Luckily 519A is not rosy pretty much at all. That’s good because I’d say rosy tints don’t have as many fans as flashlight nerds may think. I hate them, I swapped an emitter to the high-CRI 219B once and it was disgusting, I never used that flashlight since. Luckily that has changed, 219C, 519A or B35AM are more creamy than rosy, they improved brightness /efficiency a lot and it worked - it’s pretty safe to say Nichia emitters have never been as popular as they are now.
1 Thank
I just bought another one yesterday.
This is my favorite Sofirn light!
However, I’m not a fan of the SST20 4000K, I’d prefer any Nichia 4000K over the SST20.
1 Thank
TBF they are pretty neutral when new, they only really become rosy when dedomed and that’s a conscious choice you make when you’ve taken a flashlight apart.
I’m a huge fan of dedomed 5700k, the end result being rosy 4200k but I know it’s not what everyone likes.
1 Thank
I’m getting a bit off-topic here, but standards agencies are considering a revision to the standard for what counts as “white” light. If passed, lighting sources would be required to use a lower duv (pinker tint) in order to get official certification like an EnergyStar rating. This change is based on a few experiments conducted at different times by different researchers, where the results all gave very similar results. An overview is here, at EnergyStar’s site.
The short version is: They measured from about -30 to +20 mduv, and there were people who preferred every part of that spectrum, but the average was consistently pink. About 5% to 10% of the study participants preferred a positive duv, while about 2/3rds preferred negative duv. The overall average was about -15 mduv.
The “pink” dedomed 519A is only about -4 mduv.
What people prefer and what people view as “white” are a bit different though. For “white”, the results go from about +6 mduv at 6500K to -10 mduv at 2700K. The 519A’s -4 mduv at ~4000K fits right along that line.
9 Thanks
Depends how heat resistant your hand is 
The Q8 Plus is one of the best hotrods around IMHO.
I very strongly dislike this attitude - ultimately it’s based around the assumption that people are stupid, nothing more than that. Just take a look at the enshittification of the internet for what happens when people believe in that stuff.
Agreed, and I’m not even that old. I have plenty of high CRI lights which I largely see in the same “practical, not really boring but definitely not something I grab to put a smile on my face” category, but they don’t make me smile the same way as a portable sun does.
Agreed. I have two SP36 now, but for a long time I didn’t have any, while the Q8 Plus was one of the first Sofirn lights I ever bought. One thing might be to make a SP36 with series batteries, and throwy 6V LEDs with that smooth reflector. 719A for the CRI purists, and XHP50.2 HI for people who want a memeblaster.
1 Thank
I have 6 sofirn lights and I try many other sofirns.
New sofirns have cheap unregulated drivers with bad efficiency.
For me best sofirn flashlights are SP31V2.0, SP35, D25L. I always advice beginners to try SP31V2.0.
I think that Sofirn make mistake because we cannot buy parts for their flashlight unlike Convoy which sell everything for their lights and give a big choice of led types for almost every Convoy model.
SP33s is good light but it has to be a little more throwier and You must put a 70.3 led in it.
Sofirn doesn’t have light like Convoy L21B.
C8l is good flashlight but making thrower with led which is not native thrower led is wrong.
C8L with SFT 40 and buck driver will be much better. SP 10S have better user interface than SP10V3 which is good light.
Not everyone wants anduril, You must make something simple and efficient for wider circle of customer.
Just to clarify, the context of this discussion is “too few dollars chasing too many commodity durable goods”, the lack of differentiation, poor business models, execution, market fit, and product development practices … not the enshittification that Cory talks about in double sided platforms, Technofeudalism or even the lack of choice in online services in general, due to monopolies controlling the means of computation.
I love Cory’s observation btw, myself being somewhat trapped in Google’s Android Play Store which is heavily enshittified.
1 Thank
I require Anduril 2 on all my new lights, and I am still most interested in lights ~4000K. I agree with many of above comments regarding changes in the market: Sofirn used the be the only game in town for a lot of lights, now there’s serious competitive options. Some examples from my personal experience:
- For an “around the house” light, I used to recommend SP36. Good UI, good output, good battery capacity, USB C charging, at a good price. Nothing could touch it. Now? I almost always recommend the Wurkkos TS25. ~50% battery capacity, but you gain a very high quality emitter, RGB aux for much easier battery output reporting, a tailcap magnet, and it’s in a more portable size if it needed to be.
