I think we now enter the "grey zone" as where each persons observation play's a role.
But, physically, every tar needs a certain temperature and dwell time with glowing charcoal to be "reduced" or "cracked"
if the transition zone, between oxydation and reduction, is well formed, then only "smoke" will enter the reduction zone and that will be reduced in a constant , irreversible state of gas.
If the transition zone passes liquid Tar ( mostly the heavy tar's ) then that tar will turn into smoke in the reduction zone and only partial stay gas, rest will be condensed again.
As per initial build from Kristijan, i was very fond with the idea of air injection in the reduction zone, to eliminate the cool down from passing Tar's, hence keeping the char glowing more profound.
Having a large glowing charcoal bed is, IMO, the most important issue.
Generating sufficient heat to do so, is best accomplished by , again IMO, retaining the heat but without preheating the incoming air.
The best example to confirm that is the statement/observation from Kristijan, that when he opens the lid, the gasifier runs at its best. Hence better airflow, more free O's going down in the bed.
The dryer the fuel, the easier it gets, (known fact), to compensate for moister, you'l need O to have sufficient combustion.
I am working on a "condensate" entrapment for the charcoaling unit, but this can easy implemented in a gasifier to eliminate some Tar issue's.
PS. The better i take care about my fuel in my gasifiers, the less smoke i get ,,, So why should that be different with a good woodgasifier once the char bed is well established... ?