Yes, I have only measured the first batch samples, and they were wonderful, ‘tasefully’ low duv. I am yet to put some of my third batch samples into a light, but based on other’s experiences, they seem more in 219b pink territory (though more warmish pink vs coldish 219b pink).
The two reviews you linked are for the first batch, which had duv around -0.060. Note that duv goes up as output increases - contrary to most other LEDs - which I personally like.
The third batch samples are around -0.070 to -0.090 duv.
Here are two measurements I found on Reddit: bob_mcbob fweep
On arrival the Firefly X07A instantly became my most beautiful and well thought out light. The neat magnetic USB cover, with a little cut out windows for prying finger, is fantastic. Those of us who have pulled out a flashlight’s typical USB rubber cover accidentally (followed by multiple “oh s…” expletives) will appreciate this. The captive clip is another favorite feature. It allows me to locate the clip at my favorite 10 o’clock position and I wish more lights would have it.
The knurling’s are deep, both pretty and functional, not like for example my very favorite Convoy S21D (kidding). The cooling fins are EXTREMELY deep, don’t know if they do any cooling fast enough, but they are certainly very beautiful to look at. Functionally and cosmetically, everything about this light screams refinement.
Preliminary numbers, more to come: the tiny light is rocking 4300 lm in Turbo mode with Molicel P43B, 4200 lm with the Nealsgadgets Mystery Lishen battery (LR2170LH). The little light could amazingly lit up the side of a small hill next to my house, but considering how tiny it is, this should be the most ephemeral 4300 lm I’ve ever experienced . Also oddly and unlike most of the LEDs I’ve tested, particularly 519a, the Duv seems to become less negative with higher output (ie less rosy).
Waxing poetic about the FFL350A LED, I would say it has a warm, rosy, very addictive color rendition. YOLO, you must have this LED if you’re a collector. It’s the first LED in my small collection that makes me critical of my multiple 219b 4500k lights.
No IMHO it doesn’t dethrone 219b (for me, still the “reference standard”), but it does make 219b look sterile/ whitish/cold in comparison, sometimes. It also is the rare LED that survives an A:B white walling vs. 219b. Tint snobs among us likely know what I’m talking about. 219b makes a lot of LEDs look yellow or green, but not the FFL351A.
When I first turned on the Firefly, I was thinking to myself wait I’ve seen it somewhere else! Well that somewhere else is my S21F with 519a tint blending 519a 5700k + 2700k in Turbo mode. The color rendition is nearly identical IRL, the difference being S21F is slightly less rosy, and cannot match the brightness, either transient or sustained.
Thanks. It’s a brass D4V2 with Nichia 219b 2700k. Picture of the copper D4V2 is a couple posts above that picture, the light on the right side with green aux and green button, and also HERE.
Yes, thank you for sharing about its coming! It sits in my cart already!
I plan to go with the SFT40 version (due to their low Vf at higher currents), as the the smaller LEDs are likely to draw less power. With 4 emitters, the Vf of each matters more than with the 7 used in the E07x. (With 7, the current is more limited by the lume1 driver’s RPP resistance)
Already in the cart ? Nice! I am having a FFL “lovefest”; FFL351A Rosy Tint’s color rendering is so beautiful so this next one one will be FFL505A for sure. I already have a SFT40 light anyway, so decision is easy.
Still VERY impressed with Firefly E07’s fantastic quality, having been a hard-core Convoy fan for so long lol. What color are you getting? I am having a tough time between red and green. The red is so unique but the green is beautiful. And so is the blue!
Picture from Reddit, probably of the first group of FFL351A, the 3700k group.
Related to the high R9, this FFL351A 4000k (Rosy Tint) LED has the ability to make red flowers and fruits etc. absolutely POP at night. 219b 4500k is more whitish, probably more like daylight, and makes objects clearer, but the warmer FFL351A makes red color pops and scenes more 3D, IMHO, influenced by my photography hobby. Both LEDS are great and beautiful, and both have strong points.
If you do order, remember the very important Boost Drive (lower peak brightness, but more efficient and higher sustained brightness ), stainless steel bezel, and deep-carry clip are options that you have to check. I also checked on flat button (raised button annoying to use for me) and cool RGB button (even though I don’t know if D4K has the RGB voltage check function). The LED of course is 519a 5700k dedomed.
I do not know if ToyKeeper’s choice of steel bezel is cosmetic or function (more durable, anodized aluminum affecting tint?) but I already have one for my D4V2.
Cheating ANSI to get high throw numbers, or, there are TWO SIDES to every story?
For a 4300 lm light, I was disappointed to see my throw measurement of the E07X, a measly ANSI throw of 135 m. I had even tried to “doctor” the result by using one of the most powerful (current wise) available, the Molicel P45B.
