Thanks and you have to be right according to physics 
. I’m just trying to understand the magnitude of current that actually passes through the emitter. Is it 99.9, 99, or 90%?
At the numbers we’ve seen, I was doubtful that the SST-40’s would survive. Current doesn’t disappear, of course, but it does take alternate paths, where available. I can accept that driver current losses are perhaps so minimal as to be irrelevant, though, if that’s the case.
It’s hard to understand how anyone here could be measuring more than ~9.2A without burning the bond wires within a minute or two. koef3 had the test emitter on a fan-cooled heat sink, so in an actual flashlight the temperature should rise further and faster. However, we don’t actually know the current with the tail cap installed. (nkresho’s rig surely comes close)
Still, as pointed out by others, no one here has burned an emitter… yet. Maybe it’s inevitable?
Yes, koef3’s tests were on an earlier run of emitters, but could they substantially improve the bond wire capacity without a redesign? If improvements were made, the product page and datasheet would surely show higher limits. They continue to sell these as 5A emitters.
As noted by koef3 in the test, the SST-40-W datasheet is still labelled “Preliminary” and is dated 2015, so Luminous doesn’t care so much about product information. As others have stated, there’s still no part numbers for temps below 6500K in the datasheet. As nothing’s changed in the document, the datasheet still lists 5000mA as the maximum.