- The SC21pro used to be the only option for a tiny Anduril light. I’ve bought and modded many of these (putting 519A in many). However, for my own collection, Emisar’s D2 has completely taken over what I used these for. The D2 can be bought with the LED I want, the 2 channel nature makes it a size-to-feature ratio powerhouse, again has RGB aux for easy battery level observations, and it also has really high quality construction (nice to show off).
- I’m going to gift some pocket throwers to some friends this holiday season. Sofirn’s IF19 and Wurkkos’s TS11 are really the only picks, and while I think the construction of Sofirn’s is prettier, there is no way I’d ever pick it. It uses a worse LED (SST40 vs SFT40), and lacks the RGB aux that the TS11 has (again, for easy battery level reporting).
There are other examples from my experience and even more from others (driver type is a good example too). I don’t think market has moved away from 4000K Anduril lights, I think it’s moved away from Sofirn’s offerings in that category because they’re not as competitive as they used to be.
Not trying to be harsh, I really hope to see more competitive options from Sofirn and value the brand a lot. Hoping good things come from this discussion.
6 Thanks
You can buy parts. Just message them, they aren’t advertised.
Ya that’s a good idea. I haven’t really touched my sp36 since I got the q8 plus. But I don’t think it should really be a priority for sofirn right now, there’s other bigger things. But down the road ya.
I think the two biggest things are
- the 4000K sst20s and 5000k lh351d’s. Idk if they’re purposely getting the greenest tint bins to try and get the most lumens while keeping high CRI, but it’s not a compromise that works. Pick a direction, either screw CRI and go cheap, cool blue, for high lumens, or go 519a’s with good tint. Imo
- Wurkkos. They’re just doing it better right now. Sofirn has better customer service in my experience but whenever I decide I need a new sofirn or wurkkos light it usually ends up being wurkkos. Maybe sofirn gets some money from that, idk, so its not a total loss. But still
3 Thanks
Most likely due to inflation, costs of living & fuel, and people don’t have as much income & money to spend on luxuries like the LT1 & Sofirn’s other great lights with Andruil, Andruil is my favorite flashlight & LT1 firmware. Maybe you should reach out to the big outdoor & camping store chains. Promote more on Amazon, etc.
Personally, if you put out some solid Buck drivers, that would be amazing.
I particularly feel like the LT1 would be amazing with a solid buck driver. Runtime on low is already impressive with the current iteration, now imagine if it was >90% efficient. The base LT1 would also be nice to upgrade with 2700K/5700K 519A.
As for anduril, for the most part I can’t tolerate E-switch lights without it.
I would also like to say the IF23 is amazing.
looked at my spreadsheet.
Last time I bought unregulated sofirn/wurkkos light (beside the TS10, it’s exception ), was 2.5 years ago. (and it doesn’t matter andurli UI, or not)
Since then, only regulated lights from sofirn/wurkkos, or TS10.
Like the TS26 really needs the efficient buck driver, I will buy it, it it will have one.
And well, for me button is also important, I will not buy that Wurkkos TD01, with that plastic button. You have 2 good rubber buttons, the one on FC11, and FC13, use these.
And the FC series, really needs the buck driver upgrade. Just imagine how good the FC11 would be.
1 Thank
Some thoughts from a perspective of a person who’s not into flashlights, and more of a casual user. I have a bunch of Anduril 1 lights, including SC31 Pro (not sure if it’s changed to Anduril 2 now, mine’s from 2020), and I don’t use any of the Anduril features, aside from the basics like click on/off, ramp up/down and sometimes lock. Furthermore, having the gimmicky features (blinkies, aux lights, multiple hidden modes, etc.) makes me think twice before giving or gifting anybody such a light, least they hurt themselves or accidentally set something on fire. I would much rather have something with an efficient driver, proper USB-C charging (unlike my early SC31 Pro that can only be charged with a C to A cable), simple UI that I can hand over to anybody, and a good tint and CRI. One of my recent favorite lights from Sofirn is the SC21 non-Pro (because unlike Pro it has a regulated driver and no Anduril) in 4000K which is my current EDC. I don’t have anything against Anduril, but having an efficient driver is a higher priority.
