Best measurment for Amp driving force W or uF

what is the best measurment for the driving power of an ampilfier. Should it be measured by output power in Watts in RMS or should it be total capacitance in uF. Some amp may have high output power in Watts but low uF while some have low output Watts but high uF.
Don't know what you mean by "best"... power output is what drives a speaker and is a function of the output RMS current and speaker impedance. Power supply filter caps are sized to keep enough reserve for low impedance, high power draw (transients) moments that the power line cannot deliver quicky enough and to limit the ripple current within a specified value (5% or so).

Filter caps are generally sized for 1,000 uF per RMS ampere from the transformer secondary and some design for 1,000 uF per peak-to-peak current. Any more usually does nothing except add cost, any less can limit amp performance.

But there are other parameters involved, eg, how they are wired: two 14,000 uF connected in parallel will usually outperform a single 32,000 uF in filtering and energy delivery by halving the ps output impedance.

It's best to listen, really.
I'd say that if you are comparing watts vs. uF, then you probably need to consider damping, or DF.

However if there are 2 amps with the same RMS value but with different uF, then all other things being equal, the one with more uF will have higher dynamic peaks and lower Power supply impedance. However all other things are not constant, therefore listen to the amps before buying.

Lastly, given similar sounding amps, try and get the one with more power output.. again consider all variables.
What good does having a million watts or tons of current do if you're trying to pass it all through a 16 or 18 gauge inductor ? This is besides the point that some folks are using 20+ gauge wire to feed the signal to the speakers. How much instantaneous or sustained power ( voltage & current ) can you cram through a wire like that ?

With all of that in mind, higher capacitance power supplies are always a good thing. You get less ripple from the AC, increased rejection of AC noise, can sustain longer duration high intensity peaks, bass is firmer with increased "slam" and more consistent tone, etc... Obviously, one gets to a point of diminishing returns and depending on the amount of power being drawn by the amp and the level of reactance from the speaker, the point of diminishing returns can vary pretty drastically. As a given rule though, good sized SS amps should have at least 80,000 to 100,000 uF of capacitance at the minimum. This is especially true if running low impedance speakers with a high level of reactance / reflected EMF.

As mentioned above, there are different ways to achieve high levels of power supply reserve and filtration though, so "big uF's" are not all created equally. One of my preamps uses two 20,000 uF's with bypass caps whereas another uses four 10,000's, a dozen 100 uF's and bypass caps across all of those. The total values are close ( 40,000 vs 41,200 uF ), but as one can see, very different approaches were used. Given that an SS preamp draws such a small amount of power as it is, the main reason for using such a large quantity of capacitance would be for increased filtering of the signal and to keep the impedance of the power supply as low as possible. The higher the impedance of the power supply, the more likely it is to be "modulated" by high levels of reflected EMF. While this is much truer of power amps, the reflected EMF that an amp has to deal with IS passed onto the preamp in terms of altering the load that the preamp sees.

All things being equal, more is better in both categories. That is, so long as the quality is maintained while achieving the higher figures. Sean