I’ve always enjoyed your reviews. Hope I remembered to thank you, if not, thanks.
Unfortunately this does not work with my browser on Android.
tks for the link. I do use dark reader. built-in support would be more stable though. 3rd party extensions sometimes break with browser updates.
Unquestionably, the reviews on the manufacturer-supplied lights take an enormous amount of work, but is it really that beneficial, especially when multiple reviews of the same light end up repeating the same work?
What do I expect from a teardown? Some idea of how the light works, how well it is built, its modification potential, what kind of parts it uses, etc. For example, I saw a photo of a certain Zebralight driver board, and was able to read the part number from the microprocessor (a particular PIC model). By looking up the specs of that MCU, I could see it had enough memory etc. to be able to run Anduril. So that tells me there is some hope of being able to reflash the Zebralight and turn it into an Anduril light. THAT is information that I find interesting.
Also, look at the driver pics on lygte-info. You can see from how some of them are put together (multiple boards at right angles etc) that it would be amazing for a light with that driver to survive a few drops or vibration. If a light is made like that, the photo is valuable user info.
As for how the light is doing in reality, I think it is enough to say “I got this light, I liked it for X features, and ran into Y issues. Here (url) is the manufacturer’s page about it where you can find many nice pictures, so I won’t bother taking more marketing-style photos of it. Some comparable lights are Z and W and here are a few comparison notes.” And if it’s a technical type of review with a runtime graph, a vibration test would also be great, but I haven’t seen anyone doing these. (Vibration test = put the light on a paint shaker and run it for a while, at freezing and hot temperatures if you really want to be hardcore). Any light calling itself “rugged” or “tactical” should be tested like that by the manufacturer, but I doubt that many budget lights actually are. This is one area where Surefire used to be pretty good.
I’ve been told porting Anduril to a PIC would be a major pita due to the MCU dependant tricks used.
But ok, you have a point.
I totally agree
Seriously though, I try to make all reviews a tear down - can't really tell what's goin on til you open it up and look, if possible. Always seem to find something. Just wish I know more about the electronics, but I'll try taking high qual pics of all the electronics. Guess if a light is so tightly glued, it can always be cut open, and if I knew a light was free and can be destroyed, then there would be no hesitation.
Sometimes I do feel guilty though nit-picking at details on a donated light. One light I was given to review I felt so bad about ripping it, so below rated specs, that I never did end up posting the review on it. I feel though that was a mistake and won't repeat it - should have posted it, many do, and sort of gloss over the fact the manufacturer heavily over-rated it, but I won't do that anymore (sorry freeme!).
I do feel giving manufacturers benefit of the doubt. I'm ok with spec'd lumens up to ~15% over what I test, there's just too many variables, and the maukka lumens standards had been ~12% under what I've seen spec'd from top name brand manufacturers (ThruNite, NiteCore, etc.) in the past.
I should though do and publish runtime tests, because I totally agree - it's rarely spec'd by the manufacturers. The standards clearly state the 30 sec rule, so after 30 secs, it seems ok if anything happens. Also beamshots, least on a white wall, will tell you quite a bit about the beam, though personally, I'm not too fussy about the patterns - what we have today with LED technology is so much better than the black center hole of lights from the past. But still, I can appreciate the demand and interest of overall beam quality.
There's so many other things of interest:
- parasitic drain, across modes if relevant
- battery size compatibility
- weight and dims
- accurate amp draw (various modes, cells, etc.)
- driver design details
- charging circuitry (amps, compatibility, etc.)
- quality of anodizing - not sure even how to do this, but if you can damage the light, then you could come up with a standard test of maybe sand paper, file, or some other means that can be consistently applied and compared
- the optics - usually not directly rated/reviewed but the resulting output, beam pattern, and throw are all results
Sure there's more.... For any review, always better to keep it short, well organized, and able to use as a reference on specs.
I buy all the lights I have reviewed.
If someone sends me a free light, it would get the same review.
