Interesting. The S21D is already very floody (and Iām a fan of flooders). I would think that the S21C having less output and even more flood would be a very very niche product.
Thanks guys for the detective work. Pheww nothing abnormal with my 2 lights then.
Is the picture below how you would measure this angle? If so then these are some rough numbers:
The Convoy GT-FC40 triplets: all 3 have similar angle around 70 degrees
Reylight Pineapple Copper with 219b 4500k: 85
The TIR lights do have large spill angle, but very dim spill as youāve mentioned.
Olight TIR XHP35 HI 4900k: 100 (donāt bother checking CRI on this one 
)
Sofirn IF22a: 120
Are multi LED TIR not a consideration? If yes then Emisar D4V2 with 219b has angle of 125. Hope this helps.
Yes, thatās exactly the info I was looking for. Thank you.
I feel the same way. I like quad LED output and S21D is nicely floody already. In fact I am going to experiment with swapping S21Dās TIR Optic with the one from Sofirn IF25a, which is less floody but gives a brighter hotspot. Apparently they are interchangeable.
White wall shot of the triplets. The turbo run-down at home (no ventilation, sequential pictures for ~20 minutes) reflects fairly well real-life experience (turbo start during my 45 min walk, subjective eyesight eval).
M21E always has brightest hotspot and dimmest spill for the first 5 minutes of a run-down from Turbo. Once thermal regulation and step-downs have occurred, M21E remains slightly brighter than B and F. Note M21E is the largest & heaviest of the 3. More of a jacket-pocket (vs pants pocket) light. ~7.25 oz w/ battery
M21F has brighter hotspot than M21B when first turned on, but once thermal regulation kicks in, sometimes I canāt tell these 2 apart. I tend to think of this as the M21Bās twin, but with side switch and on board USB charging. ~6.7 oz w/ battery
M21B has dimmer hot spot but OTOH has nice spill that illuminates near field well. Addition of metal tail switch and clip makes this light so much more āloveableā lol. Itās the smallest and lightest of the 3 lights. ~6.1 oz w/ battery
Essentiallyā¦ I love all 3. 
Pic of the smooth reflectors.
I would like to add that from beam shot above (post #55), it seems as the reflector gets larger (the M21E), the border of the hotspot becomes more delineated, more defined. But one, this could entirely be in my head , and two, itās only seen with white wall hunting. IOW, IMHO for academic interest only and not noticeable or relevant in real life use in this case.
Thank you for all the updates. M21F is looking so damn good. Canāt wait for mine to arrive.
Youāre welcome and thanks. Yes the M21F seems to be the sweet spot for EDC GT-FC40. I know the feeling - itās like being a kid again waiting for Christmas day .
Automatic Temperature Control (ATC): All 3 lights have ATC - and brightness seems well affected by it. Within a few minutes of Turbo start, thereās that big step-down seen in all lights that I have (not just Convoy). In addition during my walk I could feel the different lights become colder or warmer at different times. I do not have any instrument to measure precisely the brightness change, but could obtain rough estimates by taking sequential pictures, at 1, 2, 5, 10 min, etc. after a Turbo start.
Note that all tests inside are excellent to see precisely the light beam and brightness level, but they may not reflect accurately actual use outdoor wrt temperature regulation. For example the lights get extremely hot during indoor test, and never so in actual use during my walk. Outdoor ventilation makes a big difference.
When I asked Simon about the difference between all the Mxx GT-FC40 flashlight models, he mentioned the M21D is at an advantage because of its āthermal capacityā. I am guessing that as the light gets larger, itās not just a deeper/larger reflector that is different, but also the ability to dissipate heat from the inherently larger space of the head. In turn this should improve the ability to maintain brightness. Following the trend of B, F, and E, I would expect M21D to have the brightest hotspot, dimmest spill, and best ability to maintain brightness?
I have 3 Olight M2R Pro Warrior w/ XHP35 HI (flat) LED and they each have distinctly different brightness (the Orange one is brightest, Patriot dimmest) and tint (Orange least green, Patriot green hell) 
. For the cost, I was hoping for better quality control.
I was relieved when I saw the beam shots of the GT-FC40 triplets having similar color/tint. Clearly coming from the same bin and thank goodness that my comparison pictures for M21B still applicable to E and F.
The M21C/D and M3-C have more thermal mass. My testing has shown sustained brightness at approx 1000 lumens with GT-FC40 5000k in M21C and approx 1400 lumens with XHP70.2 5000k in M3-C.
Thanks. Thermal mass meaning aluminum frame of the flashlight acting as heat sink? I was just thinking about the larger air space above the LED but this makes sense to be the more important reason.
Yes, basically more metal to heat up, and radiate heat. A good flashlight needs to be able to transfer heat from the LED and driver to the outside shell/host with good thermal pathways, otherwise the thermal mass wonāt help much.
