Timber writes: ...we de-rated our flash units by 25% because we dont believe that manufacturers are being, well, accurate with their specifications of power in watt seconds. This means that if we look at a 1200ws pack, we are going to rate it at 900ws. How do manufacturers probably work the numbers? Oh, they probably do a meter reading from the very center of the light spread. In the center, the very center, there is some unevenness wherein there is likely a hot spot. Instead of taking a reading 1/ 3 out of the center, the manufacturer probably uses the hot spot to represent what the power rating of the light is. To see this unevenness, take a flash light and shine it on a wall. Flashlights create obvious hotspots. first of all watt-seconds is a measure of potential electrical energy and has only the loosest of connections with actual levels of light produced by an electronic flash unit. I have a roughly 18 year old Balcar A2400 pack which I use with the Balcar "U" heads. This combination puts out twice the amount of light of the the Norman P2000X pack and Norman LH-2400 heads the Balcar replaced. I have a couple of Elinchrom heads I had modified to work with Balcar packs, the Elinchrom/Balcar combination puts out 1.5 x the light the Norman pack and head combination did. I also had a Speedotron Blackline 102A head which I used with the Norman pack (i had an adapter cable made for this purpose.) The Speedtron/Norman combination was 2x brighter than the Norman/Norman combination. I tested with bare tubes, standard reflectors, an umbrella, with the light turned away from the meter and bounced into a large white surface and also with the same softbox (a Plume Wafer 100). All of the heads and packs had recently been serviced and had new flash tubes. I also had at one point a pair of elinchrom EL1000 monolights. These 1000 watt second monolights put out as much light as the Norman P2000X /LH2400 combination when the Norman pack was set for full output. Furthermore the elinchrom monolights were within a couple of tenths of a stop in measured output as Elinchrom's literature said they would be. All of the test were conducted in the same set up and studio over a two day period. As many variables as possible were eliminated. My point is you cannot just say "we dont believe that manufacturers are being, well, accurate with their specifications of power in watt seconds" as a flat, all encompassing statement. My testing show that a large factor in the efficiency of turning energy (watt-seconds) into light (lumen- seconds) has to do with the efficiency of the head design and flash tube design. But on the other hand it is clear that some manufacturers actively skew their results, or at least report it in their literature to make their lights look better. They do this by publishing numbers measured with a narrow angle reflector being used, or as Timber points out only measuring the hot spot. (I eliminated this as a factor in my tests by measuring the light using a large range of reflectors and light modifiers) , But the better brands: Profoto, Elinchrom, Balcar, Dyna-Lite, Speedotron, Broncolor, Comet, maybe Hensel -- do not take this route. Perhaps other manufacturers are also as honest but I haven't tested them. As a general rule I would be suspicious of any make that makes claims of "effective watt-seconds" or does not specify the reflector being used.