Turned out, the “problem” is not the light, not the battery, but the “ANSI” part of the spec. ANSI throw requires measurement 30 seconds after turn-on. And this 30 seconds is a matter of life or death, not kidding, for tiny Power Light such as the E07X. Combined with P45B, E07X has the most ephemeral Turbo LOL I’ve experienced, lasting no more than about 25 seconds. The output literally “crashed” right before the 30 seconds mark.
Then the light bulb went off, and in came a Wurkkos battery. Problem solved because the Wurkkos by not starting so brightly generated less heat and stepped down later, after 30 seconds. Who would have thought a lowly Wurkkos battery could beat the almighty P45B? I am going with 270 m so I could feel happier in this Holiday Season. Kidding aside, lesson learned!
Throw with Wurkkos No-Name Battery = 270 m
Throw with Molicel 45B = 135 m
Very nice observation and experiment. The ANSI ratings are far form perfect, and some lights are known to cheat runtimes by, for example, sustaining 30s of nominal output and immediately dropping to 10% for the remaining run, which inflates turbo runtime by as much as 10x.
The fundamental issue here is that output-over-time can be an arbitrarily complicated function, and sampling a single data point at 30s does little good in giving a picture of what the whole function looks like. The ideal would be to ask manufacturers to give a runtime plot of the light instead of an ANSI output/throw/runtime rating.
You meant like designating testing at say 20° C so think-you’re-so-clever tester doesn’t leave flashlight out in the balcony so it cools to 10° C before testing? Doesn’t work btw (grin), at least in case of E07X. The power of 7 FFL351A LEDs in a tight space on Turbo with Molicel is too much for this shenanigans I was trying to pull.
Yep, but I’m afraid even so 99% of readers won’t appreciate the fine point. Like tax codes, no matter what people will find loop holes, be legal but not in the spirit of the law. I think this is where the role of a “good” reviewer steps in, to provide a thorough subjective description.
The throw of the Firefly in real use is in between small Nichia light like Emisar D4V2 or Convoy S21D (~150 m) and a medium thrower such as M21E with GT-FC40 (~300 m throw). As such, neither number above (135 m & 270 m) describes this accurately. This note is important because for my use case, S21D and D4V2 is too dim to be a good walk light, but E07X is.
E07X joins the small list of lights with Nichia-class tint that throws enough and is bright enough for me to use for safety in my evening walk. The Turbo function is meant for temporary use ONLY, for example if I want to briefly look to see if there are coyotes or other danger ahead in the walk. The Turbo brightness is truly “something else.” Enough to lit up whole side of a hill in my neighborhood.
This is very sad but true. And I agree 100% about this being a responsibility of the reviewer–to measure objectively and to describe subjectively. Sadly there exist many, many reviews on YouTube that do not offer anything useful beyond manufacturer claims, but there are always amazing folks like ZeroAir and Weerapat to fall back to.
Taking pictures to describe the different between 2 LEDs is difficult since there are so many variables, starting with the monitor (the difference is clear on my OLED laptop, not so on my IPS LED laptop). I tried my best but if you don’t see a difference, you just have to rely on my subjective description, sorry.
To my eyes, 219b is more whitish, more like a midday light, whereas FFL351A is warm like early morning sun, but not yellow. It’s really really rosy. FFL351A has that unique tint, it out-rosies the rosy 219b. It’s the rare LED in my collection survives an A:B white-wall shootout with 219b because it has zero green tint. BTW, the 219b makes the red flower a little pink, and on this aspect alone, the FFL351a renders the red bougainvillea more truthfully to my eyes. FFL351a makes red color pops like nothing else.
Bottom line: 2 very different LEDs, different but not better, nor worse. If you’re the collector type, IMHO you “should” own both , and if you decide to go ahead, make sure to select the one with the description “Rosy Tint” since not all FFL351A’s has this tint.
Firefly has fire sale ( pun intended) 15% off, code FFBFS15NEW. I am ordering the fresh new release E04 with this LED: " FFL505A 3500K CRI95 3200lm 480m (Rosy)". The tint “rosy” is part of the spec description and I like this practice. I prefer 4000-4500k LEDs so this is a little low, but the other choice is 6500k that I like even less.
I am leery of being first buyers and hope I won’t be a beta tester. But shoot-out between these new Fireflies and D4K Dedomed 519a 4500k on Christmas day is too good to pass up.
I was going to order green but the blue is so pretty. @14500 hope you caught this.
. Also oddly and unlike most of the LEDs I’ve tested, particularly 519a, the Duv seems to become 
Nice! I am having a FFL “lovefest”; FFL351A Rosy Tint’s color rendering is so beautiful so this next one one will be FFL505A for sure. I already have a SFT40 light anyway, so decision is easy.
Kidding aside, lesson learned!
pun intended) 15% off, code FFBFS15NEW. I am ordering the fresh new release E04 with this LED: " FFL505A 3500K CRI95 3200lm 480m (Rosy)". The tint “rosy” is part of the spec description and I like this practice. I prefer 4000-4500k LEDs so this is a little low, but the other choice is 6500k that I like even less.