I'm suffering from a misapprehension
 

There are any number of sources that will authoritatively state the number of lumens that a tungsten fixture will throw per watt but none of them agree so let's just pick a number... say... 14W.

Litepanels claims that their Sola 12 Fresnel throws the same number of lumens as a 2,000W Fresnel fixture (e.g., Arri ST-2).

...I'm not too sure why they highlight their competition as a tungsten fixture when they're competing against HMI fixtures but let's move on...

Now, if a 2,000W Fresnel throws 14 lumen a watt then a 2,000W Fresnel would be throwing (theoretically, approximately, theologically) 28,000 lumen.

If the Sola12 is using the Luminus PhlatLight SST-90 emitter (and I'd bet a pint of antibiotic free O+ blood that it is) then it's throwing somewhere in the neighborhood of 2,000 lumen.

Clearly, I'm seriously messed up somewhere!

It's only one decimal place... what's the big deal? ;)

There is no way that 250W of LEDs are putting out only 2,000 lumens. 25W of LEDs puts out more than that. This fixture likely puts out more than 25,000 lumens if it has the same power per watt as other LED lights.

The problem here is trying to use lumens to compare. I wrote about this in my blog over a year ago when we first released LED fixtures. Its a bit more complex than can be written about in one post here though but I'll take a stab at it.

You simply cannot compare an LED fixture to a traditional fixture using bulbs and lumen output as a measure. Lumens were a measure designed for bulbs which have a 360 degree pattern of radiation. This is measured in an integrating sphere. I know of nothing else that can measure lumens. There is no reliable conversion between lumens and lux / footcandles either despite what some people say.

While LED manufacturers certainly will quote in figures in MCD or lumens for their products, it doesn't really tell the whole story and really isn't appropriate. An LED isn't a bulb, its a tiny micro-fixture in and of itself, complete with source and lens and sometimes even a mirror. Since each micro-fixture is too weak to do much on its own, we put them in arrays like pixels on a monitor (one pixel on a monitor wouldn't do you much good either). The combo of all those LEDs forms a light that is extremely useful. Sure there are other issues (like shadow rendering) with this but that's not the point we're on right now.

Once you get to real fixtures you don't use an integrating sphere to measure output, you use a light meter. A fixture is a system. Bulb (light source), augmentation and enhancement of output (mirror and or lens), etc. It converts a bulb radiation into a focused beam with more strength. You have a finite amount of photons coming out of the bulb and the "system" helps focus those from other sides of the bulb all into the same output or your "beam" is another way of putting it. Thus more and stronger light to be focused in the direction of a subject being lit.

This is the important part to prove what I'm saying: Two or three fixtures using the same bulb may have wildly different output based on how efficient the system is. A 575w par and a 575w fresnel will have different output figures in lux or footcandles based on light meter readings comparing both. They have totally different bulb/mirror and lens configurations too. Then take a 575w follow spot, even stronger if they did their job right. But they all still use the same 49000 lumen 575w bulb. That's how you know this isn't an appropriate way to compare. You would only use lumens to compare bulbs when shopping for bulbs. Hardly anyone uses a bare bulb on the end of wire in the real world to light sets / subjects. Most people are using fixtures of some kind.

Thus, its only appropriate, if you really want to know what's going on, to use a light meter to measure the output of two fixtures to compare them in lux or footcandles and not lumens (which you have no way of measuring anyway). There is no instrument you can use to measure lumens short of buying an integrating sphere. Even if you did, no one sticks a fixture in an integrating sphere because its simply not the accurate way of reading output given that the fixture converts the bulb into something that no longer radiates in 360 degrees in the outside world (outside of fixture that is). Just think, if its got a beam, its a fixture and requires different measurement standards then if it radiates everywhere like a bulb.

Richard, do you know what they are accomplishing with putting LEDs in a 12" fresnel fixture? I thought a fresnel needs more of a point source to make a beam.

As far as I know, the fixture is still in development, at least thats what they have been telling people and that's why final figures for CRI and lux / FC output have not been posted. Also as far as I know, no one has seen the interior of the thing so its hard to know what kind of LED(s) they used--in other words, 1 very large one or a matrix of smaller ones. Certainly if it has one shadow output, and uses multiple LEDs, they found a way to lick the multiple LEDs producing multiple shadows issue.

I think this is just an exercise in trying to be the first. At that price of course it will be confined to those that would spend large money to buy mainstream HMI fixtures. We wanted to do this about 3 years ago but it was going to be too expensive for the size and output. Certainly not HMI expensive but still expensive compared to our CDM 150 fresnel for example.

The B&H page for the Sola lights originally said "Available end of July" and now reads "Available in September"

... and as of Aug. 13th B&H now says...

probably you simply need to read this
http://en.wikipedia.org/wiki/Lumen_(unit)

IMO there is not a means of easily comparing cine lighting fixtures, and... IMO... manufacturers are perfectly happy that there is not a means of easily comparing cine lighting fixtures.

This is not to say that manufacturers could not get together to devise a means of doing so, just that they benefit more from obfuscation.