Now that all --alt preset (not fast) bugs are out of the way should we be switching to 3.93 for --alt preset standard encoding?.

There is no 3.93. There is 3.93alpha. It has been said countless times even by Lame developers that people should not use alphas for anything other than testing.
Although last time I tested, there was no quality problems with --alt-preset standard and 3.93alpha, the average bitrate was still somewhat higher.

So people should still use Lame 3.92 (or 3.90.2).

I'd probably still stick with my custom 3.90.2 for now.

I'm hoping I can get some time soon to add some things into 3.93 and to make sure everything is working the way it used to before it goes stable. Until then though, I don't really see much of a reason to switch over yet.

I know this has been asked before, but is there any appreciable difference between 3.90.2 and v3.92? (sorry, I forgot the answer). I've been using v3.92 --aps and it sounds good to me.

They are different in speed and output, not sure how about quality trade-off cuz I didn't do my own listening test.
This thread talked about it.

Ah well, thanks... I guess I'll trust my ears and stay with v3.92 for the moment (--aps sounds OK to me). I could switch to 3.90.2, but for psychological reasons that feels like a step backward.

Fair enough but you have to weigh the good against the bad, I remember everyone recommending to switch from 3.89b to 3.90alpha because the quality improvement was so good.

I've done a bit of testing with --alt-preset standard on 3.92 and 3.93a2 pulled out of CVS and on the average 3.93 is twice as slow with encoding (platform UltraSparc, same compiler and optimizations)

Anyone on x86 processors see a similar slowdown?

--jth

Quite the reverse, actually!! It's about 10-15% faster on my Athlon/Win2k.