I don’t think we have numbers 30% lower than most other people. Maybe 10% or 15% is more accurate.
It was TA’s old sphere calibration that was about 30% too high. He calibrated his tube based on raw led specs, but leds are not labeled so the seller might have sold him some cheap, weaker ones thinking the he will never know the difference.
Proper integrating spheres start at around $10,000, so very few people have them. A couple people, like PFlexPRO, has one, but his is not calibrated with a proper reference source. He sort of guessed to get it calibrated based on many factory lights specs. That’s not good. He may not be too far off, though.
Most people on the forums have homemade spheres. These are not calibrated “properly” either. Certain members like DB Custom, Tom E, Robo819 and RMM (Richard from MTN Electronics) all use spheres built by Manxbuggy1 (which seems to read a bit high, maybe 10–15?). They are also calibrated based on many stock lights factory specs instead of from a certified test light like Maukka uses.
Some members also have a JoshK sphere. These came out a couple years ago. I’m not sure what they where calibrated with (I need to look that up). I have one and it reads abut 3% to 4% lower than the TA Tube. These spheres are pretty complex and seem to be limited to about a thousand lumen. There’s also a lot of math involved in using it.
The TLF forum in Germany has a member with a proper integrating sphere. I can’t remember who off hand. Maybe Kof3?
Then we have Maukka in Finland who has a proper setup.
I guess my point is we need to find out how every reviewer is measuring their lumens so we know whether to believe it or not.
Back to your point, I wish we could all get together and establish some type of calibration standard so we are all on the same page.
So many lights have been measured that it’s hard for people to go back and say “Oops, these numbers are all too high”. It might even create a “Lumen War” which nobody wants.