Marmes actually had a similar problem while testing some time ago. If you're interested in all of the gory details, you can read a section of the ORPHINIT thread, starting here. His particular board (a P2B-L) had an unusual ISA Plug and Play implementation, and was reverting the configurations almost immediately, regardless of which tools were used. In that particular case, by the time the DSP reset was issued, the SB hardware was no longer at 220H, and it would fail. It started to behave when he set Plug and Play OS to Yes (EDIT: I had that backwards previously). Unless your board is even more unusual, then keropi's suggestion, combined with that change, should get things running. If not, please post the complete output of ORPHINIT /vd (which produces even more verbose output), and we'll try to get to the bottom of this.
There were likely also other hardware interactions causing trouble in that situation, which may also be relevant here. For now, the solution should still apply, as they are very likely caused by sub-optimal BIOS resource assignment.