LAME.exe's "bitrate around" and -V "'s relation, and -k option good or bad?

[sommers]

0. LAME.EXE -mj -V0 bitrate around 280 (best) !-V = variable !mj = joint stereo
1. LAME.EXE -mj -V1 bitrate around 224
2. LAME.EXE -mj -V2 bitrate around 224
3. LAME.EXE -mj -V3 bitrate around 192
4. LAME.EXE -mj -V4 bitrate around 160 (default)
5. LAME.EXE -mj -V5 bitrate around 128
6. LAME.EXE -mj -V6 bitrate around 128
7. LAME.EXE -mj -V7 bitrate around 112
8. LAME.EXE -mj -V8 bitrate around 96
9. LAME.EXE -mj -V9 bitrate around 80 (poor)

A. LAME.EXE -mj -V0 -k

BB= -k full bandwidth
Tells the encoder to use full bandwidth and to disable all filters. By default, the encoder uses some lowpass filtering at lower bitrates, in order to keep a good quality by giving more bits to more important frequencies. Increasing the bandwidth from the default setting might produce ringing artefacts at low bitrates. Use with care!

can someone tell me about "bitrate around" and -V "'s relation, 1. 2. are confusing. and 5.6.7 are confusing.

those "bitrate around" numbers were from my test.

-----------------------------------------

Lame's -k option good or bad?
BB= is what document says

BB= says giving more bits to more important frequencies for full bandwidth,
question 1: this sounds so good that everyone must use it. but it might cause ringing effect it says. so is it bad?
question 2 : disabling all filters sounds good when you cant trust anything and when the document is contradictory.
question 3: "important frequencies" in BB=, who decides what is important and what not?

[Marsia Mariner]

Just encode anything at 320kbps, CBR ++ full stereo.
Data storage is cheap and Internet speeds are high today, so why bother with VBR and smaller filesizes?

[sommers]

1. MEGAsync uploading is limited to 1m max,
2. USB3 has its steady limit of 80m bytes,
backing up 2~10 tera bytes still cost my life time with 1. and 2.

and compression is a habit that i cant break lightly and sadily..

[sneaker]

question 1: this sounds so good that everyone must use it. but it might cause ringing effect it says. so is it bad?
question 2 : disabling all filters sounds good when you cant trust anything and when the document is contradictory.
question 3: "important frequencies" in BB=, who decides what is important and what not?

1. Yes, it's bad.
2. Where is the contradiction? Maybe you just draw wrong conclusions from correct information.
3. The developer.

Encoder has to work with a limited amount of bits, e.g. 128kbps. You can disable lowpass filter. Then the encoder will put bits into higher frequencies. Now remember: we only have those limited 128kbps. So if we put more bits into high frequencies it means we take away from low frequencies. This can be a bad idea because human ears probably won't hear anything in those higher frequencies.
The developers put those filters there for a reason. They are not stupid. Stick to their defaults.

[pandy]

you may think about lowering ATH while still using -q 0 and -V 0

[sommers]

-k option
i will consider using -k as all filter disabling, so it still may be closer to the original sound data. i should use -k enable for a science project. but for human ears no -k.

what is the point of using a lowpass filter and 48000 or 44100 sampling? this could be a contradiction to me.

-mj
I still should or shouldnt use -mj for bitrate 224 region, or it just causes more alteration of the original sound without noticing the difference or getting a full benefit?

if i decide to do bitrate 320kbps.
i should use -k
i should remove -v0
i should removed -mj
i should remove all other options that just might alter data even more?

[Marsia Mariner]

-k option
i will consider using -k as all filter disabling, so it still may be closer to the original sound data. i should use -k enable for a science project. but for human ears no -k.

what is the point of using a lowpass filter and 48000 or 44100 sampling? this could be a contradiction to me.

-mj
I still should or shouldnt use -mj for bitrate 224 region, or it just causes more alteration of the original sound without noticing the difference or getting a full benefit?

if i decide to do bitrate 320kbps.
i should use -k
i should remove -v0
i should removed -mj
i should remove all other options that just might alter data even more?

For a science project, you should definitely use LOSSLESS compression.

Sampling rate = 44.1kHz means that the highest frequency to be reproduced is 22.050kHz.
Most adults actually cannot hear anything above 15 or 16 kHz, so it makes a lot of sense to discard the frequencies above these values while doing lossy compression.

If you decide to go with the maximum bitrate allowed by the specs, I recommend that you DON'T call the -k switch, just to be on the safe side.

The truth is, MP3 sucks. People with excellent ears will notice the compression artifacts even at 320kbps... not because of insufficient bitrate, but because in spite of all good-intentions, the format itself is... well, you'd better read the explanations at Wikipedia:

https://en.wikipedia.org/wiki/MPEG-1#Layer_III.2FMP3

[Hoser Rob]

Just encode anything at 320kbps, CBR ++ full stereo.
Data storage is cheap and Internet speeds are high today, so why bother with VBR and smaller filesizes?

Couldn't agree more, and I don't see the point of any of those filters.

I don't convert to mp3 unless the source is badly recorded either. Unfortunately MOST music is badly recorded.

[sommers]

>... because in spite of all good-intentions, the format itself is...
i understood.

> -k
i will ask this agian some other time, because i didnt like the lowpass filter idea that may kill the high frequencies.

i will remove the option -mj , -v, i wont use -k for now. i will use -b224 or -b256, but not -b320.

i will try to keep flac's.