A well defined set of proportionately scaled modes is a most standard configuration which can fit everyone. For a 2.5x order of magnitude and 5 modes total (4 plus moonlight) that would be, in percentage: 0.2, 6.4, 16, 40 and then 100% (2.5x-fold the previous mode driving current ensures at least doubling the light output). And bullshite anything else LoL! :-D
I noticed that no light serves my major use pattern well…
With e-switch lights I use physical lockout a lot. Light is on, I unscrew the tailcap to turn it off. When I want it back on I screw the tailcap back, find the button, turn the light on.
If the light turned on as soon as I screwed it back I could avoid the last 2 steps.
OTOH it would prevent using shortcut to moon and blast me and everyone around with whatever the default mode is.
I had literally 1 e-switch light where I didn’t use physical lockout. Not that it had one….but even if it did I wouldn’t have to.
Nitecore Tini. It has that annoyingly long click to turn on. I hated that. And it would turn on by itself from time to time anyway.
But maybe a more elaborate turn-on sequence would serve as well as a lockout? 2-click on?
I have lock out feature in my firmware, and with the ATtiny1634 I have the parasitic drain down to 1uA which is theoretically over 3000 years with a 3000mAh cell. I haven’t checked if the CN regulators leak any current yet though, but if they don’t there shouldn’t be a need for physical lockout.
I do detect weather power is back after cell replacing in E-switch only configuration. Currently I have it set to blink a confirmation that the cell is back, then the light goes back to sleep. I haven’t decided weather it would be better to just have the light turn on again, but just as you mention it could blast everyone around.
For the actual lockout I am considering multiple options. One full lockout that requires full unlock sequence, another partial lockout that requires simple sequence on each startup until full unlock is done. I had to have lockout on my BMF SRK because it almost started a fire in my backpack during a mine exploration, but it’s tedious to have to do full lock/unlock each time I take it out of the backpack for photo shoot (it’s off-switch light), so a semi-lock feature would really be helpful in these situations.
Also, an update on the drivers. I have built them now but bumped into an issue for clicky switch lights that I’m trying to solve without new driver design. Doesn’t effect E-switch configuration though.
Also, still working on putting together the firmware. I’ve implemented support for quite a few of the suggestions here, and it’s required me to re-write how my firmware works to support these options, so the suggestions have been appreciated! Keep ’em coming.
Suggestions?
Keep it simple, or as much as you feel fine keeping it. I understand certain people may be accustomed to a lot of driver features but that poses extra work which could pose a serious development hurdle which the majority of users won't really care about. I prefer it Haiku style, it already annoys me grabbing a C8T/C8F and needing to go to Sofirn's website or this forum with my smartphone because I do not @#$%ing know how to switch the group modes or whatever.
In any case, do it as you feel up to.
Cheers ^:)
Ahh, but the whole point of this giveaway is to get suggestions that I like, and then see if I can make them… for myself…
I’m not making these drivers for the masses, I’m making them for me… This thread is only about a driver development junkie looking for a fix, and paying for product with a few drivers 
Wellp, optimize then access and operation of functions in order of frequency of use. A well scaled mode set and battery check already does for me. Plenty of room for other's suggestions. O:)
Cheers ^:)
Woke up today and got idea of how to avoid seeking the button while avoiding turning on when it’s not needed.
But it’s one of those things that user won’t figure out by themselves within 2 minutes of playing with light. A bit of hidden complication.
The idea is: implement a simple twisty / reverse clicky UI. Unlike most clicky UIs - don’t turn on when the power is on but upon short disconnection. Bonus feature: add more than 1 shortcut like that and or allow more control of the light with just a clicky.
So you have a lighted tailcap that is on while the light itself is in sleep mode? I’d see this more of a lockout feature that only requires a short tap to unlock. I could certainly implement that as one of the lockout modes.
That comment was a followup on Driver giveaway: Constant current 17mm drivers, winners (finally) announced, post #2. - #69 by Agro
Unless I expect that I will use a light again soon I turn it off by unscrewing the tail to physically lock it out. I’d like to be able to turn it on by the tailcap only w/out having to seek the e-switch. As discussed, just turning the light on upon screwing back would be a disproportionate response as it prevents f.e. going directly to moonlight to minimise disturbance.
I believe that with dual-switch lights it would work the same - you normally have the light clicked off. You click it on. What happens? With a regular clicky-only firmware the light turns on. Quite a few dual switch lights do the same. But that’s just the unwelcome choice for e-switch only lights.
But user could click on and then make a short tap to turn on. That’s not something obvious but once you learn it - should be easy to remember.
And would enable basic operation with just the tail switch. Or just the tail cap.
I don’t know how are lighted switches affected.
Oh, so you mean turn on the light by emulating an off press by quickly tapping the entire tailcap?
With tail reverse clicky - just tapping the clicky.
With solid tail cap - quick quarter-turn to break connection and quarter-turn back.
I don’t think I’m fully understanding… why would you need it with tail reverse clicky?
Same use case: avoid having to find the e-switch when I have a finger on the clicky already.
Ahem, no, that would be about 342 years, if you check your maths.
But whose counting ? 
Oh, so this case is for dual switch lights, not single switch clicky.
I basically have this done, just have to add it to dual switch configuration and have it as a setting. I would not make this default standard behavior but rather an option that can be enabled/disabled.
Yes, it’s for e-switch lights with either reverse clicky or solid tail.
Oops, missed a zero and a point… Attiny1634 draws 0.1uA in sleep mode, not 1uA. The years are right, the amp draw wasn’t. My mistake.
Great work, Mike! Did you get the issue with clickies figured out? What was the issue?
And may I ask where you sourced the CN5710/5711 from?
Just got back from more traveling, things have been still for a while but I’m back at it. Still working in clicky functionality. I have the major issues worked out but still doing stress tests. I really can’t stand issues that sometimes might happen, I wan’t full reliability on off time presses.
The initial issue is that the more pins I have outputting to the strings of regulators, the faster the OTSM cap is drained. 47uF was fine when only one or two regulator strings where on, but more and the OTSM functionality started to fail. Took a while to figure out what the issue was. I’ve gone up to 100uF and that issue appears solved but I am still working on the OTSM code for some circumstances.
I get the CN571x regulators from here: https://lcsc.com/ So far they have been nothing but fast and reliable. Gotta tip my hat to Schoki, he was the one who tipped me off about these regulators and where to get them 