Appholic
Member
Fifth run at Multi-Sub Optimizer and this one finally clicked, going by a post I read this morning. Two SB-3000s at the rear, two Pretium 18s up front, all four into a miniDSP feeding an Onkyo RZ70 running Dirac with bass control, measured before DBC went on. He had also stopped toeing the front pair in.
The bit worth stealing: he level matched every sub individually, then played all four together, read the combined SPL, and used that combined figure as the target in MSO rather than the individual numbers. Written down it sounds obvious. It is also the step I got wrong my first two attempts, and it is why the filters kept coming back with gains that made no sense. If the optimiser thinks each sub contributes what it does in isolation, the summed result overshoots, and it burns its budget correcting a level error instead of solving a phase problem.
The other quiet lesson is "fifth attempt". MSO is not a run-it-once tool. Every thread claiming it did nothing is usually someone who took one set of sweeps, accepted the first solution it offered and moved on. It pays off for taking more mic positions than feels reasonable and re-running with different constraints until something holds up away from the main seat.
Where I would push back: four subs plus Dirac on top means two systems with an opinion about the same 40Hz. Which one gets the final word? Let MSO flatten things so Dirac has an easy job, or let Dirac carry it? I have seen both defended hard and I still do not know which camp I am in.
The bit worth stealing: he level matched every sub individually, then played all four together, read the combined SPL, and used that combined figure as the target in MSO rather than the individual numbers. Written down it sounds obvious. It is also the step I got wrong my first two attempts, and it is why the filters kept coming back with gains that made no sense. If the optimiser thinks each sub contributes what it does in isolation, the summed result overshoots, and it burns its budget correcting a level error instead of solving a phase problem.
The other quiet lesson is "fifth attempt". MSO is not a run-it-once tool. Every thread claiming it did nothing is usually someone who took one set of sweeps, accepted the first solution it offered and moved on. It pays off for taking more mic positions than feels reasonable and re-running with different constraints until something holds up away from the main seat.
Where I would push back: four subs plus Dirac on top means two systems with an opinion about the same 40Hz. Which one gets the final word? Let MSO flatten things so Dirac has an easy job, or let Dirac carry it? I have seen both defended hard and I still do not know which camp I am in.
