Nowhere in SCUBA advertising do you get more misleading, misunderstood, exaggerated and downright wrong information than in the realm of underwater lighting. There are only a few national standards on measuring light output and almost no one uses them or even bothers testing the light output of their finished product at all. For starters it is impossible to represent the output of a light source in a single number. Even the commonly quoted lumen output is a poor representation of total light output.
The first thing to understand is that all light sources, depending on the technology they use, have variations in output across the visible spectrum of wavelengths referred to as their spectral power distribution. Some light sources have an even, continuous output across all wavelengths. Others have spikes and troughs at different frequencies. This popular mechanics article hase a decent explanation of PSD curves:
PM's Ultimate Light Bulb Test
The Color Temperature is a way is stating the average output across the entire visible spectrum and relates the measured output from a given source to that of an ideal black box radiator. A black box radiator is any object really that will give of energy in the form of light as you heat it. The hotter the object the whiter the light. Sunlight for example corresponds to a color temperature of 5600 degrees Kelvin. Again, it is an average number so it doesn't really tell the whole story. For incandescent lights, which are close to theoretical black box radiators, it is very representative. For some of the other technologies which are not as uniform it is not as good of an indicator. Color Rendering Index or CRI is a value which attempts to quantify how far a real world light source deviates from a theoretical ideal one. A CRI of 100 is perfect and uniform, lower values are worse meaning there are some wavelengths that are not as uniformly represented (but doesn't identify which ones). Why does any of this matter? Well photographers and videographers who are trying to accurately capture the color of objects have to illuminate them with a light source that carries the light spectrum they are trying to capture. If my light source performs poorly at the blue end of the spectrum, blues are going to be underrepresented in the final image.
The next thing to consider is light output commonly measured or quoted in lumens. This is the most frequently discussed value and also the most misrepresented. Although lumens is supposed to be a measurement of the total output in all directions what is frequently quoted is a peak output which can be very misleading. Few light sources in practical application emit light in all directions. Usually real world dive lights are designed to concentrate light in a single direction. That pattern is typically described by a beam angle but even within that beam either due to limitations of reflector design and manufacture or by intentional design the light output over that beam will vary. Some beams have a "hot" center beam and less intense periphery before dropping off to nothing. Others will provide even illumination across the entire pattern.
For a discussion on measuring the total lumen output and beam angle see these Light and Motion articles:
Understanding UW Light Specs & Testing / Light and Motion - Light & Motion
Setting a High Standard: LEDs and Color Rendering / Light and Motion - Light & Motion
Without an understanding of the pattern, a comparison of ones manufacturers lumen output to another's is essentially meaningless. Are they quoting a beam peak or a true average? Sometimes you might see both numbers. In some instances the comparison itself might be meaningless. A cave diver might prefer a beam that has a high center beam output for signaling his buddies but the same light would make a poor choice for videography.
The last thing to understand is burn time. In the days of "dumb" lights the output of the light would drop off as the battery was depleted and the voltage dropped. Most modern lights come with some form of electronics to regulate the flow of electrons from the battery to the light. A common method is to design the electronics to compensate for battery voltage drop and adjust for a constant light output over the discharge of the batteries. Final duration is typically determined by a cut-off circuit designed to protect the battery cells from deep discharge. What recent testing has shown is that some manufactures have cheated a bit and step down the output of the light over time to meet a longer burn time. The step-down is subtle enough that it is difficult to detect but measured output falls off considerably. This might not be important on a night dive to the diver/observer but might be a noticeable factor to someone shooting video in consistent low-light conditions like a cave.
So where does that leave the average consumer? No better off I'm afraid. It is a complicated topic that is impossible to boil down to a single number. Don't worry too much is one manufactures lumen claim is a few lumens higher or lower than the competition like the OP. Chances are they are using the same components from the same manufacturer. Do a pattern comparison against a blank wall if possible so you have some sense of how broad and uniform the bream is (or narrow and concentrated if that is what you are after). If you are shooting video try to understand those PSD curves a little. Seek out testing from a independent source if it is available.
