I just love the friendly supportive environment that is being nurtured here.  Welcome to the Jungle, Sonny!

Quote

b. I trust my ears

This argument can not, and will not, be accepted.

Time and again, sighted evaluation of audio quality has been proved to be unreliable. This is the reason why in this community statements regrding sound quality have to be supported by blind test results.

Please follow the suggestions given above: use replaygain and do an ABX test if you want your claims to be considered as valid.

Sonny,
Now you are focussing on mp3 @128kbps. Nobody on HA will ever claim or has ever that LAME or any other mp3 encoder will produce transparent results at that bitrate.

JS is supposed to be better than True Stereo (TS). Theorically. Indeed, JS encodes 2 channels, middle (M) and side (S) instead of left (L) and right ( R). Demonstration :
If M=(L+R)/2 and S=(L-R)/2 , then it is easy to get the L and R channel back :
M+S=(L+R+L-R)/2 = 2L/2=L and M-S=(L+R-(L-R))/2=(L+R-L+R)/2=2R/2=R
The trick is : you encode M on one side, and on the other, the more L and R are similar, the less the S channel will need bits to be properly encoded (ex: if L=R (=Mono), the needed space for S channel is zero !)

So. I'm sorry for disturbing your great community with my pathetic problems. I won't bother again.
bye all

1. I didn't do ONE blind test, I've spent DAYS blind testing and made something like 10 tests, with up to 14 FILES and I always hear the distortion.

Sometimes (for the 128kbps samples) I can hear it in less than 3 seconds.
2. tell you where is the distortion ? Did you actually listen to the 128kbps samples I've provided ? Surely not!! you wouldn't even ask. 

Still, I've started this topic with two 128kbps small samples that show a STRIKING difference between TS and JS. And I didn't get any reply about it.

gday...
my two cents..

MS stereo / Joint Stereo will unfortunately always have this affect on
disorted signals.. (mp3) as you say.. stick with TS..
in other words.. JS is the downside of audio compression.. (as i see it)

you also wanted possible 3rd. sulutions..
go for MPC format. of those codecs i have listened to..
mpc have the least problems with disortion..
(actually close to zero.)

I don't regard this as polite. I spend time answering, trying to help you to find out if the problem is caused by playback conditions used etc. and you're not willing to give a simple answer to a simple question. Have you thought of this - maybe what I use for playback (good soundcard, player with replaygain enabled) makes the problem disappear for me? In fact I've listened to the 128kbps samples and with some concentration there are several positions where I can ABX (8/8) both from the original, but mostly because of things like pre-echo and watery sounding high pitched noisy sounds (cymbals etc.). I haven't noticed any position with particularly bad distortion. So again, could you please answer my question?

Huh? Who's talked about 3.94betas - have I missed it?

anyway, i've solve my problem (i stick with TS) and i don't want to argue (or be rude!) anymore
thanks for helping, really.
bye

Edit: As an (interesting?) sidenote, I had a CD player with a little "3D expander" button that would amplify the joint stereo loss by like ten times. It made some 160 and even some 192 kbps MP3s very audibly lossy. That was back before I played with -aps or ABX, but I knew I heard something very clearly. If I were to ABX with some DSP like this on for both samples, would that be legal?

i've tried musepack and found it very good at high bitrate but strangely, also very poor near 128kbps 
given that i very often use high bitrates, musepack would be a good solution... if my mp3 player would support it !!
so for now, I stick with Lame in True Stereo. 

thanks for your answer!
i've tried musepack and found it very good at high bitrate but strangely, also very poor near 128kbps 
given that i very often use high bitrates, musepack would be a good solution... if my mp3 player would support it !!
so for now, I stick with Lame in True Stereo. 

Have you actually compared joint stereo vs. -m s with this device? I ask this because it's a known phenonemon with every lossy audio compression (exceptions might be codecs without psychoaccoustics like wavpack lossy), that you'll hear bad artifacts if you substract channels (like e.g Karaoke plugins do or your headphone if it's not plugged in correctly). The reason for the artifacts you hear could be somthing very similar.

Thanks for the info, Sonny.

I've done some testing:

All encoded files you uploaded clip on decoding, so this (together with a resampling soundcard) might be the reason for the problems you hear. I attatch test.zip containing 2 processed samples of your uploads (original and 128kbps JS mp3), with lowered volume + hq resampled to 48kHz. I'm interested if you still hear the bad distortions with these samples.

Therefore,  it would be nice if you could help the regulars and developers here figure out what problem you're hearing when using joint stereo, and why it goes away when using simple stereo. It would help you too, because the joint stereo mode has been tested much more than the simple stereo mode. Forcing simple stereo will probably just waste bits, but could break some of the tuning of the --alt-presets, and in any case can cause a decrease in quality for some samples.

So, are you willing to help?


Just to check: blind testing means you don't know what you're listening to. So, it could be the original, it could be the simple stereo version, or it could be the joint stereo version. From what you've said, you should be able to pick out the joint stereo version, but not the other two. Is this the case?

Now, the recommended version is lame 3.90.3, and the recommended command line is --alt-preset standard. The download link is in the FAQ. It's a very small download - give it a go. Do you still hear problems?

Please do not assume that people here are deaf, or that your musician ears are super sensitive and you're hearing something that no one else will. I've heard lots of people claim things like that. It's very rare in practice.

This is what I would try: people have said that the sample is clipping. First step: drop the gain on the original by 6dB (use any editor you like, or wavgain) so that this can't be an issue.

Now, encode the -6dB file using lame. --alt-preset standard, and --alt preset standard -m s. Decode both files using lame. (That's just lame --decode). Listen to the resulting .wav files back2back with the original. Blind. Do you still hear a difference?

Quote

thanks for your answer!
i've tried musepack and found it very good at high bitrate but strangely, also very poor near 128kbps 
given that i very often use high bitrates, musepack would be a good solution... if my mp3 player would support it !!
so for now, I stick with Lame in True Stereo. 

I'm curious about your opinion of Ogg Vorbis -q 6, and if you can hear the difference.

i've tested the ogg vorbis -q6 against a sample encoded lame with nearly the same bitrate.
ogg sample bitrate = 205kbps ; lame sample bitrate = 203kbps.
I used this command line for lame (3.93.1) : --alt-preset standard -V 3 -B 256 -m s

My impression is that the ogg sample is overall better than the lame. Particularly, it seems more dynamic to me, and cleaner on medium and high tunes.
But these are ONLY impressions (now don't tell me i've reach the conclusion that ogg is better or anything else !!)
Anyway, this was kinda obvious : given that the Lame sample is encoded in True Stereo, the ogg sample should sound better. if i solve my JS problem, i'll try a new comparison.
btw, the command line "--alt-preset standard -V 3 -B 256 -m s" is the best compromise i've found to get good sounding results without using JS. indeed, TS needs a LOT more bits to encode the same file with the same preset (for example, the same sample was encoded in 236kbps with the alt preset standard + TS, with lame 3.90.3), so i've lowered a bit the quality down to V3 and add a ceiling bitrate at 256 (i do not make the difference between 256 and 320). surely not great, but that's the best i've found out to keep the bitrate below 210-220.