Hello.

Well now its in there, but I must say I have a hard time hearing any diference.

Ah, how gratifying. Sometimes when I was testing I couldn't tell whether it was working or not, until I replayed the sample without the volume adjustment, and realised how much quieter it was! It would be nice to able to just turn it on and off, so you can easily see how much benefit (if any) it is providing.

Peter Gabriel: Sledgehammer: The wood flute intro is usually so quiet you can't hear it at all, and when the horn section comes in, it blows your speakers. With this kernel, the wood flute intro's volume is brought up so that you can hear it well in relation to the horns, but the horns come in at the correct volume. (Incidentally, Rjlov: the timing on this is much better with the last couple releases, it used to pump too hard on this bit, now it sounds pretty good.)

I haven't had Sledgehammer crop up lately, but I'm glad to hear it's better. I expect it was improved when we went to the exp(log()/a) scaling. Before that, everything that was quieter than -10db would be scaled up to almost exactly the same volume. Now, it maintains some relationship between louds and softs.

(I actually haven't spent enough time at home to download 12b.2 to my empeg, I'm still on the previous version, which has the FM clipping)

Richard, if you're reading, do you have any other favorite examples?

Well, pretty much any classical music. I like Holst's "The Planets". I have a recording by the LSO where the start of "Mars" has samples of around 30 (that's 30 out of 32768, or around -30db). And this is actual signal, too! The loud bits of the same track are maxed out, so it has already been normalised. Most of "Neptune" from the same suite is usually completely inaudible in the car (actually, even at home on the CD player it's mostly inaudible with the amp at my normal listening levels), but I can hear it fine now.

On the other hand, you're more likely to notice 'problems' with classical music, as well. It's interesting to listen to "Fanfare for the Common Man". A tam-tam decays so slowly, that the volume adjustment can easily keep pace. So, after the third timpani note, you can notice the tam-tam actually getting a little louder before it starts fading (slowly) away again. If you have any (even vague, hand-waving) suggestions for how to avoid this sort of thing, I'd like to hear them.

Richard.