I do see how a Bluetooth app could make things easier. Its counterintuitive but it works if the app is good.
So I’m with you there. I’d be down with this. But the app has to be good or it’s pointless. I’m not sure it’s worth the effort. Unless it’s for a lamp there’s not many reasons you’d need to control a flashlight remotely. I would definitely use it for some things but I don’t think adding bluetooth has the same sort of usefulness in a flashlight as it would in something stationary like a light fixture. So idk if the cost/benefit really works out.
What about an app that allowed you to configure Anduril parameters, with an array of USB-C terminated programming cables for different chips. Create programming templates (maybe that are sharable to other users), set each parameter however you want, and just tell your phone what light you’re flashing FW to. This should be enough info for it to build FW with those parameters loaded. Plug the programming cable into your phone, hold the pogo pins to the light, hit the go button, wait until the app says done. Could even give push notifications for when a new version of FW dropped for lights you’re subscribed to.
And best feature: it swears at you if your phone’s flashlight is on when you’re using it
But this would give you a better UI without adding cost and complexity to the light. Alas I know next to nothing about mobile app/cloud dev lol
Naw, just have the user connect via usb and then ssh into the flashlight, then edit a config file with something simple like ‘ed’. And then Bob would somehow become your uncle.
To be fair, the Indium Smart was USB-programmable via an application. Couldn’t dig up anything on CPF, but here’s a blog article from 2007.
I do recall 4 Sevens struggling with both Bluetooth support on Android and the realities of getting RF in and out of metal casings. Still can’t find the announcement, but a quick search suggests the Navi Smart NS1 (random link - don’t suggest trying your luck with that vendor) was an actual product, albeit more something along the lines of Tile or Airtag before either got big.
For everyday use it’s useless; but it can be very useful for some uses.
Maybe someone wants to reconfigure a light depending on where the light is used, indoors, walking, biking, etc.
All different brightness sets, but perhaps also different K.
I know Anduril is hugely popular on this site, and obviously I have no problems with that, but for me a light that has Anduril, is in immediate no sale.
But if a phone means I could easily reconfigure Anduril into something like a ZebraLight or DQG interface, then Anduril would no longer be a dealbreaker.
Doesn’t even have to be using my phone. USB cable connected to my PC would be great too.
The absolute ideal for me would be that is could define 2 groups.
Each group has a number of brightness levels I can set, and be totally free to pick from 0-100% in steps of 0.5%
Then I would configure an indoor and outdoor group like on my DQG Tiny 26650. Low, Med, High, hidden turbo.
Configure how turbo can be activated, etc.
I hate stepdown, so the first thing I would do, is play with the settings to find a high without stepdown.
Then we could even have a special section on this forum to share configuration files.
I really could spend hours configuring my light behind my PC.
But when I actually use the light it should be hyper simple.
You got any plans in this direction already, or is it just a “in theory one could, but nothing planned” thing? I’m asking since I’ll be designing a driver for a light of mine soonish, and if you’re planning to swap chips for a future release, I’d take that into account when deciding.
Simple is better. Not only is it easier to use, also id imagine the cost of a flashlight/driver with built in bluetooth receiver will be significantly higher.
I’ve no clue what the price difference between the chips is, that said sometimes more complex is cheaper.
For the sake of easy explanation let’s assume there are two group of flashlight buyers.
Those that want 2 or 3 level light without turbo or other fancy stuff.
The other group are giga lumens with trickloads of settings.
With smart sortware a flashlight designer can sell one design to both groups.
That reduces design costs, simplified prodiction lines and less types on the shelf.
I recommend using avr32dd20 as the MCU for now. It’s a really nice chip and I think it’ll work well for the foreseeable future.
However, I think I might update the name to “Anduril 3” soonish even without a major change of MCU architecture. Instead, I’m thinking it should streamline a few things in the UI and expand the “simple mode” to include interfaces from several popular brands.
That’s sort of what I’m thinking for Anduril 3. It’d include a few popular interfaces and let the user choose one. Not via a phone though… just use the button to pick one.
In addition to giving people a choice of several “simple” UIs similar to other brands, it should also eliminate the need for manufacturers (like Sofirn and Wurkkos) to ship two different versions of each light which have different interfaces. They could just ship it with their own UI, and users could change it to something else with a few button presses.
Thanks! Guess I will go ESP32-C3 + custom firmware on one light for fun, and stick with 32dd for the rest. I kinda want to mess with wifi/BLE setup, and 5uA deep sleep current ain’t half bad. More than a 32dd, but I got 2S1P or 2s4P 18650s anyway.
I don’t want a light that I can no longer program when the app that controls it stops getting updated. A light is not something that should ever become obsolete, I have a 20 year old Mini Maglite that still works perfectly, I wouldn’t be able to say that if it was Bluetooth controlled via an app.
I understand that some people don’t want Anduril because they’re scared of the diagram but I like it just fine how it is. I have a few settings that I change loosely based on TK’s setup but after that I mostly use hold for moon, click for memorised medium and double click for very bright. I sometimes ramp when I want a specific brightness but thanks to the beauty of hybrid memory after five minutes it goes back to memorising my preferred level.
Given the rarity of my having to change settings deep in the diagram it’s a non-issue.
That’s why such things only work when open sourced, so it won’t be deprecated as long as there still is people using it.
My idea was to have the light host a WLAN Network with a minimal web GUI where you can enable/disable modes, and change brightness of each. So it only needs a wifi capable device to set up, and regular single button UI during use.
I know not everyone is willing or able to do this, but i have customized configs i flash to my lights when i get them. This allows me to configure my settings on a computer, compile the firmware, then flash it to the light. since my settings are the new defaults for the firmware, I can factory reset my lights to get back to my defaults.
I feel that. It helped me get familiar with the options and how to navigate the menus, but it gets old when you have about 20 lights with anduril, and you regularly update the firmware for access to new features.
