How much practical diffs in LED output brightness?

[jur]

For my bicycle riding, I have a Fenix L1D-CE on my handlebars, used at 80lm, and on the helmet I have an L1T set to bright (approx. 40lm).

I don't see much difference between the 2. Granted, the handlebar light is viewed at an angle while the helmet light is line-of-sight, but even on my garage door there is not really such a huge difference between the 2.

Sure I can see a difference when switching between 40 and 80lm on the L1D, which is a doubling, but it aint THAT much in practical terms, you see.

So here's the question: Is there going to be a real practical difference between the Cree P4 and Q5 LEDs (80 vs 107lm)? I wonder if it really wil be all that noticible in use, ie not in white ceiling tests.

 

[Oddjob]

Some people get caught up in numbers and others focus on real world use. Back when the PID CE was first released there was all kinds of debate on the dark ring surrounding the hotspot. The "white wall hunters" did not like the beam quality while "real world users" did not notice it. I am somewhere in the middle. I like a nice clean beam but I do not notice mild artifacts when using some of my lights.
The benefit that I have enjoyed with the new wave of emitters is longer runtimes at the same output levels. Don't get me wrong, I like a bright light but sometimes I have to remind myself that I got along just fine with a minimag for years before I discovered CPF.
I always say "to each his own" and people can do whatever they want with their lights. Because we are all different ultimately I think it is an indivdual thing. Once person's "bright enough" is another person's "not bright enough". One person's shelf queen is another person's EDC. IMO the practical difference is whatever the individual deems it is.

 

[OhMyGosh]

For me it is so much about being brighter - but being able to get the same brightness with less current. Longer run time, cooler running, better lumen maintenance (longer life).

 

[shakeylegs]

Comparing my P1Dce (P3) and the P1Dce SEQ2 with OP, I've found clear real world differences. I tend to use these lights outdoors on uneven terrain for route finding and breaking trail, and I find the smooth beam of the Q2 preferable to the ringy beam of the original P1Dce especially where footing is dicey. Potholes, roots and rocks can hide in the dark circles of the ringy beam.
Also, the Q2, even with op reflector, appears to put out more overall light with greater throw and broader spot. These are purely subjective observations but they are real enough that I prefer the Q2 and am looking forward to transplanting a Q5 into my original P1Dce along with one of David's OP reflectors.

 

[Sub_Umbra]

I try not to put much emphasis on whether or not a light seems much brighter. I know that if I increase the output by 10% I'll be able to see more with it whether I may conciously notice the difference in brightness or not.

 

[lowatts]

Sure I can see a difference when switching between 40 and 80lm on the L1D, which is a doubling, but it aint THAT much in practical terms, you see.

So here's the question: Is there going to be a real practical difference between the Cree P4 and Q5 LEDs (80 vs 107lm)? I wonder if it really wil be all that noticible in use, ie not in white ceiling tests.

Going from 40 to 80 lumens won't be perceived by the eye as a doubling in illumination. Not sure how it works with light, but with sound the human ear needs 10 times the power to perceive a doubling in loudness.

In the case of going from 80 lm to 107 lm between bins at the same power, that's about a 27% increase, that means it's possible to increase run time by roughly that much at the same brightness level. IIR the Cree is about 40-50% more efficient than the Luxeon, and that was a big deal, the difference was really noticeable; it looks like we're quickly getting to that % improvement again over the early Crees.

 

[Gunner12]

According to Flashlight reviews, the L1T produces 26 lumen on high and the L1D-CE produces 61 lumen on high and turbo(no difference unless you use a Lithium battery), not the 80 or 37 lumen claimed by Fenix.

Human eyes do not detect brightness linearly. For a perceived double in output actually requires four(?) times the output.

 

[OhMyGosh]

Your eye is funny about perceived brightness. It is very very good at adapting to the average light level and color balance. If you view two lights separately, you will need a large change in brightness (and color) to notice much difference. But if you compare them side by side at the same time you can detect very slight differences in color and brightness.

 

[2xTrinity]

Going from 40 to 80 lumens won't be perceived by the eye as a doubling in illumination. Not sure how it works with light, but with sound the human ear needs 10 times the power to perceive a doubling in loudness.

