RMS = root mean square (it is the square ROOT of the MEAN or average, SQUARED)
RMS mode is able to recognize the on/off cycles of PWM and adjust the frequency in order to accurately achieve wattage set-point.
AVG mode is NOT able to accurately compensate for the on/off cycles of PWM. It averages the 0vdc and the 5vdc (or what ever the pulse voltage is) and concludes that the output is half of what is actually is so ramps up frequency to double of what it should be.
AVG has its uses, but does not belong in a circuit using PWM. If you want a stable output from a manually adjustable power supply, AVG circuitry is a win.
The reason you see an improvement with the provari over the twist or spinner is due to range capability.
The twist and spinner are capable of something like 3.3vdc to 4.8vdc range.
The provari is capable os 2.9vdc to 6 vdc.
What you see as an improvement in vapor production/volume, and TH is due to a higher wattage range achievable at higher voltages.
Mechanical MODS use neither. They are a function of battery voltage and resistance of atty.
Voltage^2 x resistance = Power
In order to "tune your vape" you must select the right Ω atty. As your battery drains, the wattage output decreases.
Advantage of a mech MOD is there are no electronics to fry, will presumably last longer, and is able to use lower Ω attys.
Advantage of VW APV is you can repeat your favored wattage regardless of what Ω atty you put on the device (see limitations below), will give you greater control to adjust wattage output without requiring an atty change.
APV limitations: Built in safety cut-off does not allow use of attys below 1.2Ω.
Both serve their purpose. I have one of each.
Simplified Explanation
AVG averages the voltage to determine how much to apply to reach setpoint. Useful in a constant output device like all of our original PVs and in VV using a POT (potentiometer) to regulate voltage.
RMS uses the square ROOT of the MEAN of the SQUARE'd average of PWM (pulse width modulated) or AC (alternating current) supplied voltage. You can't perform a straight average of a sine wave as the highs and lows will cancel each other out returning a value of zero. So the square root of the MEAN is used which will always return a possitive number, and the final result is squared in order to in order to return an accurate "running average" over the length of the varying signal.
RMS mode calculates the voltage required to maintain an accurate power output even though the voltage is pulsed to the atty.
If you use AVG mode on a device that applies voltage by PWM: The highs and lows are averaged. For an equal distant pulse wave expecting a 10w output the voltage will increase until the true average is achieved. It can hit more than twice as hard as you are expecting.
Because:
Total Power (wattage) generated is calculated by the average power out put over time.
Example:
10w for 30 minutes and 0 watts for 30 minutes would = 5w per hour average. (In AVG mode, in order to maintain a setpoint of 10w it would output 20w in order to achieve it, not regognising the "off" time).
In RMS mode, it knows the difference between the on time and off time and calculates the voltage required to maintain the power setpoint accurately.
In AVG mode, it doesn't know that the power is being pulsed and increases voltage in order to compensate.
So what good is AVG mode? Use it when you are in VV mode, you should encounter no problems. Also useful for any steady voltage output scenario.
RMS mode is only needed because our power supply (in newer APVs) pulses voltage to achive wattage setpoint.
Using RMS mode while in VV mode: same accuracy as using AVG in VV mode.
A thesis could be written on the subject to explain it in greater detail, I tried to keep it simple so the non tech guy/girl can grasp the basics.
(BOTTOM LINE)
RMS mode for pulsed atty power.
AVG mode for constant atty power.