Of the 2, I'd use the FCE3000 primarily because it fits the size of your tent better... and could cover a slightly larger tent at that... the spiderfarmer is 450w and too much for .64m^2. when you dim significantly, the diodes to produce light in same consistent way from what i udnerstand, so it is better to run closer to 100% than below 50%, in general... plus, cheaper than the sf4000.
Diode count puts it at 0.25w/diode which will match spec sheet on bridgelux.com... so the specs are accurate as opposed to looking through rose-colored glasses.
Alternative -- SF2000 200watt option from spider farmer... 0.33watts per diode of samsung lm301b (better than their 450w vesion for some odd reason). 0.25w is what the spec sheet is based upon, so i expect the efficacy is "close enough" 2.5-2.6umol/J and they list "2.7umol/J." if they use top bin diodes, it would push 2.7umol/J. most manufacturers do not buy the sj/sk diodes for the warmer white chips.
This smaller SF light would be 800PPFD or 35DLI over 12 hours per day... this is near ceiling for atmospheric co2 and loosely controlled environment. And, over 18hours for autoflowers it's up around 51-52DLI which is near max for co2 + tightly controlled environment over 18hours per day. so, you'd want to dim 66-75% for autos in a 80x80cm space. 100% for photos 12/12... (66-70% in vege 18/6)
Relative to your tent/box size, the SF2000 is probably best option for you. FCE-3000 is slightly ovepowered, but if you plant to up-size, it could be the way to go, too. unless you want a 120x120 area in future, i'd avoid the 450-600w fixture(s).
With both you want a little breeze going over them... that is more than enough to keep the diodes running at a good temp for best efficacy.
----- extra info
MARS Breakdown
https://www.mars-hydro.com/fc-e3000-led-grow-light
specs look accurate enough...
800-some umol/s in .64m^2 is roughly 1250+ PPFD -- some of which will miss the plant, but that's WAY ove what ambient co2 will allow. Over 12 hours that will provide 52DLI -- 38-42DLI is roughly ceiling for ambient co2 conditions and loosely controlling environement to avoid too cold or too hot at minimum.
That's relative to 12hours per day in bloom -- so autoflowers are 150% of that... 78-80DLI over 18hours in a .64m^2 area.
bottom line -- it's overkill for 2.5x2.5 -ish area. photoperiods maybe 75% in this space? and with autoflowers, you would need to run it at 50% or so 18hours per day.
if you tightly control environment and add co2, then you can push to 50+ DLI... take bottom line suggested dimming and multiple by 1.25-1.33x. 90-100% for photos 12/12, and 60-65% for autos 18/6
i'd look into the bridgelux chip on bridgelux's website... make sure longevity curve is similar to samsung and all is well. Pretty sure they are also 1/2watt chips and they spec sheet will ikely show them run at 0.25w per chip for the best results. run them hotter and they die faster.
*the FC-XXXX are accurately specified.. i am not 100% certain of the FCE-xxxx.
SPIDERFARMER
https://www.spider-farmer.com/products/sf-4000-led-grow-light/
Okay, i think they slightly fibbed on efficacy... watts per diode are up near .37watts per diode... so, unless they are using top bin, i don't expect them to reach 2.7umol/J .. maybe with some rounding but think it's more than 0.1 off. still good.. and if they do have the best diodes of each CCT used, then 2.7 is definitely possible. IR/UV/the reds, all bring efficacy down (relative to parts used)
otherwise looks fine... good light. fairly good price (unless you want to negotiate on alibaba).