I download many mp3s, convert them to wav files, edit them for various
things (equal loudness, amount of silence at the beginning and end,
reduce dynamic range, change tempo), then reencode them.

Most of the d/l are at 128 kbps. (Of course, I have no way of knowing if
they are first-generation.) Is it a good strategy to reencode them at
192 kbps? I know that the losses in the very first encoding can never be
recovered, but I'm hoping that 192 will minimize further losses. I'm
willing to accept the larger file size (vs 128).

Where possible I try d/ling songs with the highest bit rate (320). So if
they are first generation, reencoding at 192 will introduce new losses.

I encode using CDex, v1.5.1, which uses Lame encoder v 1.30, engine 3.92
MMX. Settings: Version MPEG1; Quality, Very High; Mode, stereo; Output
samplerate, 44100.

I load the songs into my laptop and use it for DJing to a crowd of 120+
senior citizens. So I don't need a "golden ears" answer.

Thanks,

Ray