I didn’t read the whole thread, so sorry if someone already posted this, but i want ability to set stepped ramping in classic mode order. Low-medium-high-low-medium-high. No bullshitting with click and hold when i want change direction.
I see no problem with reverse -click and hold, it can stay, but carousel is interesting option. And programmable ON mode option would be nice too.
Can always hold from off which by default cycles in that direction. Pretty certain it cycles in that direction when holding from memory too?.
Like TK has said, she’s making it fit into what most prefer, can’t do everything to suit every single individual as that would be a huge job to code.
Ah, makes sense. What was your old account? The admin would probably help you get access to it if you asked.
Fortunately, BLF isn’t the only place where people talk about flashlights. Anduril v1 was made for BLF… but I started on Anduril 2 because of what people said elsewhere — Reddit, Facebook, Amazon, Banggood, personal blogs, etc — places where people didn’t think I’d see it, where they pulled no punches. People outside of BLF had a number of complaints, some of which were worded with a lot of anger, but the underlying points were solid.
So I took those critiques / rants / reviews and looked for ways to solve as many as possible. Some of the main themes were:
- It was too complex for new users out of the box.
- The full UI had too much, but Muggle Mode had too little. People wanted something in-between, something more Goldilocks with all the basic flashlight functions, safe limits, and no disco modes or config menus. No weird modes to get stuck in.
- All the core functions needed to be as few clicks as possible, because a surprising amount of people have difficulty counting clicks or reliably executing click sequences longer than 3. It may seem strange, but more than one said they literally can’t count to four while clicking a button.
- Muggle mode was too easy to enter/exit by accident.
- Muggle mode was way too limited and felt like an afterthought (because it was).
- The term “muggle” is insulting.
- Config menus were too easy to access by accident.
- Multi-click functions were too close together.
- Lockout mode was good, but too slow to enter/exit.
- Some fairly specific extra features would be nice.
So I removed muggle mode entirely, and replaced it with something in-between, with the core essential flashlight functions and nothing more. It’s pretty similar to a “normal” UI from other companies, designed for people who just want to get stuff done and don’t want any weird extras getting in the way.
Tried to address the other main complaints people had, too.
However, I also encountered an entire category of complaints which I can’t really do anything about without turning Anduril into a completely different UI. The ideas people had in those critiques aren’t bad… they just require a different solution. And to address that, I’ve been working on something else:
People should be able to choose a UI at time of purchase, or change it fairly easily at home.
Because no single UI can be and do everything for every person. One size doesn’t fit all. It may be possible to cover 70% of what people want, or maybe even 80, in one program… but it’ll never be 100. But there’s a very simple way to get closer to 100%: Offer several different UIs so people can choose whatever they like most.
To help accomplish that, I’ve been pushing toward a few goals:
- Make reflashing easier. For example, with easily-accessed reflashing pads and commercially-available pogo pin adapters. This allows reflashing in under a minute with no soldering.
- Make firmware available to everyone. In other words, free software. Publishing everything, chasing down license compliance issues, teaching companies how to participate, etc.
- Make firmware easier to create or modify. For this, I created a microkernel and UI toolkit which handles the hardware details and allows people to write user interfaces in an easier, portable language which works on many different lights.
There’s still a big piece missing though:
- People need to actually make the other UIs they want.
I made the UI I like, and a few other examples for reference, and some documentation about how people can make their own… but people still need to actually do it.
Indeed, it can’t.
This is the most common criticism I’ve encountered in the second category — things I can’t do without sacrificing core functions. It’s not a bad idea, it’s just … a fundamentally different type of interface. It deserves to be its own UI, not just some obscure option in Anduril.
The closest I can potentially work in is the “Off -> 2H” button action. It’s currently mapped to something nobody really seems to care about, so it’s prime real estate for something else. It could be mapped to full turbo, or momentary turbo, or momentary mem, or … whatever.
But if double clicking is too slow, 2H doesn’t help. It isn’t any faster.
That only leaves the 1H action, ideally responding as soon as the button is pressed, for a true momentary… but activating turbo as soon as the button is pressed would create a ton of problems on most non-thrower lights. Like, literally every time the light is activated, it would immediately go to full power. Want to take a bathroom trip in the middle of the night? BAM! Now the user is wide awake and temporarily blinded.
So I’m not going to remap “Off -> Press button” to turbo.
However, that could easily be the starting point for designing a completely different UI…
That sounds like the beginning of a brand new UI. So… make it. Don’t use my silly toy interface; it’s designed for purposes which are completely different than your needs. It’s not what you want… so make what you want. The whole point of free software is to empower people to scratch their own itches instead of depending on someone else to do it.
As they should. Diversity is a beautiful thing.
On that note, in case it matters, there’s also another option: Add a clicky switch. Anduril already supports the use of a second switch, a clicky switch, for momentary purposes. It simply turns on at the memorized level as soon as power is connected, like most tactical-style lights. But it requires a physical button, which isn’t something I can provide with a firmware update.
That how I configured my HDS long time ago and never changed because I still find it convenient. Momentary activated, 1H is the lowest level, 2H is the burst level. There is a slight delay with 2H though.
But my HDS is 100 lumens max and I don’t know if I would do the same with a “pocket rocket” EDCed. I’m pretty fine with the 2 (low) levels momentary of Anduril.
Preach it, sister! One UI can’t be everything to everyone. What you’ve done with the FSM toolkit is a great way to address that.
