I have sent 37 leds in the mail on friday.
And yesterday evening I did an output test on the led (will call it the 4th gen. LG 3535 UV led). Do not expect anything as accurate as my white light tests, my method of testing UV leds has many imperfections leading to probably large errors (method here). Errors are caused by fluorescent bleaching of the used copier paper, variation of fluorescent die in the paper, but also higher than average output drop upon warming up of UV leds. On top of that the new led has a vastly different emission angle which also will cause extra error the way the led illuminates the paper. The measured voltages however are as correct as my normal led tests.
So I was quite surprised how close this led tested to the 3rd gen. version I tested 2 years ago.
The main difference is the much lower voltage of the 4th gen. led which makes even this lowest bin more efficient than the 3rd gen. that was tested, so it can not be direct driven on a single li-ion anymore but must be regulated (i.e. by a 7135 based lineair driver). And like the 3rd gen. led the sweetspot for me is 1.5A.