It is what it is. Advertising related to phases has always been sketchy (from everyone, not just Asus). So much more goes into VRM design, but phases and power stages are easy to quantify, and so people latch onto it.
If they Asus style of forgoing doublers was in any way not up to the task, changing power stages would have no effect at all.
I agree, I will hammer any maker for pulling the same stunt and I have.
There's certain disadvantages of the Asus system.
If you only have 4 phases, you only have 4 phases of control. Removing the doublers basically turns those stages into parallel phases. Even if you have 4 stages under each phase, you only have 4 phases not 16.
The biggest issue besides scaling, or not being able to turn on/off phases to meet demand is the issue of balancing. You have 4 stages getting fed one signal. You have one draw point, and so because of mass production, you'll always be loading one stage harder.
It makes them faster at times but it also makes them wasteful. It also makes stuff cheaper because you don't need fancy controllers or smart doublers.
In order of preference...
1. 16 actual phases
2. 8x2 phases with 8 smart doublers or smart controller that can balance.
3. 8 big phases with redundant stages.
I really would like to know why Asus did a respin with the big fancy controller. I doubt Asus will fess up and it would be very expensive to do the testing of the power systems properly.