Infinitely Adjustable output vs many modes?

22 posts / 0 new
Last post
gcbryan
Offline
Last seen: 1 year 10 months ago
Joined: 05/07/2010 - 20:42
Posts: 2528
Location: Seattle,WA
Infinitely Adjustable output vs many modes?

Do you prefer the ramping or ring control for infinitely adjustable light output or multiple modes where you know what the output will be?

Initially I thought variable output was a great idea but the more I use it the more I prefer to have a well designed UI with many modes which I know what the actual output will be.

With ramping in particular you never really know if you are wasting battery power. I liked the ramping on my Storm (headlamp) and Proton Pro until I got the Zebralight H51f (headlamp) and had all the choice of output that I could want (6 levels at least) and they were easy to access and I know what level each one is.


I see a lot of interest in the Sunwayman these days. Has anyone had one long enough to start to prefer a well defined but more discrete output like the Zebralight UI or is the ring system the way to go for most of you?

Edited by: gcbryan on 09/17/2011 - 03:40
Tido
Offline
Last seen: 3 days 9 hours ago
Joined: 05/28/2010 - 15:28
Posts: 189
Location: Berlin, Germany

I have played around with the "infinitely" (really just 255 stages) adjustable model early in the development of the BLF-VLD and ditched it in the second release. It either takes ages to get to the high or low end of the range or you rush through the levels so fast that you miss the one you are after.

The eye is also very bad at distinguishing light levels. Level 220 will look no brighter than level 200, so what is the point of having those extra 20 levels? I usually create just nine levels spaced on a logarithmic scale (1, 2, 4, ... , 128, 255).

gcbryan
Offline
Last seen: 1 year 10 months ago
Joined: 05/07/2010 - 20:42
Posts: 2528
Location: Seattle,WA

I've found that to me the best spacing is based on 3's...3 1.3.3,10,30,90,270.

Hikelite
Offline
Last seen: 3 weeks 6 days ago
Joined: 07/13/2011 - 16:18
Posts: 3614
Location: RO

Tido wrote:

The eye is also very bad at distinguishing light levels. Level 220 will look no brighter than level 200

The eye is not very bad.

It's just not too much difference. Compare 220mm vs 200mm, except you don't need something to fit a small box that 20mm will be totally ignorable compared to all other big objects in a house, sofas, tables.

A glass having 2.5mm thickness is almost impossible to differentiate from a 2.89mm glass, because of how it relates to everything around you.  

 

PilotPTK
Offline
Last seen: 4 years 5 months ago
Joined: 09/04/2011 - 21:53
Posts: 1736
Location: Michigan, USA

Hikelite wrote:

Tido wrote:

The eye is also very bad at distinguishing light levels. Level 220 will look no brighter than level 200

The eye is not very bad.

It's just not too much difference. Compare 220mm vs 200mm, except you don't need something to fit a small box that 20mm will be totally ignorable compared to all other big objects in a house, sofas, tables.

A glass having 2.5mm thickness is almost impossible to differentiate from a 2.89mm glass, because of how it relates to everything around you.  

 

 

The eye is quite amazing, actually.  We have a huge dynamic range and can 'see' things from VERY dark to VERY bright.  Our eye, however, is pretty darn bad at distinguishing between small changes because of that range we have available.  If memory serves, brightness perceived by our eye is about log[2.5] for double the brightness.  For flashlight brightness steps, log[2.5] steps will appear very natural.

I am currently extremely busy with work. Please do not expect a response from me quickly. I will be dropping in as time permits, but the amount of time I can dedicate to responding to topics and PMs is very limited.

okwchin
okwchin's picture
Offline
Last seen: 1 year 6 months ago
Joined: 01/07/2011 - 04:41
Posts: 1144
Location: Australia

Measurements are one thing, they have a linear correlation, however our eyes have a logarithmic response to light, so the 200 to 220 lumens will appear much less than the absolute value difference of 10%, totally indistinguishable unless side by side, and even then, not really noticeable.

