Mmm... maybe someone should ask... the software writers?? =) I had to hear this from the EXS group. Feel free to contact us through e-mail (support@...) or toll-free phone if US (800-877-6377) or 320-235-9798. We are always here. Concerning the "mystery byte," here's the values for that. I couldn't replicate values of 0x2C (decimal 44) with it. Perhaps Markus can e-mail use the source file for that and we can see what's happening. Can you do that Markus? The byte is a bitmapped byte, with these values OR'd together: 1=One Shot 2=Disable Pitch 4=Reverse 8=Custom Vel Range So 0x2C isn't possible-valid, neither is 0x70. Of course, Emagic may be using extra bytes internally for something, I don't know. I usually get 8 as this byte. And I've heard clicks with this set at 8. Anyway, the best answer I know of is that the EXS is extremely accurate, that's why you guys like it. The EXS isn't the first sampler to react adversely to non-zero initial samples at certain times. In fact, Roland's click too - that's why you see many pro Patches set with the Attack to 1 instead of 0. My theory is that the click comes from the playback engine ramping up inaccurately try to play the first sample if it is non-zero, or especially if it is far away from zero. The best workaround is to increase the Group Attack offset to 1. We are currently putting this into the software as an option, so people don't have to do it manually. Garth Hjelte Sampler User