Hi everybody!
After lurking around for a while on HA I never use anything besides preset standard (or sometimes medium) with lame 3.96. Before this, however, I used Lame 3.89 (beta) to rip a large part of my CD collection (various genres - pop, rock, electronic etc.) . I used the following settings: CBR 192 Joint Stereo, q=2. As you might have guessed, all the talk about how CBR sucks, and Lame 3.90.3 or above is the way to go I feel a bit uneasy about my older MP3 files. However, it is not possible for me to ABX the Lame 3.89 files – they sound perfectly transparent to me even on very good equipment (I used a very decent Sony CD Player and Sennheiser HD 600 headphones). While it is easy for me to spot any 128 kbit MP3 due to the missing high frequency content and most Xing or Blade MP3s due to their heavy pre-echo or swishy/ringing sound, my 192 kbit Lame 3.89 files sound great to me – no discernible pre-echo and no obviously missing frequencies. Still, my vague uneasiness remains – compared to --preset standard, my files are certainly inferior, but to my ears they sound too good (i.e. perfect) to justify a re-rip.

What are your opinions?
Is there something seriously wrong with Lame 3.89 as compared to later versions? What is it and how could I notice it when listening to my MP3s?
Has it serious problems with the Joint Stereo implementation (I noticed that my Lame 3.89 files have in general more MS frames than Lame 3.96 files, but this alone should not be a problem, right?)?

I know that listening to MP3s is a highly subjective business, but I would be grateful for any thoughts on my question. I also know that my settings with lame 3.89 (CBR 192 JS, q=2) would not be considered ‘transparent’ by people with golden ears and/or experienced HA members, but can it be considered good enough to be ‘nearly transparent’ (i.e. transparent to the ‘joe average’ listener)?

Thank you for your help!

This is the only comment that matters. These are primarily for your own listening right? If you're happy with them, don't waste time re-ripping and re-encoding, just enjoy them!!

I will second what John33 said.

The only thing i'd even be remotely concerned about is if you start doing alot of mp3 testing/abx tests, that you *may* become more sensitive to particular artifacts.

If you can't abx now don't worry about it. However, the more you will mess with abx the more chances that your hearing sensitivity will shift. This happened to me because of abxing samples posted here. Now I can nearly always abx any random sample in encoded in lame 128k - I couldn't do this 6 months ago and I am glad i don't have a low bitrate collection. 192k artifacts should not be annoying in this case.

If you want to re-encode my suggestion is that you rip to lossless first then transcode. If you don't have space, archive the lossless copies on another drive ,DVD or even CD's. This way you won't need to re-rip if you decide you need more quality headroom, another format etc.. and you can easily go lossless once you have the space.