 

Infinitely variable brightness is useful when you have your hand on the control all the time, and you keep on changing viewing distance but want to maintain a relatively constant target brightness, Interms of practical use, a 6 mode would cover practically everything you need for working, more is nicer

Whats more important is the UI, the UI is what makes or breaks the torch. Having 3 modes, on a reverse clicky with memory causes me grief when starting up with a mode lottery, while an electronic clicky with shortcuts to high and low really makes a torch work.

I like rotary magnetic ring control for the ability to set brightness easily, by feel, and during operation to any level, however its the battery life / brightness that tends to be a little unknown.

The V10R, vs the M10R... The M10R has 3 modes + strobe. Modes are well spaced, make sense, yet I still prefer to have the V10R. Maybe if it had 6 modes, that would give me more output options.

The HDS rotary is FAR from budget, but is an example of a torch with a rear electronic clicky that has 4 modes, each of which has a key press shortcut. Three of the modes are programmable to any brightness, and the last mode is controlled by a tail cap that turns through 350 degrees and can be completely operated with a single hand. It can be programmed to have no mode memory and start-up on the rotary dial. Interms of UI, this is the cumulation of the benefits of a button click, shortcuts based 4 mode torch, with the addition of a rotary interface with 20 brightness steps, and as far as I've seen around, pretty much the closest to the perfect UI.

"like everyone else - I’m looking for my next “last” flashlight" -  ohnonothimagain

ainu
Offline
Last seen: 6 years 4 months ago
Joined: 08/10/2011 - 05:20
Posts: 32
Location: Europe

I think there is a waste of energy with fixed mode, because I always use the mode That ensures a light equal or more to my needs.
With variable instead I can choose with more precision, and consequently less waste.

mitro
Offline
Last seen: 1 year 6 months ago
Joined: 01/25/2011 - 11:58
Posts: 707
Location: Montgomery, IL USA

I like easily repeatable light levels. While infinitely adjustable is nice, I much prefer many pre-set levels because I generally know how long the cells will last at a given output.

Anyway, I think the Zebralight UI/level spacing is just about perfect. (Shocking, right?) But if anything I'd have it combined with the UI of a HDS clicky where you could pick your 6 modes from 20-30 levels.

Boaz
Boaz's picture
Offline
Last seen: 1 hour 19 min ago
Joined: 11/07/2010 - 09:31
Posts: 6537
Location: Birthplace of Aviation

okwchin wrote:

Measurements are one thing, they have a linear correlation, however our eyes have a logarithmic response to light, so the 200 to 220 lumens will appear much less than the absolute value difference of 10%, totally indistinguishable unless side by side, and even then, not really noticeable.

 

Infinitely variable brightness is useful when you have your hand on the control all the time, and you keep on changing viewing distance but want to maintain a relatively constant target brightness, Interms of practical use, a 6 mode would cover practically everything you need for working, more is nicer

Whats more important is the UI, the UI is what makes or breaks the torch. Having 3 modes, on a reverse clicky with memory causes me grief when starting up with a mode lottery, while an electronic clicky with shortcuts to high and low really makes a torch work.

I like rotary magnetic ring control for the ability to set brightness easily, by feel, and during operation to any level, however its the battery life / brightness that tends to be a little unknown.

The V10R, vs the M10R... The M10R has 3 modes + strobe. Modes are well spaced, make sense, yet I still prefer to have the V10R. Maybe if it had 6 modes, that would give me more output options.

The HDS rotary is FAR from budget, but is an example of a torch with a rear electronic clicky that has 4 modes, each of which has a key press shortcut. Three of the modes are programmable to any brightness, and the last mode is controlled by a tail cap that turns through 350 degrees and can be completely operated with a single hand. It can be programmed to have no mode memory and start-up on the rotary dial. Interms of UI, this is the cumulation of the benefits of a button click, shortcuts based 4 mode torch, with the addition of a rotary interface with 20 brightness steps, and as far as I've seen around, pretty much the closest to the perfect UI.

the beauty of the hds is it comes in one stupid style that demands that  you either love it or hate it ..Personally i hate it ...certainly don't love it enough to offer up my left tes**** for it . If perfection lies in only one super model than I'll take the bus .it's a light ..not an experiance in photon display

 I find the varible brightness rings to be an awkward move with one hand and a chicken neck wringing with two .. either way makes light jump all over the place while raising or lowering light levels /It's impossible to flick your thumb and hold a steady beam of light . The zebralights micro clicks just seem friendlier and who on this planet doesn't really love a side clicky ?

