Armytek Wizard Pro vs Zebralight H600fc MKIV Measurements

Directly onto the spectrometer.

Do you have any idea why the SC64c with an XP-L2 emitter that is supposedly less efficient than an XHP50.2 can manage 106 lm/w at average 400 lumens whereas the H600Fc MK4 only manages 89 lm/w at 400 lumens? Is it because the XP-L2 is 3V and the 50.2 is 6V? If so, maybe ZL’s 3V driver is efficient but ZL’s boost driver efficiency is behind the competition. I wish ZL would send you every single model to test so we know what we are buying. Driver efficiency is an important factor to me at least.

I measured 1312 lumens for the H600fc MK4 at 30s compared to your 1124 lumens at 30s. Strange because most of the time my testing setup measures lower than yours for the same light. I’m wondering if ZL found a higher bin or is it just unit to unit variation. That’s quite a huge variation though considering all the identical Olights I measured have near identical output.

A bit above 600 lumens according to the graph.

Efficiency is lm/w. Doesn’t matter how long it runs if the output is super low. Even my Armytek Tiara Pro 18350 w/ the E21A 2000K 9080, which is the most inefficient of the E21A series, has been on 24hrs/day for the past months on the lowest mode and still going. This doesn’t prove the driver is efficient just that it has a great firefly mode.

How are you calculating 135 L / W at that setting of just over 600 lumens?

In my opinion, SH H04RC (don’t confuse with discounted H04) headband and bracket is much better than AT. The quick release bracket allows the pocket clip installed on headband and it’s very sturdy. also the anti slip strip is extremely useful in retaining your headband adjustment. But on the other hand, if you buy discounted H03/H04, you get the mushy silicone rubber headband which almost always requires the top strap to be used. Otherwise the light will bounce even when walking.

[Clemence]

I haven’t seen the new H04 headband yet so I am looking forward to checking it out when my E21a H04s come. That is, of course assuming China ever gets back online.

Okay, but Lumens per watt at different output levels seems to be cherry picking. I feel like the correct way to test something like that would be to do it across all output levels and average the results.

No, IMO the correct way to measure it is by removing the driver and adjust the output so both drivers outputs the same. Every driver design has their own sweet spot.

[Clemence]

Problem is the drivers between flashlights don’t have equal output modes to compare so we can only only rely on whatever graphs and efficiency measurements reviewers provide. Also we have VERY LIMITED flashlight efficiency data to use. Besides Maukka, I don’t think anyone else does efficiency measurements. Measuring efficiency is a huge undertaking as it requires a lot of time to run the battery down. I am grateful with whatever data we get from reviewers here. Therefore we can only get a rough comparison to get a general idea.

Maukka’s graph includes several lights overlaid over each other. The other flashlights were all running at higher output than the H600fc, so the comparison makes sense. Flashlights are less efficient on their higher brightness modes. Again these are not exact differences in efficiency. Even measuring two lights of the same model at the same output mode will result in different efficiency results. Just look at my output measurements for the two Wizard Nichia units or my H600fc measurements vs Makka’s H600fc measurements (but that is a rather extreme case).

The information we have on flashlight efficiency is indeed very limited, and I understand that we can only rely on any testing that has been done so far by generous forum members. This does not mean, on the other hand, that we can draw conclusions on driver efficiency if the limited data available is not comparable.

The H600fc MK4 effiency of 89L / W was tested at 410 lumens. The graph showing other lights only shows output and runtime of different lights with different emitters and different CRI levels. Where are you calculating or getting the Lumen/W at those higher levels? I still don’t see how you can conclude that a comparison can be drawn between different lights and where the lumens/ watt data is coming from. As per my previous question, how are you calculating 135 L / W at that setting of just over 600 lumens for the Olight?

To make matters more complicated, each driver will have optimal efficiency at a specific output. So Driver A may be more efficient at 3 amps, and Driver B might be more efficient at 1 amp. So driver A may be more efficient than driver B in one setting, and the opposite is true at another setting. As far as I am concerned, none of the data above dictates that one driver is more efficient than another or vice versa.

135 lm/w was calculated by Maukka. I dont know how it was calculated

Do you have the link or reference? I’d like to see at what setting this was done on as I am curious myself :slight_smile:

Thanks

It’s in the link in my post above.

OK, so I did a little digging and this is what I see.

The Olight H2R tested 135 L / W at 600 lumens, which is impressive.
The Cree XHP50.2 4000K highest bin 70 CRI is rated at 1328 Lumens at 25C.
The Cree XHP50.2 4000K highest bin 90 CRI is rated at 1073 Lumens at 25C.

That makes the low CRI version about 24% more efficient than the High CRI. This can go up to 30% if the lower bin High CRI was used.
If we were to add 24% efficiency to the Zebralight at 89 L/W, it would translate to 110 L/W at the 410 Lumen level.
The math is oversimplified, but we can probably assume that the Olight is most likely more efficient than the Zebralight in the 400-600 lumen range.

From what I recall reading in old forum posts, the Zebralights are not as efficient in the higher settings, but are tuned to perform better in the lower and medium settings. Not sure how true that is, but it may be possible.

Great to see that Olight is releasing lights that are that efficient nonetheless.

Yea the H2R is really good. The unit I have has amazing 4000K tint to boot (–0.006 DUV at turbo and –0.003 at top of ramp) but that’s sadly the only Olight with good NW tint in existence.

Olight efficiency is getting even better now. The only 2 brands that provide runtime curves or specify step down points are Fenix and Olight so I’ve been studying their flashlight specs and comparing them with run time curves of other lights from flashlight reviewers. These two brands are really impressive in their super high efficiency compared to the rest. I just wish they provide better emitter options. Until then, there’s good reason to buy Zebralight and Armytek over Olight and Feix.

It was calculated by dividing the integral of lumens over time (lumen*hour) by the total battery energy (Wh).

lm/W = (lm*h)/(W*h)

Maukka compared just some flashlights with different LEDs and its just fact about run time in high mode , but not about efficiency.

Lumens / watt = efficiency. Not sure how you can say it is not. Take a look at the links to Maukka’s posts. Also please take a look at Maukka’s comment just above yours on how he calculates efficiency.

My Zebralight H604c does very strange things