Also, you handled that very well 
I like the improved ability of putting a strobe mode into momentary. This will be another improvement for light painting photographers, a market in which the FW21 and FW1A are fairly popular. I would like to see adjustable strobe brightness too, but that might be asking too much, especially if you are trying to simplify the UI! I have heard many complaints for light painters that it is too easy to press the wrong buttons and completely messing up the settings.
BTW - The newish Light Painting Paradise Lightpainter flashlight currently has the best UI for light painting.
In the abstract I think it would make more sense to have lockout at 3C, but I would keep it at 4C. My reasoning is that not only do Anduril 1 and Narsil use that, but other lights seem to be adopting it too. We will never have consistent shortcuts across all user interfaces, but it is nice when they develop, and is worth encouraging.
Exactly. That’s why I’m not sure. By itself, I’d say it’s an upgrade to make lockout shorter. In the context of larger trends though, I’m not sure if it’s a good idea or not. I didn’t do it last time because I wanted to maintain cross-UI consistency… and I’ve kinda regretted it ever since. So maybe I can do it this time, or maybe I shouldn’t, for the same reasons as last time.
ZebraLight uses 4C for battcheck, so it would still be consistent with some lights.
Most valuable clicks are :
1 click
hold
double click
click&hold
Using firmware where battcheck is click&hold from off, so you need to press only twice. Or you can use it for lockout
My main concern is that a locked out light should always stay locked out until I unlock it.
I would not want my light to "exit" lockout mode unintentionally while it is in my pocket or bag after something accidentally pressed against the switch a few times.
I prefer an electronic lockout (click sequence) instead of a mechanical lockout (untwist parts) unless I am storing the light for a long time period (and want to avoid parasitic battery drain).
I am not opposed to using one click sequence to "enter" lockout mode and a different click sequence to "exit" lockout mode or to click sequences that are different from other user interfaces but I do understand that consistency is preferable when possible.
If the same click sequence is used to both "enter" and "exit" lockout mode, then I am more interested in a lockout mode click sequence that may be harder to use but very unlikely to accidentally "exit" rather than a lockout mode click sequence that is easy to "enter".
If 2C is turbo and 3C is LOCK you may need turbo and get the flashlight locked…… I think It’s really dangerous
Mashing the wrong combinations is a user issue. This can be alleviated somewhat through design but ultimately the user needs to hit the right combos otherwise its not really an input output system just a mash and hope system.
Anduril 2.0? Sounds great to me! The future of flashlights is exciting because it just keeps getting better. I’m waiting for a massive shift in thermo management where higher power can be sustained for longer periods of time, like a moving diaphragm that draws in cooler ambient air and pushes out the heat. Just thinking out loud. Go ToyKeeper 
I’m still not sure about 3C/4C for battcheck/lockout or lockout/battcheck.
4C makes lockout harder to hit by accident, but also harder to use. And on some lights, it gets used a lot because the lights turn on easily by accident. So it seems like it’s worth optimizing, to make a common operation easier.
The main concern I have is compatibility. It could be a pain for people who switch between old lights and new lights. Most of Anduril2’s other changes are less likely for people to trip over, or at least wouldn’t happen as often… but lockout is used quite a bit.
I’m using a FW3A at work about everyday and always twisting/untwisting the head because it’s faster than 3C / 4C to lock / unlock it.
Hitting 3C and 4C doesn’t make a great difference pratically because 4C only needs a fraction of a second more. What would make a difference is if the flashlight was automatically turned on after unlocking it. With this feature, it would be better to lock / unlock it with 4C (harder to hit by accident).
I agree with this, I personally prefer loosening the tailcap for lockout, BUT, I would actually use the digital if it was “immediate” in the light’s activation afterwards.
For example, it would treat the 4th click as the same as 1C when you release it, and go to your memorized mode. Or, if you hold the 4th click, it will act as though you are doing a 1H and go to the bottom of the ramp. Lastly, if you add a 5th click, 5C, then it will behave like 2C and go to the top of the ramp… I don’t know if I explained that clearly, but to me, that makes the most sense as far as usability.
To summarize, I would make Lockout into more of a 3C + standard operation (1C, 1H, 2C, etc.). This would be very easy to remember because all you have to do in lockout is… whatever you were going to do, +3C (in front).
This would effectively remove a pause and 1C from the current process, and be much smoother, I think.
Anduril 2 already turns on while exiting lockout mode. It also has an optional function to automatically lock the light after being in “off” mode for N minutes.
So the question is about which of these two styles it should use:
- Faster style
- Off 3C -> Lockout
- Off 4C -> Batt check
- Lockout 3C -> On (memorized level)
- Lockout 3H -> On (ramp floor)
- More compatible style
- Off 3C -> Batt check
- Off 4C -> Lockout
- Lockout 4C -> On (memorized level)
- Lockout 4H -> On (ramp floor)
Additionally, to configure the auto-lock function…
Lockout 5C -> Auto-lock config menu (choose how many minutes before it locks)
Lockout 5H -> Disable auto-lock
… and to configure aux LEDs:
Lockout 7C -> next aux LED pattern
Lockout 7H -> next aux LED color
The faster style spaces apart the lockout mode functions so they all have a gap between. 3/5/7 clicks. The more compatible style would be 4/5/7 so the user is more likely to get the wrong thing by accident. However, the auto-lock thing could be moved… perhaps to 10C/10H. Then it’d be 4/7/10 clicks.
The slower / more compatible style also increases the chance of accidentally going into momentary (5C) instead of lockout (4C), since they would be only one click apart. It’s a complaint I’ve heard a few times, that people end up in momentary and then can’t figure out why it won’t unlock.
In any case, turning the light on from lockout is faster than before. Instead of 6 time units (4C, pause, 1C), it’s down to 3 or 4.