HDS problem // ugly ..1 design 1 shape /ugly

 

mitro wrote:

I like easily repeatable light levels. While infinitely adjustable is nice, I much prefer many pre-set levels because I generally know how long the cells will last at a given output.

Anyway, I think the Zebralight UI/level spacing is just about perfect. (Shocking, right?) But if anything I'd have it combined with the UI of a HDS clicky where you could pick your 6 modes from 20-30 levels.

  I agree with you the zebralight ui is very nice .....

 I'm of the opinion that if you were given 20 different ramping lights  and asked to pick a medium in varying circumstances ..  you would have always picked almost the exact same lumen level ....and that the variance would be so slicght you'd have a hard time picking out which one was avtually brighter or dimmer out of the 20 lights ...

.Now to prove my theory ...I need sunwayman,jetbeam or zebralight to give me 20 lights to test Silly

καὶ τὸ φῶς ἐν τῇ σκοτίᾳ φαίνει καὶ ἡ σκοτία αὐτὸ οὐ κατέλαβεν

Boaz
Boaz's picture
Offline
Last seen: 1 hour 19 min ago
Joined: 11/07/2010 - 09:31
Posts: 6537
Location: Birthplace of Aviation

gcbryan wrote:

I've found that to me the best spacing is based on 3's...3 1.3.3,10,30,90,270.

 

this ONLY works on the zebralight because it's a 3 mode light

6 regular modes would be insane

4 would be pushing it

a three mode light works really well for 90% of the time

you need a moon mode  / a low / a med/a high for runtime  /and then a 100% output

the zebralight gives you 6 modes but the medium  mode twice is just redundant ..they had a space there and had to use it Smile

 my optimal spacing is 2 /30/180  with hidden 0.3 moon and a turbo mode of 350   ..5 speed no 2nd medium nessesary

καὶ τὸ φῶς ἐν τῇ σκοτίᾳ φαίνει καὶ ἡ σκοτία αὐτὸ οὐ κατέλαβεν

SashiX
SashiX's picture
Offline
Last seen: 4 years 3 months ago
Joined: 10/07/2010 - 15:46
Posts: 2045
Location: Spain

Well, 3-4 modes well spaced may replace (for me at least) adjustable brightness. Memory is a must here, of course.

Eltehs
Offline
Last seen: 6 years 5 months ago
Joined: 08/03/2011 - 02:49
Posts: 9
Location: Latvia (EU)

gcbryan wrote:

Do you prefer the ramping or ring control for infinitely adjustable light output or multiple modes where you know what the output will be?

What did you think about Maglite 200 UI - 3-4 programmable modes, changed with reverse clicky ? Mode programming by moving light with tailcap pressed.

Ways
Offline
Last seen: 12 months 4 days ago
Joined: 05/10/2010 - 15:30
Posts: 40
Location: UK

I prefer variable brightness to modes generally speaking and further prefer control ring to ramp but it depends on how it is implemented, for example the ramping on the D11v2 is good but it is the ramping along with fast access to min/max that makes this light for me.

Sunwayman v10r and its control ring i really like and although it does not have shortcuts to min/max the short travel of the ring makes that a moot point for me. 

The daddy of all variable UI's for me is the Varapower 2000 it is genius, simple click on/off and twist the switch for level although the light itself is too large for EDC if someone made a similar UI in an 18650 host with an XM-L it would be a winner for sure.

The downside of not knowing the battery drain at any given level is irrelevant for my purpose i will use the lowest level i need to accomplish the task at hand i do however check the battery drain at lowest & highest setting and that gives me the range for max / min runtime and i consider that to be enough. 

On the other hand i really like the AKoray K-106 with programmable modes. It's not a light i use very often but the fact that it has modes that i have set to be as close to my ideal as i can get makes it something i keep coming back to.

 

Bottom line... variable all the way, be it twisty / turny ring or ramp but with ring for preference.