In the case of going from 80 lm to 107 lm between bins at the same power, that's about a 27% increase, that means it's possible to increase run time by roughly that much at the same brightness level. IIR the Cree is about 40-50% more efficient than the Luxeon, and that was a big deal, the difference was really noticeable; it looks like we're quickly getting to that % improvement again over the early Crees.

With the eyes, changes in intensity of incident light, or lux, are easier to detect at very low levels, when the pupils are more dialated. Also, an increase in lumens will genrally be more noticeable if it is used to light a larger area to the same intensity, rather than the same area to a higher intensity.

Also, there are more variables than with hearing. Things like uniformity of light make a differrence as well. Using a light with a very bright hotspot up close may actually hinder visibility by causing the pupils to constrict, while if using a light with the same lumens, but uniformly diffused, then it translate to better visibility. That is why I find I can differentiate between lights with similar output, such as the modes on the Fenix lights, using ceiling bounce, or when lighting up distant targets very easily, but not as well comparing the hotspot brightness up close.

IMO a much more practical difference would be noticeable if the Fenixes, instead of just dropping in a more efficient LED (such as a Q5 as opposed to a P4) actually adjusted the current so that all but the highest level would be the same brightness as before. That would lead to, in general, about a 33% reduction in current at all but the "turbo" level. Battery life might be extended by more than that much, due to the fact that batteries typically sag under heavy load. That is the beauty of variable brightness -- with increased efficiency comes longer battery life at low levels, or a higher max, whichever is needed at the time.

 

[B0wz3r]

With the eyes, changes in intensity of incident light, or lux, are easier to detect at very low levels, when the pupils are more dialated. Also, an increase in lumens will genrally be more noticeable if it is used to light a larger area to the same intensity, rather than the same area to a higher intensity.

Also, there are more variables than with hearing. Things like uniformity of light make a differrence as well. Using a light with a very bright hotspot up close may actually hinder visibility by causing the pupils to constrict, while if using a light with the same lumens, but uniformly diffused, then it translate to better visibility. That is why I find I can differentiate between lights with similar output, such as the modes on the Fenix lights, using ceiling bounce, or when lighting up distant targets very easily, but not as well comparing the hotspot brightness up close.

IMO a much more practical difference would be noticeable if the Fenixes, instead of just dropping in a more efficient LED (such as a Q5 as opposed to a P4) actually adjusted the current so that all but the highest level would be the same brightness as before. That would lead to, in general, about a 33% reduction in current at all but the "turbo" level. Battery life might be extended by more than that much, due to the fact that batteries typically sag under heavy load. That is the beauty of variable brightness -- with increased efficiency comes longer battery life at low levels, or a higher max, whichever is needed at the time.

Actually, pupil contraction has very little to do with it. At a given level of light intensity, and when changes are small as you mention above, the difference will not be enough to cause an alteration in pupil size, let alone invoke Whytte's reflex in the iris.

The reason it is easier to detect changes in weaker stimuli is because the amount of change in stimulus intensity is less than for high levels of stimulus intensity. For example, if I ask you to gauge the weight of a one ounce weight, and then add a one pound weight to your hand and ask you to decide which is heavier, the task is absurdly simple. However, if I give you a 50 lb sack of potatoes, then add a one pound weight to it and ask you the same question, you cannot tell the difference. In both cases, it is the ratio of the amount of change to the previous stimulus intensity that is at work. That's how all the perceptual systems work, by comparing relative amounts, not absolute intensities. Even at low levels of illumination, as long as the change is close to the difference limen, it will not be a large enough difference to effect the contraction state of the iris.

-B0wz3r
(PhD in Perceptual and Experimental Psychology, University of California Santa Cruz, 2005).

 

[drmaxx]

, but even on my garage door there is not really such a huge difference between the 2.

Sorry if I am stating the obvious, but your eye does not see 'lumens' but 'lux' The difference between 40 and 80 lumens is not that big of a difference to the eye and will be additionally distorted by having two different reflectors that disperse the amount of light differently.
Especially for biking the reflector and the amount of useful light provided by the light makes all the difference.