WML shootout

[Unobtanium]

Arisaka and Rosch Works are using Don McLeish designed reflectors and xpg2 pills assembled by malkoff in their SL1 and E1 lights, respectively. Claiming 250 to 280 of lumens or so. I used my m31 drop in in a VME head for this test.

The M600U is a very successful wml featuring Surefire's TIR technology, and the M300c is its new little brother. Both are very popular.

The Streamlight Protac 1L is a 180 otf lumen 2.0oz light preferred by Frank Proctor so strongly that he had a special mount made just for him and that light by Unity.

Below, I compare the above as well as the high and low beams of my 2010 grand jeep cherokee. The target is a red 1 gallon gas jug at 50 meters wedged between the two small tree trunks center frame at waist level, with the larger light colored oak as a backdrop.

Jeep hi beams:

Jeep low beams:

SL PT1L:

Malkoff 280 otf lumen M31 in VME head with AR coated sapphire lens:

Surefire M300c (KE1F head):

Surefire M600U (KE2C head):

[Byron]

Criticism that is meant to be constructive and is hopefully taken that way:

When taking beamshots, a cell phone camera (Galaxy S4 in this case) is a subpar choice. Not only will it have a lot of difficulty with fine detail at low light (note the gas can is just a blob in the best lit photo, and the trees lack any detail), but I am curious how much control you had over the exposure. Does your S4 allow you to manually fix the exposure length, f-stop, and ISO? If those three values were not 100% fixed between shots, the comparison becomes moot.

I would also highly recommend using a tripod. It will not only allow a longer exposure (that will more accurately reflect what the human eye can see), but will ensure that your composition does not change from shot to shot.

When sharing beamshots, it's helpful to share the exposure length, f-stop, and ISO used. Do you recall what they were in this case? I don't see them stamped on the EXIF data.

I would also recommend using a fixed white balance so that the camera does not compensate for beams that are more yellow or blue in tint. This is one of the few things I do see stamped in the EXIF data and it shows "Auto" rather than a fixed setting.

Finally, I'd recommend setting a different benchmark/control than vehicle hi-beams. Those things are putting out thousands of lumens that will make any CR123 light look pitiful by comparison. I'm sure the weapon lights appear much brighter to the human eye (especially if the eyes are somewhat dark-adapted) than those photos.

[JV_]

If those three values were not 100% fixed between shots, the comparison becomes moot.

Agreed.

[Unobtanium]

Criticism that is meant to be constructive and is hopefully taken that way:

When taking beamshots, a cell phone camera (Galaxy S4 in this case) is a subpar choice. Not only will it have a lot of difficulty with fine detail at low light (note the gas can is just a blob in the best lit photo, and the trees lack any detail), but I am curious how much control you had over the exposure. Does your S4 allow you to manually fix the exposure length, f-stop, and ISO? If those three values were not 100% fixed between shots, the comparison becomes moot.
Yes, I could fix all of those. However, after comparing photos taken with fixed values, to what my eyes saw, I chose not to. I agree 100% that this is not the most scientific method. However, the photos I took are representative of what I saw with my naked eye, minus the note you made about resolution, and about a bit of the spill in what I call the "10% zone" (where my eye could have maybe 200/20 vision due to lack of light, but the camera saw black). However, the photo's I've posted are very near to what my own eyes saw. This is subjective instead of objective, but it's the best I care to do (I don't want to go and buy a better potato just for this purpose, I'm sure you understand)

I would also highly recommend using a tripod. It will not only allow a longer exposure (that will more accurately reflect what the human eye can see), but will ensure that your composition does not change from shot to shot.

When sharing beamshots, it's helpful to share the exposure length, f-stop, and ISO used. Do you recall what they were in this case? I don't see them stamped on the EXIF data.
I set them to auto. As above, I selected all those values as fixed and tried to tailor them to what my own eyes were seeing. In the end, "Auto" ended up replicating what I saw, best. Basically, I went out and shot 100 goats in the boiler room while grazing instead of shooting 100 blocks of gello in a lab. Some people will find more value in what I did, some folks won't. I just wanted to represent what my own eyes saw rather than a fixed value, as the fixed value was radically different in the case of my potato.

I would also recommend using a fixed white balance so that the camera does not compensate for beams that are more yellow or blue in tint. This is one of the few things I do see stamped in the EXIF data and it shows "Auto" rather than a fixed setting.
Again, I tried this. It either made the lights VERY warm, or VERY cold. The "Auto" setting on my S4 was pretty darn close (as pictured) to what my eyes saw. For reference, my Jeep's headlights on high looked to me about 4200K, my M600C was around 5500k, my M300C was around 5000K, my M31 was around 6500K, the SL PT1 was around 6000K, give or take, to my eyes. I've spent a ton of time messing with LED's in various tints and know that people see tint differently, so give me +-500 on those numbers, with the most variance on the Surefire, as they had a touch of green in the tint. That is actually what I liked most. They had almost no "gray out". My S4 actually captured "gray out" and lack, VERY WELL! in these photos. Notice how the Malkoff module makes you strain to see anything, even when it's got some light on it? That's how it felt in person. EXACTLY. While the Surefire, if you touched it with light, it was lit. I don't think the SF had so much higher CRI values, as I feel it just allowed more contrast with the introduction of some green/yellow into the tint spectrum.

Finally, I'd recommend setting a different benchmark/control than vehicle hi-beams. Those things are putting out thousands of lumens that will make any CR123 light look pitiful by comparison. I'm sure the weapon lights appear much brighter to the human eye (especially if the eyes are somewhat dark-adapted) than those photos.

They do appear brighter to the human eye, and yes, especially when dark adapted, but the photos were pretty representative, minus the "10% area" I mentioned above. That really gives photos like mine a "tunnel vision" feel where you get an abrupt "cut off" instead of the gentle natural "fade" of human vision.

No hard feelings here, I agree with what you said, but provide my rationale above, given my equipment limitations.

[Unobtanium]

What impressed me most was how far 1 CR123 powered lights have come. Here is a quick comparo at my apartment complex of my M300C and M600U:

This illustrates the utility of each light in a small structure. From the door frame to the far wall in the kitchen is 6 meters. The edge of the door frame the lights were held against (at about neck height, aimed directly across the room as humanly possible) is about 1.2m from the wall seen to the right with the boxes stacked against it. The room is a total of roughly 4m across, with the first door on the left appearing at 3 meters beyond the entry door, and the second door at 5 meters. The second door is offset into the kitchen by roughly 1/2meter.

SL PT1L:


Surefire M300C:


Malkoff M31:

[Unobtanium]

Again, Byron, I know you will comment on the M300C and M600C 100 yard comparison, and how they both appear so similar, and blame it on the auto-adjust of my S4. However, this photo is identical to what my eyes saw. The two lights in "real life" are literally that close to each other. Look at the reflection from the white square (ventilation) on the roof center-frame and compare it. Very subtly different. I really think the auto settings on this phone provide great light-to-light comparison. My HTC Incredible before it did best with all fixed values.