   

Tido
Offline
Last seen: 3 days 9 hours ago
Joined: 05/28/2010 - 15:28
Posts: 189
Location: Berlin, Germany

I have picked up this project again today. It is basically the 18650 version of this torch, where I replaced the puny XR-E Q5 with a neutral white XM-L T5 driven at 2.8A. I also wrote a basic driver that is using the magnetic selector ring as a simple up/down switch to cycle through the modes. That was the point where I got stuck, as I absolutely suck at designing UIs. Maybe some of you can come up with something decent.

This thing got two inputs. First the selector ring, which, for technical reasons, can only be used to detect movements clockwise or counter-clockwise.  Second is the reverse clicky switch at the tail. It can be used at start-up to determine how long the light had been powered on before it was last turned off. This is what most cheap driver PCBs do.

I'd like to go with the standard BLF-VLD approach of having two mode groups, one main group and an extended group. The extended group contains all modes available, around ten different constant light level, strobes, beacons, distress signals, etc. The main group consists of 3-5 modes, which can be chosen from the extended group freely and in any order. The hard part is how to change from the main group into the extended group and how to program a chosen mode into the main group. I was thinking about as special turning sequence (e.g. up-down-up-down-up-up-down) but this seems a bit awkward. Another way could be clicking sequences, like three short taps at the tail switch in a row.

Well, any suggestions are welcome.

gcbryan
Offline
Last seen: 1 year 10 months ago
Joined: 05/07/2010 - 20:42
Posts: 2528
Location: Seattle,WA

Eltehs wrote:

gcbryan wrote:

Do you prefer the ramping or ring control for infinitely adjustable light output or multiple modes where you know what the output will be?

What did you think about Maglite 200 UI - 3-4 programmable modes, changed with reverse clicky ? Mode programming by moving light with tailcap pressed.

I've never seen that light. I've had the Akoray K-109 where there were 3 programmable slots and 5 choices. That's OK if you are only going to have 3 modes (which isn't bad).

Zebralight is perfect (to me) in that you have 3 modes and a double click in each mode makes that 6 modes. It's not cumbersome and the outputs are well spaced as mentioned each is approximately 3 times the last output.

old4570
Offline
Last seen: 3 hours 14 min ago
Joined: 09/06/2010 - 02:44
Posts: 3322
Location: Land of Oz

Ive got 2 variable output lights , and they are very nice , if only they were XM-L , they might see more use ... 

As for ease of use , very easy to use . 

One of the variable drop ins is in my Surefire P6 , with the twisty tailcap , and still easy to use . 

 Always remember , the easiest thing in the world to do , is to expel hot air from your lungs and through some vocal chords ..
The resulting sound may , or may not be worth listening too ….

 

df2dot
Offline
Last seen: 9 months 1 week ago
Joined: 02/15/2011 - 14:15
Posts: 520

 

problem is the cost, fancy UI gets a bit pricey. if i am paying $30+ it better be perfect.

I think ring control for brightness + strobes if you like. 2 separate switches to boot

okwchin
okwchin's picture
Offline
Last seen: 1 year 6 months ago
Joined: 01/07/2011 - 04:41
Posts: 1144
Location: Australia

Tido wrote:

I have picked up this project again today. It is basically the 18650 version of this torch, where I replaced the puny XR-E Q5 with a neutral white XM-L T5 driven at 2.8A. I also wrote a basic driver that is using the magnetic selector ring as a simple up/down switch to cycle through the modes. That was the point where I got stuck, as I absolutely suck at designing UIs. Maybe some of you can come up with something decent.

This thing got two inputs. First the selector ring, which, for technical reasons, can only be used to detect movements clockwise or counter-clockwise.  Second is the reverse clicky switch at the tail. It can be used at start-up to determine how long the light had been powered on before it was last turned off. This is what most cheap driver PCBs do.

I'd like to go with the standard BLF-VLD approach of having two mode groups, one main group and an extended group. The extended group contains all modes available, around ten different constant light level, strobes, beacons, distress signals, etc. The main group consists of 3-5 modes, which can be chosen from the extended group freely and in any order. The hard part is how to change from the main group into the extended group and how to program a chosen mode into the main group. I was thinking about as special turning sequence (e.g. up-down-up-down-up-up-down) but this seems a bit awkward. Another way could be clicking sequences, like three short taps at the tail switch in a row.