Looking at output graphs, it looks like 1+ FET to me…
Can’t find any concrete info on the driver.
For the UI, now anduril 2.0 have simple mode, that is actually quite easy, for non flashlight people.
charging on USB C to USB C is a must, as they won’t know why it’s not charging (and good if it’s on the flashlight body, so that the a non flashlight person don’t have to disassemble the light )
1 Thank
Amazon US listing has:
Driver: Constant current buck driver
Here’s ZeroAir’s runtime graph for non-Pro (full review):
And here’s one for Pro (full review):
2 Thanks
Looking at the specsheets, your observation makes sense. I haven’t seen a teardown, so perhaps the OP can clarify this.
I strongly agree with this. Most of my lights are from Wurkkos (and all except one with Anduril) and I had none from Sofirn (until I got an SC28 as a gift).
Besides TK, many people suggested regulated drivers and better LED options, specially Nichia. None of my Wurkkoses have either (albeit the Lattice Power LEDs in my original TS10 from last year are very nice, ditto the SFT40 in my FC13), so I agree that both these suggestions would help Sofirn differentiate and better compete with Wurkkos.
These three advantages, plus my great previous experience with the TS10, and the FC13 great throw performance (360m) for a light of its size and form factor, were what led me to buy mine. I’m sorry to say the SP31Pro wasn’t even considered.
Another reason that led me to buy Wurkkos instead of Sofirn was the former’s being much more present and in touch with the community: at least at the time I made these purchase decisions, Wurkkos was everywhere and Sofirn mostly wasn’t. In that regard, I think Sofirn is already working to remedy this, @Barry0892 outreach here (eg this very topic, and others, asking for input from the community) is both refreshing and promising.
+1 to this. @Barry0892, Anduril is definitely not the reason you’re not selling more lights (nor is 4000K for that matter).
I strongly agree with this. Treat all people as stupid and you will quickly be drowning in stupidity (it’s a kind of self-fulfilling ‘prophecy’) while driving away the non-stupid people. Sure, there’s a large amount of stupid people around (the prevalence of Winblows PCs, iDiot phones, and orange-tinted demagogues are proof to that), but these and other stupidity-catering ‘brands’ are already firmly entrenched and trying to out-compete them on stupidity alone will probably not lead to success.
4 Thanks
I agree with ToyKeeper’s answer.
The biggest factor for me is that I want an efficient driver (buck boost etc.) like one of the better designs like thefreeman’s driver.
Second, is good emitters. nichia 519A and 719A, luminous sft40, osram clspm1 or cslnm1, cree xhp50.3HI (80 cri minimum).
Of course, Anduril with standard flashing pads.
Combine this and it will be an extreme recommendation and I’ll buy a new flashlight.
2 Thanks
Okay, just another what-should-be-painfully-obvious observation…
None of these things mentioned affects The Muggle.
The Muggle doesn’t give two shiites about buck regulators or the UI or what flavor ATtiny it has or any of that. All these things are what The Enthusiast is all wet and sloppy about.
Now, I’m assuming that Barry’s referring to overall sales on their site, Amazon, etc., and not what The Community™ is buying up.
Like I said, The Muggle outnumbers The Enthusiast probably 1000:1 if not more.
And as for stoopit, too many of those people are buying “9990 lm” 3×AAA lights. Admit it. C’mon, admit it.
A documented problem with “focus groups” is that they “don’t have any skin in the game”. Kind of a repulsive expression and I have no idea what it actually means as far as origin, but I get its meaning. Focus groups want X and Y and Z. Demand it, even. They’re must-haves. But these same people, when it comes to BUYING it… won’t. People might push “green energy” that comes exclusively from windmills… right up until they have to pay 3× to their “clean” supplier what “dirty” energy costs.
