. 229
3. 232
4. 218
5. 206
6. 225
7. 213
Average: 213.9
--alt-preset standard -b32
1. 185
2. 229
3. 231
4. 218
5. 206
6. 224
7. 213
Average: 213.5
--alt-preset standard -Y
1. 173
2. 179
3. 177
4. 164
5. 169
6. 173
7. 172
Average: 172.7
--alt-preset standard -b32 -Y
1. 173
2. 178
3. 176
4. 164
5. 169
6. 173
7. 172
Average: 172.5
--r3mix (just for kicks)
1. 169
2. 223
3. 227
4. 213
5. 201
6. 218
7. 200
Average: 205.6

=========================================
Here's my personal conclusions:
1. -b32 hardly affects the bitrate, so why bother using -b96 or -b80
2. While -Y does bring the bitrate down, because all -Y really is is a lowpass, some of the wrong tracks get lower bitrates(3, 4) and higher bitrates(1, 2) than they should according to aps.

[guidryp]

Holy Smokes!

You ripped 7 albums 5 times just to experiment? My Hats off to you.
I will look at just plain -Y on one album.

Peter

[Amadeus93]

Originally posted by pacohaas
1. -b32 hardly affects the bitrate, so why bother using -b96 or -b80

The principal reason I use -b80 is because some of the material I encode is mono, so it doesn't need as high a bitrate. On stereo material, you're right - it behaves the same as aps without -b80.

[JohnV]

Originally posted by pacohaas
2. While -Y does bring the bitrate down, because all -Y really is is a lowpass, some of the wrong tracks get lower bitrates(3, 4) and higher bitrates(1, 2) than they should according to aps.

Again... -Y is not a lowpass.
Just to put it simple (I've explained this so many times already during few days), using -Y can directly affect to the bit allocation of the whole frequency range of MP3. If it affects, the effect will be lower bitrate. -Y also can retain some high frequency content, so it's not a lowpass.

It's totally expected that those tracks/CDs which give the highest bitrate with APS, show the greatest drop in bitrate with -Y (unless the music is really complex). The reason some trakcs/CDs give higher average bitrate doesn't necessarely mean it's "harder to encode" complex music. It's just as likely that the bloating is because the SFB21 issue, and if this is the case, -Y gives very high drop in bitrate.

[pacohaas]

Originally posted by JohnV
Again... -Y is not a lowpass. 
Just to put it simple (I've explained this so many times already during few days), using -Y can directly affect to the bit allocation of the whole frequency range of MP3. If it affects, the effect will be lower bitrate. -Y also can retain some high frequency content, so it's not a lowpass.

I understood that, I wasn't saying it was just a lowpass, I guess I was trying to point out that it might not be the optimal solution to lowering the bitrate because it only lowers bitrates based mainly on frequency data and not highly optimized and tweaked as --alt-preset standard is. So I agree with you, and here's some data to add:

[b]--alt-preset standard --lowpass 16
1. 173
2. 178
3. 175
4. 164
5. 168
6. 172
7. 171
Average: 171.8

======================================
As you can see, this data is very close to the --alt-preset standard -Y data, which backs up my original correlation between a lowpass and -Y (and again, yes I know that's not what -Y is)

Holy Smokes! 
You ripped 7 albums 5 times just to experiment? My Hats off to you.

thanks, when I do a test, I want to do it thoroughly. I picked some well mastered/produced CD's(because I knew they would be higher bitrate that my other CD's) and some good compilations(to test a range of recordings) to get a thorough test.

[pacohaas]

which brings me to yet another idea that I've been throwing around since I read in the source code that ABR was reccomended at lower bitrates:

2-pass encoding:

pass 1: something like r3mix or aps -Y
pass 2: ABR at pass 1's bitrate.