I’ve been thinking about changing that too. With the manual memory option added, it seems like that should be easier to access than the config menu.
So, while on, perhaps something like:
3C: toggle ramp style
4C: enable manual memory at current level (“lock” memory at the current brightness)
4H: go back to automatic memory
5C: config menu (this would be changed everywhere a config menu exists)
Also considering remapping lockout mode’s aux LED controls from 3 clicks to 7 clicks, to make it consistent with the way that works in “off” mode. Or perhaps switch to some other method entirely for configuring those.
Same here. That’s why I’m not sure. It seems like something which could use some discussion.
That’s an interesting idea. I assume it would act like holding from off, meaning it would ramp up from the floor?
It was like that originally, but people were concerned it was too easy to exit. I guess one of the first things new users like to do is open up the light to look at the battery, which would take them out of the simple mode.
I kind of liked the original muggle mode better. It was on/off only at ~130 lumens, with loosen-tailcap-to-exit. A 3-step version might be good too, like 10/80/250 lm with click for on/off and hold to change brightness.
Ordered a new Robin’s Egg Blue D4S today and asked Hank if he could also supply one of the keys… we’ll see how that goes. In the meantime, I have keys coming from Oshpark and a new USB ASP so it can be dedicated to this key… pins also on the way. Between contactcr and myself we’ll get some of these made up as soon as the slow boat allows us to. (I actually did pay expedited shipping, we’ll also have to wait and see how that pans out)
Aside from thousands of hours of flight tests, and tens if not hundreds of thousands of hours of subsystem tests across the different systems, the MAX also had (napkin math estimate), somewhere in the range of 1/4 to 1/2 million hours of in service use before the problem manifested itself.
That would not have been easy to catch in flight test, unless there were a specific risk identified to test for. But if there were a specific risk identified to test for, it could also have been pre-empted in the development phase. That would have involved the failure analysis fully understanding the interaction between the hardware, the software, and the pilots to recognize the potential for what was intended to be a small nose down adjustment to iterate multiple times into a large nose down adjustment without the necessary response being recognized, and reach a point where it becomes physically difficult to recover from.
The investigation will likely be looking quite extensively into why that risk wasn’t identified.
There is a parallel because it would have been possible to design a test programs for the D4v2 that included leaving a light off in muggle mode for a sufficient (whatever that means) length of time, with batteries in differing states of discharge, but tests targeting such specific, difficult to predict faults don’t usually come to mind without an extensive FMEA effort. That level of effort is supposed to happen for aircraft, but that’s a far more safety critical case than even a high output flashlight sold only through two specialty shops not frequented by the general public.
Even with that level of effort, while it could credibly catch an effect like performing a voltage check in sleep mode causes the firmware to also do a temperature check in sleep mode, I’m skeptical a reasonably diligent engineer would identify a risk that temperature measurements in sleep mode might be inaccurate, much less that this would only occur below a certain battery voltage. As a result, it would be highly unlikely to determine that a muggle mode resting state test needs to be conducted at a range of battery voltages.
Well, if I pretend to order again, it comes up with 0.8mm by default. My original order details don't give the thickness, so I'm not sure I was shipped the right PCB thickness. In any case, doesn't matter now - I worked around it.
I can’t tell by my OSHPark email if I ordered the correct thickness or not, they’re in fab as of a couple of days ago. Pins and USB ASP due here in a week on the expedited shipping, I have a tracking number and will be watching it (If I can remember to) [Not to worry, my phone will be watching it and let me know if/when it moves]
For others looking at this. If you do not explicitly choose the “2 oz 0.8 mm” option (with a check mark), you will get a 1.6 mm board. Same goes if you do not choose any options at all, or choose another option (such as swift service), you will get a 1.6 mm board. The lead time on the “2 oz 0.8 mm” boards is 2-3 weeks. During the order wizard you may see the “2 oz 0.8 mm” option and think it is the specification of your board, but no, it is an option you can (and must for this) choose.
Thank you for posting this, I’d have been disappointed that I had to wait again… at least this way they’ll start making the correct ones now instead of a week or so later
I ended up with the boards the wrong thickness too. I don’t have your soldering skills, so I’ve ordered another set in 0.8mm and will just have to wait again.
I keep learning new things. Just not always early enough in the process. :-}
Please noooo. I like to be the same all my Andúril flashlights user interface and if some big change happen I need to disassemble all of them and reflash. Ok a few one has programming pads but I have more than 10 without and some of them really hard to open.
The only way that a flashlight like this is ever “safe” is when it’s mechanically locked out or the battery is removed. That is still true with the simplest or most complex of operating systems, and with or without the muggle mode software glitch.
If you want a “safe” flashlight you should go buy something from the local hardware store…but even one of those has melted a hole in my living room rug when kids got a hold of it. Maybe something from a toy store?
Just my own observation, since I now have several lights running NarsilM and Anduril… but it seems the timing between the clicks of the D4v2 needs to be faster than in many other lights (eg. D4S, SP36, GT Mini, FT03, PL47). I also had a bit more difficult time configuring the aux-LEDs initially (requires faster clicks than I’m used to with the other NarsilM/Anduril lights).
I’m not saying I don’t like the faster clicks, just that it felt a bit different that what I was accustomed to. Was the timing in-between-clicks set to a faster timer for the D4v2 compared to the previously-released flashlights? (or maybe changes in a recent Anduril version?)
GuppyDrv - 8 clicks to config - Great!
H17F - 8 clicks to config - Great!
Guppy3Drv - 10 clicks to config - Even Better!
Emisar D4 - 10H to config mode - Great
Biscotti - 10+ clicks to config - Great!
LD-4 drivers - 10+ clicks & password option to config - Superb!
Anduril, RIOS - 4 clicks to config - Too easy to access and change the levels/ modes.
Then you either hit the button again once in config or have to wait for the sequence to finish if you realize.
Most annoying.
I’ve asked you why but you’ve given no logical reasoning why you’re so adamant to have it set up like that in Anduril.
I could be wrong, I know I don’t think like most other folks here but it makes no sense to me.
Personally I would have ramp config set to 10H clicks, keep it out of the way and use the 4 clicks from off for something used more often.
It’s your puppy and I won’t say any more on the subject but thought you would be open to sensible suggestions on improvement.
(I actually did pay expedited shipping, we’ll also have to wait and see how that pans out)
[Not to worry, my phone will be watching it and let me know if/when it moves]