The Enthusiast might be willing to pay double the price for a thick-shelf copper-pcb unibody tubelight with buck driver and flashing pads and wah-wah-wah what everyone else is demanding, but The Muggle sees the Crapfire POS special at half the price and buys that instead because it looks tacticool.
Suggestion? Do ONE thing in each class, and don’t make it suck.
Make 1 tubelight with a thick beefy shelf and plenty of thermal mass that can sustain the advertised brightness without stepping down for a decent amount of time, with a sideswitch that doesn’t suck and no retarded “thumb-guards” around any tailswitch, and make every other feature top-notch, and sell it as a premium light, touting those features over what the competition offers.
Same for a soda-can light, compact thrower, mini/micro thrower, keychain light, diving light, etc.
Having a lineup of 30 different tubelights, each a minor variation on a theme, doesn’t help sales. And an alphanumeric soup of new model numbers doesn’t help, either. To this day, I can’t keep track of all the SPs, SCs, WKs, TDs, etc., let alone the ‘S’ and ‘A’ suffices and various versions of each.
Colors are great, and I have no idea what would be better, whether to make color variants a rare treat, or just come out with multiple colors from the start, so I can’t offer any advice on that.
The UI? I get it, everyone and his grandmother wants andy3 and the like, but I don’t know WHY. It’s got the best ramping I’ve seen (doesn’t linger for 20min on the bright end nor fly by the dim end), but is it to be able to set max temperature? to have candle mode? what? I’m seriously asking.
I gotta say, to a person, EVERYONE who’s played with it, loves my Acebeam EC50gen2. Click on/off, press’n’hold for LMH. Done. Not even strobe or turbo or nuffin’. No shortcuts. It Just Works.
Other lights? I gotta explain steps vs ramping, just to have them figure out how to ramp down after ramping up… and up… and up…
Skilhunt E3A as well. Simple on/off AAA twisty. No modes. Beautiful beam. It Just Works.
That’s what people want. Normies, that is. My gf has a SP10A because she has no need for firefly (I got the 'B) and is better suited for LMH+strobe that the 'A has, vs FLM+H that the 'B has. But most times? She presets it to one thing and keeps it click on/off, rarely changing modes.
I went years with my 1-mode '502.
So I’m genuinely curious as to what complex UIs have that people demand.
The reason I bring it up is that especially now that amazon listings include the andy flowcharts, how many normies were simply scared off? They see that and think they gotta learn morse code just to turn it on/off.
And sad to say, that’s why we sometimes end up with retarded UIs that make you click through multiple brightnesses, and maybe even strobe, just to turn it off again, because they’re aimed at people who never graduated Walking And Chewing Gum Academy and can only handle simple clicks.
It’s projection of the worst sort to think that normies, ie, The Muggles, the ones who actually buy the lights 1000× more frequently than The Enthusiasts, want and even demand, what it is that “we” want instead.
So yeah, you can do one thing right, or do dozens of watered-down variations instead. And a great light that’s efficient, powerful, built like a tank… but has a horrible ringy rainbow beam, just won’t sell, at least not for long. Taking a reflector that’s been used for one LED and throwing in a different emitter might not go so well. Same with green beams, etc.
Just looking at the lights I use most often, I can say they’re simply ONE of each class. I might rotate through tubelights, but for around the house, I’m using my TS10 almost exclusively. I’m close to a charger, so who cares about runtime? Other than that, I got my iR2D2 on a neck-lanyard, my IF22A and/or GTmicro for throwing, my DC7 (Q8ish but with onboard charging) and Cometa just in case, and pretty much that’s it. In the kitchen is my Tacklife with assloads of diffusion film to make it useful.
Most of those lights have “find me in the dark” indicators, which is important, but mainly, I have one light for each function. And that’s my point, that The Muggle needs ¼ in holes, not ¼ in drills.
Do one thing. Make it right. Don’t let it suck.
Repeat for each additional “class”. Tubelight, thrower, compact thrower, behemoth, lantern, etc. Don’t spread yourself (or your resources) too thin.
Just puttin’ my stank on it…
8 Thanks