best LED light 4x4

Read wiki on Daily Light Integral (DLI) .. get the gist of it. roughly 500 watts of a high efficacy light can do it... or maybe 600-650 watts of a low efficacy cheaper light can do it. 16sq ft is roughly 1.5m^2. So, for photoperiods/12 hour operation requirements you want 900*1.5 = 1350 umol/s PAR produced by the light. This is "900 PPFD" and over 12 hours that's up near 40 DLI.. think it is 38.9 if memory serves. This may be 5% too much.. maybe 10%. IMO, better to err slightly on high side. Over time can mitigate the loss of 10% from dimming over 50,000+ hours of use. If you stick to autoflowers on an 18 hour cycle, you need 2/3rds of "1350 umol/s", or about 900 umol/s for 16 sq ft to reach same 38.9 DLI. You can see that DLI is 1:1 inversely related. If it runs for 150% of the time, it needs 2/3rds of the power to give the same DLI. So, if you plant to grow autos or photos, exclusively, it puts different requirements on your equipment. If yo plan to do both you should be realtive to 12 hours operation requirements for obvious reasons. "requirements" is more of a suggestion for upper end of ambient CO2 conditions. actually max dli you can provide will vary from another garden due to varying factors like ambient co2, temp and RH%. you can save money with cheaper lights but when you get up to 500-600watts it becomes at lot easier to recoop any additional day-one cost by saving on each electric bill. You can work out a break-even analysis with this info. If you use the light year round, it should last 6-7 years before you get to 50,000 hours and they go downhill faster after that point (see longevity curve of diode) Go for lights with 2.7-2.9umol/J efficacy. Lights listing a higher efficacy should be looked at with skepticism. Far too often manufacturers either bastardize the numbers or plain lie on the spec sheets. You can always reference the diodes from the original manufacturer's site (like samsung or osram etc). The power at which they are run is the key. If the spec sheet says they are run at 0.2 watts to get the given specifications, any power above that will reduce efficacy and also reduce longevity of the diodes. Specifically for samsung 301 chips -- you want 4-5x as many diodes as watts.. or divide watts by diodes and should be .20-.25. After that it's just about build quality.. is it flimsy or does it have a good heat sink? They often give the highest efficacy of all didoes used and pawn it off as the efficacy of the light. or.. worse.. based it on the 6500K diodes (cooler CCT = higher efficacy) specs even though they don't use any of those diodes, lol.. or, like i said.. plain lies. mars hydro generally has good prices in north america. they have really improved the accuracy of their spec sheets over the last 5 years or so. Comparing what you find in your region to the "FC" series (fc-4800 for example) is a good way to ensure a high quality light with high efficacy. if they diodes are running up near .4w - .5w ,, this should be a "cheap" light. they are not bad, necessarily, but should be cheaper because they will cost more to operate and they will dim/die faster while running hotter in that time too. efficacy = longevity efficacy = lower operation costs for same strength of light. efficacy is king. distribution of diodes is no better than the 'bar' style for sure.. that's common sense you can see with your eyes.