You would first have to pick a definition for "run-time". I started using the 50% of max brightness value for a couple of reasons.
First of all, I was used to working with "half-lives" in the nuclear business.
Secondly, the brightness of a light has to change by a factor of two for the human eye to detect the change. Also using the 50% value for both regulated and non-regulated lights tells you the same thing. It is true, that for a truely regulated light, there will probably not be a second 50% decay in brightenss.
It could be usefull to report a second run-time value that would indicate a time to really usefull, but lower level of light. Here again, a value would need to be agreed on...75%, 80%, 90% of max brightness. The 50% and 75% values for really regulated lights are almost equal!
Reporting Run-Time as time to 50% brilghtness and (say) 90% brightness would tell you if the light was regulated or not.
By the way, I became a "designated" tester after I purchased a logging DMM with a PC interface and also found out that my ISP would give me either a 5 meg web page or a 5 meg FTP site for free! Folks kept asking what the runtime was for different light, so I jumped in with both feet and started plotting runtimes for those lights in my collection. Kinda learned as I went along. /ubbthreads/images/graemlins/grinser2.gif