" Pollenated weed plants grow seeds instead of synthesizing THC and other cannabinoids. Well, they still synthesize these compounds, but in much smaller quantities."

As a medical user that grows it all the time, WRONG.

Most of the useful cannabinoids are produced and contained within the trichome. Seeded sections of flowers tend to have higher concentrations of these trichomes, as it's a protective measure against UV damage by the sun to the developing embryo, due to how thin the tissue of the calyx is.

Exactly. I've got a 55 gallon freshwater, using a single 50w 6500K panel (I'm only driving it at half power.) Everything is well-lit and the plants grow so quickly I'm having to remove an entire jungle every time I'm changing the water.

And anyone with half a brain knows they can just go to the manufacturer direct in most cases and get what they need. They have no qualms selling to you at the price they give to wholesalers and retailers. None.

On the other hand, finding someone that knows EXACTLY what you need (my job) per a given situation is not easy, which means most people are still stuck listening to the marketing of fly-by-night ebay companies and less than reputable LED grow light companies that claim to have patents when they're just reselling stuff from China.

"This Wikipedia page [wikipedia.org] has several examples of the efficiency of different kinds of lights. Most of the LED examples they give show around 50-100 lumens per watt."

Quit relying upon Wikipedia. It's outdated. Cree's already pushing 300+ in the labs and have 200+ commercially available. Most LEDs are pushing 130-150 right now.

You've listened to wayyyyy too much marketing and read too much ill-educated nonsense on cannabis forums.

The light cycle itself triggers flowering. Red light just happens to be a bit more efficient than blue for photosynthesis. I've done flowering under pure blue light and still obtained the typical stated yield from the seed supplier's website.

HPS gets used for flowering because the intense green output, which can go through the canopy, down to lower sections of the plant, where the green has overall superior quantum yield, as red and blue will not penetrate that far. IR has minor effect.

Guess what? We've got white LEDs pushing well past anything HPS lamps can do in terms of lumens per watt (really we're looking for photon flux density) and at color temperatures much closer to the natural light of the sun. HID? Finished.

Correction, XPG2 is a 5W LED, not 5A. 1.5A drive.

"Marijuana plants like direct, intense sunlight. Unfortunately LEDs aren't very scalable. As you increase the current they start to run into physical limitations and the efficiency goes to hell. A 100W led may only put out 50lm/W, where a 1W led could put out 100lm/W."

I'm sorry, we've got plenty of LED systems out there pushing 2,000+ umol from several feet away, like any HID. And typical 100w LEDs are about 130 lumens per watt. Cree has LEDs available for the consumer that at 1w drive get 200+ lumens per watt (Cree MK-R) We figured out the Auger/watercooler effect and are working around it. We've got LEDs that dump 150+ lumens per watt at 5-10A drive current, now (Cree XPG2.)

"HIDs are actually extremely efficient (Around 100lm/W) and scalable (bulbs go up to 1kw+)."

We've got 1,000w LED packages in 70mm x 70mm size, with better efficiency.

" To get and equivalent amount of light out of LEDs with some sense of efficiency, you'd need thousands of them."

Plants care about photon flux, not lumens, which is weighted at green wavelengths for human vision. That's not saying that green light isn't useful in itself, but lumens mean jack shit when it's overall photon flux density making the real difference. I'm questioning how out of date your information is. Seems like way early 2000.

And thousands? Seriously?

http://tinyurl.com/mxq5w2b (PDF WARNING)

Try a couple hundred, easily fit on a 30mm x 30mm COB array board. Those are 5W LEDs each. The Cree MK-R is a 15W LED. 100 of those is 1.5kW.

" This may be suitable for plants that can deal with indirect sunlight, but it is not ideal for Cannabis."

http://i.imgur.com/5sCX9NX.jpg
http://i.imgur.com/KDI9NNX.jpg
http://i.imgur.com/cu2IsVO.jpg
http://i.imgur.com/0sJiDxs.jpg

Uhhhhhh..... what? Speaking as a medical cannabis user and grower, and as a landrace genetics preservation specialist for a Dutch seed bank, you're totally, absolutely, utterly wrong.

Let's take one of the crops I've done. Wheat fodder grass for animals. Using current traditional soil methods, one acre of fodder grass will require 100,000 gallons of water. Most of this water is lost via transpiration of the plant and evaporation from the soil.

I can build you a 1/8 acre building, load it up with recirculating vertical-stacked NFT channels, and you could produce that same acre of grass using 1,000 gallons of water.

http://www.youtube.com/watch?v... - have one of the UK projects I worked on.

If you consider a strip of metal barely rigid enough to support the strip a heat sink, sure.

I don't use heat sinks unless I'm dealing with insane power. Like 1,000w in a 70mm x 70mm package.

"And she's been unwilling to loan me a unit for test purposes, dammit ;) "

Your sister does my exact job.

What sort of unit are you looking for? I've got tons. General purpose, microgreens, hell I've got technology to grow some plants without any light at all (very useful for the cattle/sheep ranchers) with a 99% reduction in water requirements.

"If you've got evidence to show, please share....our electric bills will thank you"

Sure. http://ledgrowersforum.co.uk/

How about a cola as big around as my leg?

"LEDs aren't there yet. You can't get cannabis to flower properly under LED lighting, nor can you get the sort of growth rates you'll get under HID lights"

Yea, explain why I made it on High Times Pix of the Crop six years in a row using pure LED lighting.

Or how about you go roll over to http://ledgrowersforum.co.uk/ and learn how dead wrong you are.

"The parent post already stated HPS and LED lights have near the same efficiency..."

The parent post is wrong. Wikipedia is a horrible source for the rapidly-evolving world of semiconductor lighting. I've already got LEDs with 200+ l/w efficiency (Cree MK-R.) In single color LEDs, going to equal weighting and ignoring the weighting of green, you might as well just say 330+l/w, which is double any HID (excepting the almost pure-green low pressure sodium lamp, which can do 220+ l/w.)

"HPS is in the >100 lumen/watt ballpark"

Only in their more inefficient ranges like 250 and 400w. 600w can get up to 160l/w.

"LEDs, while capable of much more efficiency when operated at currents below their maximum ratings, usually operate near the same 100lm/W efficiencies when operated at maximum current"

Current LED tech is 130+ l/w and Cree's already popped 200+ at room temp 5600K 80mA drive with their MK-R, and in the lab, they've already hit 303+ at room temp 5150K color temp and 350mA drive.

" But a 400W LED fixture would produce nearly the same heat overall. It just wouldn't get as hot ;-) "

Comparing 400w HID to 400w LED, full power, the LED will produce on average 15-20% less heat, with that heat difference made up as light output.

"apparently you have no idea just how much heat LED arrays put out."

I design LED systems globally, son.

Try again when you know who you're talking to - THE LED expert on this site.

And a hydroponics light is a bullshit term because there is no specific light for hydroponics alone.