I have had two manufacturers contact me about doing reviews. I told them I would report on the good and the bad.
No free lights so far. Perhaps my lack of a paypal account had something to do with it.
Pulling some lights apart is easy-peezy.
Some take divine intervention. And way more than just a screwdriver.
Since I’m buying lights for my own use, I don’t want to take the chance of bricking one just to see the guts.
Though I do like seeing teardowns - particularly if it’s something I own.
I think the majority of reviews here are done for interest. And yes there are a few that do look “sponsored”.
Spending days to make runtime graphs, do a write up, and take good photos is certainly not something to do unless you are in it for the joy of contributing.
And then - sometimes - crickets.
All the Best,
The runtime graphs particularly seem time consuming if done manually. I’ve done a few informal ones without even any graphs, and it was a pain. I think HKJ has a fairly automated setup and that must help a lot. I don’t know how other people do it. It would be nice to have a standardized, simple setup for this, e.g. ESP32 board in a box with light and temperature sensors and some logging software. You’d put the light in the box, start the program, and later download the measurements.
I find the studio-quality photos nice to look at, and if the reviewers make those to enjoy exercising their photography skills, that’s great. But they do look like marketing. I’m happy with a few phone camera snapshots of bits that I can’t easily see in the manufacturer’s pictures, with the rest as text. I reviewed a number of lights on CPF, and the reviews were just writeups of my impressions. I don’t remember ever including pictures, and if I did, they were unquestionably crappy pictures intended to just convey some kind of information.
Also, maybe it’s just me, but while I do care about beam quality, I find I don’t look at beam shots very much. It’s enough to have a text description: smooth floody beam with pleasing warm color, that sort of thing.
No damage to home but work is without power for another 1 to 3 weeks. The wind blew hard here, but no trees down. We were on the east side of Ida and its surprising how 30 miles can change the landscape.
The reason so many trees and electrical lines are down is because we had a record rain fall this year, and it rained for weeks daily soaking the ground making it soft and muddy until the day Ida hit.
I’m glad to hear you are ok, and to see that you have internet. I hope the area recovers quickly.
Doing runtime by hand sucks. I used to do it by videoing a lux meter and taking sample readings at various time intervals during playback. This gets old in a big hurry!
Automating runtime is fairly easy if you can come up with something to log voltage over time.
The sensor setup is something Terry Oregon came up with in his huge post about testing photo diodes for looking at light waveforms.
I then used his design to start my own measurements.
This works well as long as you are careful not to let the light overpower the photo-diode and clip the signal.
The OSRAM diode is very linear as far as I can tell. This same diode is used in many of the inexpensive Lux meters on the market.
I use PICO log software that comes with my PICO oscilloscope. But anything that can sample a voltage over time will do the trick.
A computers sound-card will work - just be careful not to exceed the input voltage max or it will fry your card (ask me how I know).
Some use a logging multi-meter (kind of expensive) or some sort of software/hardware to do A to D conversions and create a time-voltage log file.
I then use my Lumen Tube to get reference Lumen levels to correct the voltages measured to the actual Lumen output.
I make the plots using Excel or Libra Office Calc.
All the Best,
Thanks pennzy, I appreciate your words
When I wrote what I wrote, I was not complaining in that way, but just stating that even after many work for a review, sometimes people don’t care about reviews that are not “the mainstream” or of certain brands that are well known, well “adverstised” and even sold in some GBs here on the forum. But these mini rants of mine shall not be taken into account
Ooops, sorry! I never tested it, as I normally use the site as it is!
I just remembered that thread and posted in case you guys didn’t know it and wanted to try using it Sorry if it doesn’t meet what you are looking for.
As for the suggestions of Forsyth P. Jones, I guess they are interesting but I am not sure if they are feasible in some ways, specially because it is supposed to a) envolve money, b) envolve people’s availability when we know that sometimes we “leave” for a while or for good!