Well, any suggestions are welcome.

 

Detect turn movement only? so are you able to detect position relative to a centre, or just relative change. If its just relative change, theres great potential for the UI to get 'lost' interms of position.

For one which is relative to centre, (i.e. spring loaded to middle position, and you turn it clockwise or anticlockwise momentarily), I have a torch with such a UI, and it works pretty well! Strange, and has its quirks, but it works.

"like everyone else - I’m looking for my next “last” flashlight" -  ohnonothimagain

Tido
Offline
Last seen: 3 days 9 hours ago
Joined: 05/28/2010 - 15:28
Posts: 189
Location: Berlin, Germany

okwchin wrote:

Detect turn movement only? so are you able to detect position relative to a centre, or just relative change. If its just relative change, theres great potential for the UI to get 'lost' interms of position.

For one which is relative to centre, (i.e. spring loaded to middle position, and you turn it clockwise or anticlockwise momentarily), I have a torch with such a UI, and it works pretty well! Strange, and has its quirks, but it works.

No momentary or spring loaded switches. The ring contains a magnet and its position is picked up by Hall-effect sensors. So normally this design could pick up the absolute position of the ring. But of course there is a twist...

The ring has six positions but only four of those are equipped with sensors. Let's say the positions are numbered 0 to 5 counter-clockwise. There are sensors at positions 1, 2, 4, 5. If the ring is in positions 0 and 3, the driver can't distinguish between those two. Therefore it can not be used to read the absolute position of the ring. But it is possible to deduct the direction of the ring being turned. If, for example, the magnet was detected at position 1, and next it is detected at position 2, the ring has been turned counter-clockwise. If it just "vanishes", the ring must have been moved clockwise into position 0.

okwchin
okwchin's picture
Offline
Last seen: 1 year 6 months ago
Joined: 01/07/2011 - 04:41
Posts: 1144
Location: Australia

Boaz wrote:
 I find the varible brightness rings to be an awkward move with one hand and a chicken neck wringing with two .. either way makes light jump all over the place while raising or lowering light levels /It's impossible to flick your thumb and hold a steady beam of light .

Ive never had an issue with magnetic ring torches flying out of my hands, infact my V10R is a great single handed torch, with brightness easily adjustable with just my thumb. Not as great for tactical grips, but thats where the HDS works well. Not the best looking thing, but not butt ugly either. The beauty is within!

"like everyone else - I’m looking for my next “last” flashlight" -  ohnonothimagain

okwchin
okwchin's picture
Offline
Last seen: 1 year 6 months ago
Joined: 01/07/2011 - 04:41
Posts: 1144
Location: Australia

Tido wrote:

The ring has six positions but only four of those are equipped with sensors. Let's say the positions are numbered 0 to 5 counter-clockwise. There are sensors at positions 1, 2, 4, 5. If the ring is in positions 0 and 3, the driver can't distinguish between those two. Therefore it can not be used to read the absolute position of the ring. But it is possible to deduct the direction of the ring being turned. If, for example, the magnet was detected at position 1, and next it is detected at position 2, the ring has been turned counter-clockwise. If it just "vanishes", the ring must have been moved clockwise into position 0.

Ah ha! needs more thinking than just a 6 pos switch...

Doesn't make it any easier, especially with 0 being uncensored. Will sensors 2 and 4 show some small change with the switch in position 3? that could help determine pos 3 vs pos 0, and therefore allow you to setup 0 as an off/minimum position.

"like everyone else - I’m looking for my next “last” flashlight" -  ohnonothimagain

Tido
Offline
Last seen: 3 days 9 hours ago
Joined: 05/28/2010 - 15:28
Posts: 189
Location: Berlin, Germany

The inputs are latched,  they can only be on or off. Besides, absolute positions would be a bit useless anyway. There are no markings on the ring or the body, so it's impossible to see which position it is in.

The up/down  works quite well and I think it is just about perfect for moving within a mode group. What I haven't quite figured out is how best to switch between mode groups and lock in new modes into the main group. Using three or four short taps on the tail switch might be best.