Olight M2R Pro Warrior was a popular thrower with great reviews, its beam perfectly designed/shaped with TIR optics for use in my evening walk: large, bright central hotspot to provide excellent illumination far field, dim spill but very floody (wide angle) to illuminate near field just enough.
The limited edition of these M2Rās uses Cree XHP35 HI (flat) LED, which has a lower, warmer CCT of around 4900k compared to the regular version. It is therefore one of the very few EDC throwers with a warm beam. The lower CCT enhances depth and color, as opposed to my Nitecore MH12S with its very cold, very blue 6500k CCT.
Olight M2R was my most favorite walk light, then the Convoy GT-FC40 triplets came and showed just how green the beam color is . This is unfortunately something that I canāt un-see.
Just took my first evening walk with my new M21F GT-FC40 4500k. I immediately applied some DC Fix (as I always tend to do) and it proved to be a very nice experience. Not as bright as my TC20 V2, but I wasnāt expecting it to be.
Since I applied the DC Fix almost immediately I canāt say much about the performance without, but with it applied it is a fantastic blend of throw and flood. Tint and color rendering are superb, brightness is more than adequate on 35%, and heat is well managed at that level.
The light handles very well, although the complaints about the recessed button being hard to locate and easily confused with the charging cover are well founded. The light is deceptively long, longer even than my Thrunite Catapult V6, but itās relatively thin handle and good balance make it very comfortable to use for an extended period. The UI still has its rough spots, but overall Iām very happy with it.
Of course the value here is incredible. $37 for this light plus the included 21700 is fantastic. If you can live with the UI quirks, I highly recommend it.
Good points. This latest (?) Convoy UI for the M21F and M21E, for me anyway, is still not a match for Anduril. But it at least is moving in the right direction versus the 12 Groups of M21B. Battery level is now directly accessed with 5 clicks. BUTā¦ the 5 clicks have to be fast, too fast for me so I missed it a lot of times - do you have same problem? I donāt know whether the problem is with the feel of this particular button (itās not as light or as easy as with my all time favorite - Sofirn IF25aās) or with the firmware, but itās really hard so far for me. I found Andurilās 3 click to battery so much easier to access.
I found the clip to be a must for these lights to not confuse power button with the USB charging port and also to quickly āblind-locateā the button. I took picture below to show decision-process for clip-on versus screw-on clip for potential buyers. Hope it makes sense . (The screw-on clip is MUCH better to use and in appearance.)
It would be a pretty simple matter to drill your tail cap on the Eā¦just position them accurately. Plenty of meat to do so without worrying about pullout from tension/use.
Thanks for the suggestion. Have to stay with clip-on because I am on the clumsy side as a handyman and me taking a drill to this beautiful light would honestly be like watching Death Wish 3 :person_facepalming: .
Same Olight M2R pictured above versus M21B, now illuminating a real scene, my front garden. Shots of throwy lights are difficult, as sometimes (as in this case) part of the scene is in the hotspot, and part in the very faint spill. I tried my best to adjust expo so it doesnāt favor one light or another. As usual, pictures taken in RAW format with Canon DSLR, white balance 5000k.
At any rate, GT-FC40 has remarkable ability to render colors as seen in real life: from lavender plant left front, to the brown bark of the tree in the back. In fact when it comes to a light brown surface area such as frequently seen in gardens and nature trails, I actually prefer GT-FC40 over Nichia 219b. The very strong magenta tint of 219b could be a bit too much at times.
My beloved M2R is not as beloved anymore lol. Actually in real life the green tint is not that bad, as long as you donāt do back to back comparison. But here you could see how overwhelming and ādestructiveā that green tint is. When even garden rocks look green you know your LED is in trouble.
Comparison with a Cree LED is easy - some are infamous for a horrible green tint and unfortunately my XHP35 HI 4900k is a victim. Another one that Iāve experienced is XHP70 Neutral White in Thrunite T2, which was a true green hell .
It is the gold standard Nichia that for me makes the most interesting & tough comparison for the GT-FC40, and next up is the very well known 519a, here in 4500k guise. I was so excited when I placed the order - what not to like about a new Convoy with Nichia. At any rate, 519a has spectacular Ra and R9 (in the LED used by Convoy Ra 93.6, R9 82, R12 80, Duv 0.00091 ), but does have a yellow/golden tint, reflected by its positive Duv.
I still love my Convoy with 519a a lot (the same way people like a variety of chocolate ), but the yellow tint while not nearly as bad as XHP35 HI does seem to show up in A:B comparison, more subtle but there nevertheless. To my eyes, rock and bark of tree take on a yellow tint, more yellowish than live view. All pastel/light color subjects such as the leaves of the lavender plant take on a slight yellow tint. Probably as a consequence, subjects donāt stand out from each other (color āpopā in photography), as seen with either Nichia 219b or Getian GT-FC40.
I was about to ask this. Need to be same type of AR lens for apple to apple comparisonā¦
Thanks for sharing cannga, all your pictures are great. But, I had hard time following those many āeditā across all yours postsā¦