I have ordinary stepped lights (L/M/H etc) and I would use one of those if I absolutely must know the runtime.
Unfortunately the ācoolā factor of the ramping UI is somewhat ruined by the āblinkā. It gives the impression the light is glitching or the battery is struggling.
There are some people who really want fully regulated drivers and call FET drivers āarchaicā or ājunkā since the top end output is dependent on state of charge. The blip in the ramp gives all users a reference to know that below this point, output is fully controlled by PWM of the 7135. Above is PWM on the FET, and is therefore subject to field conditions.
To continue your sports car analogy I would say that putting fully regulated drivers on these lights would be like putting a governor on that sports car, thereby limiting the output.
āyes,this LED could output 20% (10 - whatever the number may be) more light with an open FET, but then we wouldnāt know how long it would run - see how much better life is with a regulated driver?ā
I also think so. So far there have been more people who either didnāt know what it was or simply didnāt like it than ones who liked it.
I rarely take spare lights. The situations when Iām in the risk of running out of energy seem to be the same where weight is critical. Well, thereās always a cell phone anyway.
The tradeoff is: either know well how much runtime you have or have as much runtime as you can (by using as much light as you want and nothing more). Itās a flawed one. As Mike C noted, this can be resolved by checking voltage periodically. Though I donāt like the elaborate effort involved. Turn off. Wait a second so the light doesnāt notice a multi-click. Check voltage. At that point you know the voltage which you have to translate to energy which in turn you have to translate to runtime. In practice each translation must be approximated.
Current indicator helps with one approximation. I think there should be a better way of deriving runtimeā¦
Despite I find the blink āawkwardā, and despite I regularly check the battery to avoid being āout of juiceā (even carrying a spare on a holster 
), I find the blink useful for the mentioned purpose: to mark the point when battery rain will increase!
If it was user selectable - without the need to āprogrammeā - it would be nice, but I can live with it and that is another thing that makes the Emisar D4 my EDC light since I got it! :+1:
/\ +1
Those look great! :+1:
Can you share photos of your other weathered lights?
My red Convoy S2+ spent some time tumbling in the dryer inside a small bottle of gravel, broken glass, and screws.
I have a C8 with this weathered look. I putted it in a bottle and spinned it with a drill at low speed for an hour and a half with rocks
thatās kind of like buying jeans with holes in them. To each his own. I kind of like that look 
Or like buying an AR-15 and then throwing it down the driveway a dozen times and telling people that youāre an āoperatorā and your shit just gets used a bunch.
I use my stuff, but I donāt want to wig it up, just to impress the ladies.
Chris
The guitar industry calls it ārelic-ingā and I often like the results.
These lights look fantastic and Iāll probably do something of a similar effect on a few of my lights to blend in a couple larger dings Iāve made.
Fascinating! 
After the D4v2 came out, I made some from-scratch firmware for it, with a few different interfaces available. The main one is called Anduril, and can work both ways ā smooth ramp or stepped levels. Switching between the two is quick and easy, just three clicks while the light is on. So you can have smooth ramping or precise runtimes as needed. It also has compile-time options to enable/disable the blinks at the top/bottom/middle of the ramp. Does that help?
I like the blinks and they donāt bother me, but Iām not as nit-picky as some here.
If anything, TK has saved me valuable seconds standing there with my thumb on the button, wondering if Iām at the top, or bottom of the range. Over the lifespan of this light, the blinks will buy me more time for other thingsā¦like women.
For this, I salute you.
Chris
Very useful to know, thank you.
Thanks, it was actually your red S2 that inspired me to add the weathering, and to build my latest project. Iām making a bedside light with a single 219C and a lighted tailcap. Have all the internals ready, just waiting for that red host to arrive
Iāll take some pics of the other lights Iāve done this to later, but i already have an album of my last project here
Can clearly see your more extensive approach gives a more realistic result, than my sandpaper only method.
Gotta give the bottle/tumbling a try :+1:
nice work on the table
this chart is approximate from your data
CuTi w nichia starts out 8% dimmer than the alum w nichia, but after 20 seconds the CuTi is 50% brighter. The CuTi also lasts 50 seconds longer before it dies.
now about those blinks:
They mark the brightest level at which the light can maintain constant regulated brightness (~120 lumens)
The blue line (consistent flat regulated brightness), is just below the blinks.
After hearing all the recommendations to use high charge non protected cells, I bought some Sony VTC6 from imrbatteries.com. Surprisingly, I detect zero lumen gain compared to the Panasonic protected NCR18650Bs that I bought from Gearbest. I swapped them back and forth and charged the VTC6 to full capacity and still no detectable difference in light output. I then tried comparing with my Astrolux S42 and also no difference in detectable brightness. I canāt understand how that can be the case. The NCR18650Bs are supposed to go up to 5A only. Is it possible that it can run at much higher amps than 5A?
Got my Titanium D4 this week. My initial impressions:
Pros
Cons
Thanks Jon-slider for the very useful comparison graph! I wonder would the cu/ti make the XPL HI last longer also. I love my D4 XPL HI 4000k. The tint is the best Iāve ever seen. Even better than my nichias.
Itās possible that you have a defective D4 or a defective VTC6. Or maybe your eyes isnāt that sensitive to the output difference between the two cells.