Hi, I’m not doing that many led tests lately but here is one more. Cree XM-L3 was introduced recently, with an increased current rating of 5A compared to 3A for XM-L2. This seems a major upgrade, let’s check that. In appearance it is different from XM-L2 in that the phosfor is not limited to the die area but spread out over the entire led, in fact it looks like some cheap chinese knock-off. Luckily they added their logo in the corner for compensation 
. The XM-L3 U4 1B led (6500K) was obtained from Kaidomain.com.
For comparison I tested another modern 5050 footprint led as well, a N5-bin 5000K Luminus SST-40 from the Convoy aliexpress store.
In the graph I included the data from my XM-L2 T6 3C test from 2014 and a SST-40 N4 BA test from 2017 to see any progress over the years.
About output bins: I’m not always sure that the bins claimed by the sellers are correct although from these 2 leds they seem plausible. And I came to care less about bin claims over time, the leds that I test are usually the ones that are commonly available for a fair price, and those are the most relevant I think, whatever their honest bin. And a better bin does not change the performance characterisics. Sometimes better bins than I test are available, i.e. Cutter.com.au has special bins sometimes.
The test was done like all my more recent emitter tests, for links to my other led tests, see my sig-line below this post. I described the method in detail in my XP-L test, with two minor upgrades since that should not matter significantly for the results: I used my Integrating sphere no. II instead of no. I, and for the current I used a clamp meter, which appears to measure 0.1A lower than the power supply current-reading that I used for my earliest led tests.
Some further remarks about my led tests: 1) just one led was tested which I hope is representative for the whole batch, reflowed on a DTP copper board (XM-KDlight 20mm), 2) I used my large version II integrating sphere with high quality luxmeter, 3) the output numbers and voltages were measured with the led close to ‘steady state’ for each current, so warmed up and settled, you should be able to get these numbers in a well heatsinked flashlight. Mind that these are output numbers of the bare led, in a flashlight there will be losses from light obstructions, lens and optic, 4) output is in ‘djozz-lumen’ defined as 1/550 of the output of my Sunwayman D40A on high setting, which I hope is close to the real lumen, but at least is consistent over all my emitter tests done in integrating spheres. Last year I checked the djozz-lumen with lights measured by maukka and the djozz-lumen seems 7% to 11% high (depending on spectrum) compared to maukka’s calibration.
Data:
I leave most of the interpretation to the reader, just a few things:
*both leds were tested to death, the XM-L3 died at 11A, the SST-40 died at 10A. I see now that I omitted the 10.5A data of the XM-L3 in the graph (I=10.5A, V=4.36, lm=2004), sorry about that. Both leds died from blown bond wires.
*The XM-L3 U4 output follows the SST40 N5 output almost exact, but with 0.2V higher voltage for most currents.
*The 3A -> 5A increased rating of the XM-L3 compared to XM-L2 has little to do with actual increased current performance, it seems just an upgrade on paper.
Conclusion: the XM-L3 seems not a revolutionary upgrade from both XM-L2 and SST-40 although it dies at a slightly higher current than the others. In general if I look at these single-die 5050 leds since 2014, not much has happened in 6 years 
:
