The perfect UI; the perfect torch?

Jaz

Newly Enlightened
Joined
May 4, 2007
Messages
29
There are countless multi-mode EDC type LED torches on the market. So you'd think that by now some manufacturer would have come up with one that could be called "perfect".
Obviously what's perfect depends on the user's preferences, but reading specs, UI setups and reviews on CPF I certainly haven't found one that comes across as perfect. Rather each one is just OK, or good, but has one or two or three things which stuff it up.
I have to admit I haven't used much of a range of LED torches myself – I hardly qualify as a "flashaholic", with a collection of a grand total of 2 torches (Fenix L2D, Jetbeam C-LE).
But here's what I would consider the perfect 1AA torch. I've tried to keep it completely realistic. If some manufacturer happens to stumble across this design and manage to make something like it, then good for them (and us).

Firstly, the LED: a Cree R4, when it comes out, to give 140 lumens from a AA on turbo with about an hour of runtime. (About what the specs say we should get.)
Now for the tricky part: the UI. Most will agree that 15 or 25 or 250 modes are too many, but really I think there's nothing wrong with having a full range of brightness levels. The problem is arranging them well so the options are there but don't get in the way when they're not wanted or needed. (I'll get to how this will be managed below.)
So how about 7 lumens, 25 lumens, 50 lumens, 90 lumens, 140 lumens - or something similar.
And those annoying strobe and SOS modes, since manufacturers can't seem to get rid of them. (Don't worry, they'll be out of the way.)

Some aspects of my idea for arranging these modes have, as far as I know, not been used yet, probably because they involve a combination of twisty and clicky switches. Therefore I'll have to describe the design of my imaginary torch to get this idea across…

The head would screw into the body (as with the Jetbeam C-LE), rather than vise-versa as in Fenix torches. The threads would be conductive as in Fenix torches, and the positive contact region of the head would also incorporate a metal ring which when coming into contact with the body would turn the torch into the 140lm turbo mode (as in a Fenix L2D-CE). Because the head screws into the body, though, the way this is achieved would be slightly different:

http://www.jas00zer0.110mb.com/torch.html

Sorry I can't embed the picture in my post, and apologies for the deplorable quality of the diagram. I hope it gets the idea across.
The red part on the head is a cross-section of the ring that would come into contact with the narrowed battery tube (conducting from the negative end of the battery). The part of the raised ring which electricity is conducted through is the sloping outer part. The narrow flat face (blue) is covered with a plastic coating to prevent reverse polarity.
It's not shown in the diagram, but the tail end of the torch would contain a regular clicky.


This is how the modes would be accessed (I tried to draw a chart, but it got too complicated):
The torch is turned on with a full click. If the head is tightened beforehand, or afterwards, it goes into the 140lm constant on mode. At this stage this is the only mode in the turbo setting: half-pressing the clicky switch will simply turn the torch off until the switch is released again. It will come back on in 140lm constant mode. The torch can just be used as a single-mode, full brightness torch.

If the head is loose when the torch is turned on, or when it is loosened from the tightened on position, the first mode it goes to is 7lm. A half-press then takes you through 25 lumens, 50 lumens, 90 lumens, and back to seven – similar to a Fenix, without the SOS.

Strobe and then SOS mode are accessed by tightening the head to turbo, and then doing a quick ½ second or so loosen-tighten. A half-press from either strobe or SOS takes you back to constant 140lm.

From 7lm, 25lm, 50lm or 90lm a quick tighten-loosen (not loosen-tighten as above) does two things:
1) It sets this mode in the memory. When the torch is turned back on it will come on in this mode.
2) It locks this mode in, making the loosened position a single-mode setting like the turbo one. The light can be flashed on and off but the mode won't change.

This locked-in mode is exited with another quick tighten-loosen.

And that's all. It might seem complicated, but really when you think about it this design enables the use of the torch as a simple single- or double-mode one, while still having 5 brightness levels and the strobe and SOS tucked away there. I think that's a close to perfect as we'll get.

Now, if a manufacturer were to make this torch as I've described it, or a similar one, with the exterior look of Jetbeam and the circuit efficiency of Fenix, and sell it for about US$60, I honestly believe their sales would go through the roof.

Reckon it'll happen - sometime?
 
Last edited:
Wow Jax, you've put a lot of thought and effort into that. Unfortunately I wouldn't buy a torch with that complicated a UI :)

My perfect torch, after thinking about your post, would be:

Small 1xAA SSC HAIII torch with a forward clickie (for momentary or full) and a multistage bezel (say 20 steps) going from: dim locator flash - 0.1lm to 180lm (logarithmically) - 180 lm strobe. No need for multimode, soft presses or memory etc, the torch just comes on at the bezel setting you last left it.

Fenix, build it and I'll buy a couple of dozen!
 
more than three light levels is overkill

... so a FLUPIC 2.2 driven light, where the user can set the brightness, is plain perfect
 
I have a fenix P2D Q5 and a fenix L1D R100 as EDC lights. the P2D is always set to "turbo" while the L1D is always set to "low". I like the idea having a lamp which gives me instantly maximum brightness and contrary another which will give me just a bit of light and simply is away only one or two clicks if I need and want a lot more. for me the UI of those fenix's are perfect and I wouldn't want it other as they are, though I haven't made any use for strobe and SOS yet.

I expect to arrive my DBS V2 in a few days and this is the flipside of the coin: I choose the model with the SM-1 switch because it is supposed to be a superthrower and thats it, pronto. I won't need nor wan't more than a simple on/off switch with that one, just on click and the thing has to throw as bright and as far as it can, additional modes would I personally consider just gimmicks within that light. but since this lamp will obviously not be an EDC, I'm just fine with that, its always a matter what you need and want a specific tool for.


regards, holger
 
Wow Jax, you've put a lot of thought and effort into that. Unfortunately I wouldn't buy a torch with that complicated a UI :)

My perfect torch, after thinking about your post, would be:

Small 1xAA SSC HAIII torch with a forward clickie (for momentary or full) and a multistage bezel (say 20 steps) going from: dim locator flash - 0.1lm to 180lm (logarithmically) - 180 lm strobe. No need for multimode, soft presses or memory etc, the torch just comes on at the bezel setting you last left it.

Fenix, build it and I'll buy a couple of dozen!


Well, I'd buy it but I'll be snobbish here and ask for it to built in the US.


Karl
 
Well, for me the perfect UI would be a combination of the Surefire Titan (most intuitive brightness regulation) and an forward tail clicky (if you want to just flash or signal).

To bad that
a) surefires patents wont make such a thing affordable the next 19 years
b) the added lenghts of those features would make a light for to bulky (the titans potentiometer almost doubles the lights lenghts, and a good tail-clicky is also 1-2cm).
 
more than three light levels is overkill

... so a FLUPIC 2.2 driven light, where the user can set the brightness, is plain perfect
Would have to disagree as I find the (4) levels of the NovaTac 120P to be perfect!
There's not a day that goes by that 1 of the 4 levels are not being utilized.
 
I find the (4) levels of the NovaTac 120P to be perfect! There's not a day that goes by that 1 of the 4 levels are not being utilized.

+1 on the 120P. Fast & easy to click on any of the four levels the user selects ... and just as easy to change any or all of those four levels as needed.

The UI on the Surefire U2 is also fast & easy, but limited to six preselected levels -- and a few years ago, six levels would have represented a miracle in design & execution. Now they up the ante with the UA2 and UB2, which build on the simple UI and add more lumens & variable focus.

more than three light levels is overkill
Unless you want or need the ability to adjust the output to the job.
 
ahh you obvious haven't used the dial on a Spy yet.

Twist the detent knob to access 6 visually spaced levels, and if you really have to you can reprogram each level and get strobes and such.

5z5d05e.jpg
 
Wow Jax, you've put a lot of thought and effort into that. Unfortunately I wouldn't buy a torch with that complicated a UI :)

:oops:
It's not really that complicated though - single mode or double, with an optional sequence of 4 brightness modes and the SOS and strobe out of the way...


Actually, I hadn't thought of the dial-type brightness control. That would be ideal. Is this in use in any single AA torches that anyone knows of?

(Trying to keep this immaginary torch completely realistic, and the costs lowish - sub-Fenix if possible.)
 
My favorite is still the VB16 line. Two position rocker on body to increase/decrease output and a push button on the end to turn off/on at whatever level it is currently on. No instructions to remember.
 
The problem with coming up with the perfect UI/perfect torch is that what is perfect for me is not for you.
 
@Jaz, this is a definite improvement over current popular click-click-click-click-click style of UI. However, it seems far more complex than is necessary.

The problem with most multi-level UIs is that they dictate a specific intended usage pattern for each light. High-first makes lights useless for tasks where minimizing glare is important; low-first makes the light useless for prompt attention getting or 'tactical' use for persons so inclined; and finally all multi-click interfaces make lights useless for signaling. There can be no ideal UI for all people and all uses as long as that UI has tradeoffs which get in the way of using the light for certain classes of tasks.

The way around all of these problems of standard UIs is a UI where the light level is controlled by a different mechanism than is used to switch the light. The brightness level also needs to be adjustable while the light is off so as not to fall into the 'high first' or 'low first' dilemma. If the end user is in a situation where low required, then he/she can set the light to low before turning it on. If the user needs high first, then he/she can set the light to high before turning it on.

All that is necessary for this kind of UI is an independent brightness selector, like that used on the Gladius and SF U2. These UIs have the advantage of being dead simple: one detent on the selector matches one brightness level. All the switch does is turn the light on or off.

What your design does is overload a two-stage brightness selector to support more than two brightness levels. While this is much better than the fifty presses model used by far too many manufacturers, it adds far too much UI complexity to what should be a simple system. It's a much better idea to support multiple levels simply by adding more detents to the brightness selector than to resort to a hybrid selector + click click click interface.

I think a better tack towards bringing UIs that don't suck back to medium tier lights is to find some way to make something inspired by the U2/Kroma/Gladius switching ring system affordable to Fenix-level manufacturers. The ideal UI already exists--the only problem is making it cheap enough to be widely used.
 
... It's a much better idea to support multiple levels simply by adding more detents to the brightness selector than to resort to a hybrid selector + click click click interface.

I think a better tack towards bringing UIs that don't suck back to medium tier lights is to find some way to make something inspired by the U2/Kroma/Gladius switching ring system affordable to Fenix-level manufacturers. The ideal UI already exists--the only problem is making it cheap enough to be widely used.

Yep, exactly.
Although I imagine the mechanism also takes up a fair amount of space, so a AA torch with one might not be very EDC-able (have to admit I haven't actually used one).
 
My perfect UI is physical 2 mode switch. Programable for each mode. All you have to do, losen bezel and off-on for cycle thru modes, tighten bezel in desired mode to lock

The NDI is about as close as ive found in my price range.

With the New LS coming out, maybe it will be fully programable, meaning even modes etc can be changed..:party:

Crenshaw
 
Jaz you clearly have a lot to learn.
Have a look at the McGizmo area of the forum there are lots of flashlight design drawings and images of prototypes.
Your idea of using a AA battery is noble but futile.

If you are really interested in a light with a UI as close to perfect with a AA battery have a look at the PHOTON proton.
A single button achieved so very much. And it is small and built and you can but=y one today.
That said there are drivers like the above mentioned flupic which can be programmed to do whatever you want them to do.
Lights like the Novatac which allow the end user to set the levels he wants.
Keep at it. At the moment the Photon proton may be your best bet.

If you are hell bent to make a flashlight of your own to sell look at your numbers carefully. multiply by a factor of 5 and then make sure you are prepared for delays from sub contractors design failures several version of prototypes.
The cost of developing a light which will sell and be competitive is a lot more than you think it is.
Use the prior experience, brains,input of those here on the cpf. There are some of the most amazing flashlight minds in the world here.
And to be frank you really need to improve your ability to make a drawing and be able to post it here. It is not that tough. Using a flow chart to describe your idea of a UI also isn't that tough. If you really want to get your ideas across I suggest that you learn how to do it. There are way too many people with the skills to communicate designs, user interfaces, optical designs, PIC programming, driver design. You will find it very tough to achieve greatness with on off linked bar room napkin drawing and little else which is original.
Am I being harsh?
Sorry if you think so but the seas aren't always smooth.
Really if you are serious go back to the drawing board and do it right. make your ideas clear and concise. Have your hard hat on.
Oh and what ever you do no matter what Do not endeavor to create a flashlight on a presale basis. Several have gone down this path and have never returned. It is tempting...Just don't.
The CPF is an extremely supportive community they are also forgiving of the first mistake or two..But not much more than that.
You will find fast supporters here some ready to encourage..and some who will run at the first sign of trouble. be careful. Most importantly HAVE FUN!!
BTW Fenix has sold more lights than they ever expected to at the beginning.
At first their quality was questionable. Their designs were off the shelf. (most still are) and their costs are low. and still are.
Don't feel sorry for them they are making money hand over fist.

Also for a ring switch UI look at the surefire U2. It uses a hall effect sensor. They have New lights with an even more advanced version of this UI coming out soon.


Yaesumofo
 
Jaz you clearly have a lot to learn.
Have a look at the McGizmo area of the forum there are lots of flashlight design drawings and images of prototypes.
Your idea of using a AA battery is noble but futile.

If you are really interested in a light with a UI as close to perfect with a AA battery have a look at the PHOTON proton.
A single button achieved so very much. And it is small and built and you can but=y one today.
That said there are drivers like the above mentioned flupic which can be programmed to do whatever you want them to do.
Lights like the Novatac which allow the end user to set the levels he wants.
Keep at it. At the moment the Photon proton may be your best bet.

If you are hell bent to make a flashlight of your own to sell look at your numbers carefully. multiply by a factor of 5 and then make sure you are prepared for delays from sub contractors design failures several version of prototypes.
The cost of developing a light which will sell and be competitive is a lot more than you think it is.
Use the prior experience, brains,input of those here on the cpf. There are some of the most amazing flashlight minds in the world here.
And to be frank you really need to improve your ability to make a drawing and be able to post it here. It is not that tough. Using a flow chart to describe your idea of a UI also isn't that tough. If you really want to get your ideas across I suggest that you learn how to do it. There are way too many people with the skills to communicate designs, user interfaces, optical designs, PIC programming, driver design. You will find it very tough to achieve greatness with on off linked bar room napkin drawing and little else which is original.
Am I being harsh?
Sorry if you think so but the seas aren't always smooth.
Really if you are serious go back to the drawing board and do it right. make your ideas clear and concise. Have your hard hat on.
Oh and what ever you do no matter what Do not endeavor to create a flashlight on a presale basis. Several have gone down this path and have never returned. It is tempting...Just don't.
The CPF is an extremely supportive community they are also forgiving of the first mistake or two..But not much more than that.
You will find fast supporters here some ready to encourage..and some who will run at the first sign of trouble. be careful. Most importantly HAVE FUN!!
BTW Fenix has sold more lights than they ever expected to at the beginning.
At first their quality was questionable. Their designs were off the shelf. (most still are) and their costs are low. and still are.
Don't feel sorry for them they are making money hand over fist.

Also for a ring switch UI look at the surefire U2. It uses a hall effect sensor. They have New lights with an even more advanced version of this UI coming out soon.


Yaesumofo

Yeah, I'm sure I do have tons to learn. :duh2: I'll have a look at the McGizmo part of the forum.
I definately wasn't thinking of manufacturing/producing my own flashlight. (Teenagers can't really do that sort of thing. :rolleyes:)
Just posting an idea, nothing more - certainly not looking for "greatness".

Jaz.
 
I can't thnk of any good reson a teenager can't do exactly that with the correct ideas tools and knowledge behind him. Never sell yourself short.
Yaesumofo


Yeah, I'm sure I do have tons to learn. :duh2: I'll have a look at the McGizmo part of the forum.
I definately wasn't thinking of manufacturing/producing my own flashlight. (Teenagers can't really do that sort of thing. :rolleyes:)
Just posting an idea, nothing more - certainly not looking for "greatness".

Jaz.
 
Top