Somy Nex said:
...I also don't see anyone coming up with a clear cut definition/methodology of measuring runtime output...i see no reason why fenix should be singled out....if some of these were addressed, maybe then this thread, despite the very useful discussion within, would have more overall worth in my eyes.
One reason why there's been no suggestions is many people aren't aware of the factors explained in the previous posts, which is why this thread has educational value.
You're right Fenix is no worse than many, also that Surefire isn't more accurate, they simply de-rate their numbers. We know from much experience, that various examples of a given Surefire light can have large variations in output due to manufacturing variation.
Once you understand the issues (which this thread will hopefully help with), it's actually not hard to
suggest an improved lumen definition for consumer flashlight use, or an improved lumen testing methodology. What's hard is devising one that's practical and not burdensome for all manufacturers, and one that achieves adherence in a global market.
E.g, you could just say "all mfg should have an integrating sphere and test every light" according to xyz procedure. Assuming you could even mandate that in a global market, that degree of regulatory intervention would stifle many mfgs. Also what body would enforce that, investigate violations, make rulings, etc?
One possibility is for ANSI, ESO or some other standards body to produce a definition for flashlight lumen ratings and a test methodology, which manufacturers could voluntarily use. To solve the "my light doesn't produce that output" problem, the standard would also have to define allowable manufacturing variation, or what methods are used to ensure manufacturing uniformity.
That way the mfgs that can afford it would follow the standard and get to use the "ESO lumen" nomenclature, the others would continue using the current ad-hoc lumens. Some means of compliance enforcement would still be needed.
It's more complicated than first appears, which is why this thread is good, as it spurs thinking about such things.