Still, I’ll be following the thread development and at least I can try to modify some things in my reviews (whenever possible) to provide some information “for the sake of flashlight science”
Comments about reviews are always welcome. I should think.
How else can the reviewer learn to improve their offerings.
All the Best,
PS one of my pet peeves is a review with many many ultra high res. photos.
Downsize those suckers so a page can load in my lifetime!
Actually if all you are after is runtime.
A simple photocell will convert light to volts.
Then something to convert analog volts to digital and log that over time.
All the Best,
Yes if I do runtimes I’m planning to use a simple phototransistor and an MCU board. I hope that’s enough. At first I was going to write Raspberry Pi Pico but then realized ESP32 would make things even simpler, since I could download the data and monitor things to my laptop by wifi, or put long term tests (sublumen moon modes that are supposed to last for weeks) online so people could monitor them over the web (almost as exciting as watching paint dry!).
I guess a thermal sensor would also be worth having, I’d like to have an automated way to shut off the light if things get too hot: any ideas about that?
Once we have AVR-1 Anduril drivers we’ll be able to have flashlights with real API’s like God intended ;-).
Ahhh but has anyone ever done a Freezing/Boiling test yet?? Video at the middle of the page. Boil it good!
"Run over by a truck, dropped in boiling water, set on fire and even sent to SPACE (111k ft)"
> Automating runtime is fairly easy if…
Thanks! That post looks very helpful for the electronics part. I’ll probably end up going a different route for the software.
A Benetech GM1020 and an Elitech RC-4 will do. The MINIX diagram was done this way.
It was some years until I truly learned just HOW MUCH time it takes - not to mention the investment of real money in equipment - to do good solid comprehensive reviews. We have many here - some still around, some gone and moved on - who have done great reviews and/or teardowns for us. I am super appreciative of that and also have approximately zero sense of entitlement that would suggest I should complain when I don't see what I maybe wish I would. If I were to do so then I'd probably start feeling a little guilt over not investing that time and money myself to give the same thing back.
When I see the "marketing slick sheet" type of reviews...well, they are what they are and they serve their own purpose for people who are shopping for the light or perhaps new to the hobby, etc. Sometimes those reviews add a lot of great photos that show things not otherwise seen in product listings, and sometimes those can be very helpful to a modder even if measurements are not given on key parts (although it is fantastic when people take the time to add measurements for us). The reviews that go a step further with beamshots and especially run time or light quality graphs and such increase the time and dedication exponentially. Add teardown to that - assuming someone is willing and also willing to risk damaging parts or finish - and the time goes up still, plus the added time of inputting the information into a forum post/repository.
Not everyone is interested in the techy details or the modding details and this site serves all kinds of interests for folks. I just take those great tech and teardown reviews as a gift and try to express my thanks for their efforts, and when I can't find something I want/need about a light, I enjoy the exploration of discovering it myself on the bench in front of me.
As for a wiki type repository...man those are hard to keep up with and there are SO MANY lights these days. And compiling that info from so many available sources (here, youtube, reddit, websites, datasheets, etc, etc) could be a real labor of love. But that said, we've done it before and even if those things don't keep up with the future they're still great and worthy of the efforts (such as the excellent emitter repository thread here). I don't know how folks like Parametrek keep up with it and do such a good job...solo, and for the long haul. Based on some past experiences, my opinion is that it's also good to keep in mind that trying to use forum platforms as a sort of encyclopedia or reference base is difficult at best. Organized threads can be done, and stickied, and search fills up the gap. Trying to inject a more defined structure into an organic environment like a forum has a lot of drawbacks, as opposed to a for-viewing web page and such.
I think sometimes we take these people for granted and just expect them to keep on doing it or that they'll be around forever. I'll be the last one to suggest that someone isn't doing reviews the right way or should feel like they need to invest more of their own time or resources. Mukka, HKJ, TK, all the great guys mentioned above, the steady-reviewers who churn out videos all the time...it's great. If anyone falls short of information then either I can try to add to it, find it elsewhere, or explore it myself.