it doesn't matter... whichever has a better resulting coverage - distribution of light - will be better if all other factors are the same. PPF or umol/s PAR is the key specification to compare in that regard. if it produces the same rate of PAR photons, the results will be very similar and only deviate due to dsitribution of light.
So, if 2 lights better covers the area, then sure...
in most cases it'll make very little difference so i woudln't overspend for either context.
400w can mean drastically different things due to differences in efficacy -- how well it converts electricity to photon production in the wavelength range plants can use (PAR).
High efficacy = quality. If something is priced high but not around 2.8-2.9 umol/J efficacy, it's over-priced garbage. The best way to assess this is to compare to the testing paramaters on the diode manufacturer's website, like the LM301 from samsung.com. It's tested at 0.2w per diode, so longevity and effeicacy is tied to that fact. Piping moe rpower through that diode will cause it to die faster and operate less efficiently. Slight deviations are not so bad, but doubling and tripling has a significnat impact. Any LM301-based light up around .4-.5w per diode and higher are not going to last "50,000" hours and it's efficacy is nowhere near the high end even if it advertises something close to 3umol/J.
umol/J, simplified... just multiply watts x umol/J and it should equal the umol/s PAR output of the light.. one more way to double-check the light manufacturer's spec sheet... many, if not most lie through their teeth, so you always have to compare to the diode manufacturer testing specs. Those can be trusted.
lights are simple. if they use a good diode, use a proper amount of diodes relative to power, and have some decent heat dissipation, it will be an excellent light regardless of brand name. Learn how all that jives together and it's easy to recognize a high quality light... and easy to size one correctly for your needs.
400w with a high efficacy light will be around 1100umol/s PAR output. Less efficient lights will fall short of that.
12-hour operation
1100 / 900PPFD = 1.22 m^2 coverage for ~39 DLI (daily light integral) That's how much a high efficacy 400w of light could cover -- assuming shape of lights fit area of coverage.. i assume you can apply common sense like that when looking at options. the math is a ballpark. Always have to adjust based on observed growth pattern due to local variables. this gets you close.
18h
for autoflowers you could cover 50% more area while operating 18 hours per day. Or, with vege phase of photoperiods, you run the light at 67% and cover the same 1.22m^2.