A quick look at the chart here on BLF looks like 686 lumen at 2A per emitter.
686 * 0.85 = ~583 lumen (15% loss for NW-seems a bit high)
583 * 3 = 1749 lumen. At a 10% loss, it becomes ~1852 lumen.
My 3-mode WW version pulling 3A:
881 * 0.85 = ~749 lumen per emitter with 15% loss for WW.
749 * 3 = 2247 lumen.
At a 20% loss, it looks like ~2114......
Well after a lot of thinking i thought NW would suit me best, although i have never owned a NW light but the nice color rendering combined with high lumen output would be great. But if it isn't 3A and the 2200 lumen i was hoping for i might consider the skyray. But i'll see what ric has to say.
The SkyRay looks really bluish to me, even more so than the Trustfire 3T6 (i have them side by side). This NW is 4000, maybe 4250k. IF you look at Cree's XM-L chart, the difference from 3700k to 8300k is 240-280 lumens. That is a whole 16.7% range. So 15% is quite belivable. I have a Xeno E03 WW and NW. With the same battery and tailcap, it is 38 and 44 lux reflected respectively. 15.8% difference. And do remember, lux meters are calibrated to have peak sensitivity at 2583k warm white, so that 38 lux for WW has the advantage. This is a real-world measurement.
If one really needs the lumens/lux, you can't go wrong with the CW 4A with short bursts, switching to medium for long use. If you live in the tropics and is hot whole year round, you'd need very high mass to soak up the heat, seriously even those 8 fins on the TF-3T6 does nothing much to cool over here unless you have a very strong breeze.
Well, i'll ask Ric if he can send me the so-called 3 mode 4A direct-drive drivers. I'll probably get the CW direct-drive 4A light as well to complete the collection. Remember, Chicago-X has the WW 3-mode direct drive and he gets 3A while I get 2.8A.
Trust me, if you own a lux meter, you'd see that the higher you drive, the faster the lux drops, it does not look what it seems. The difference in form factor between the regular 2 or 3 batteries "longish" Sky Ray/Trustfire/SZBOM vs this Dry is well worth the price, operated in medium/low mode this light is another version of the extreme runtime (actually this has higher lumens efficacy).
https://budgetlightforum.com/t/-/2525 (do play with the lumens calculator, input the different drive levels, and heatsink levels)
I can tell you, NW at 2A drive level, 150 reflected lux (vs 130 for TF and 160 for SR, both cold whites), are all around in the same ballpark.
And if you dig neutral white CCT, seriously there are no other options. 
You can get the Sky Ray, but mine is sitting here direct drive on either a SLA / 3S3P pack because the driver blew and many others also. I thought Vin posted shots of this and his Dry CW with the 3-mode 4A driver.
BTW, our eyes/pupils work in such a way....that 0.78A and 2A in closer quarters, is actually quite close. I just tried it last night briefly along a dark pathway in my estate....well my wife is afraid of cockroaches. LOL!
The difference on the lux and DMM is there, but in reality 2A shuts down the pupils somewhat. Of course you'd tell a difference if you are using this for throw, btw this light is not for throw and not useful as a dedicated thrower past 100m, well of course it still works and the output is appreciated, but a Fandyfire STL-V6 would blow this out of the water in throw. Correct tools for the correct use. BTW you guys do know the hotspot size matters a lot for lux (throw)? Twice the diameter drops the light intensity, eg your 4000L just became 1000L.
In a nutshell, if you are basing on paper 3A looks very much brighter than say 1.5A be it close up or for throw, then that is going to disappoint. In fact for lux throw figures, you'd need 4X the lux for twice the throw. So yes, say for my WW driven at 3A, it probably still does not reach 2X farther versus a 0.78A drive. :) (remember the 3A is going to put out a lot more heat)
No offence guys, i'd echo Chicago X and Vin's comments at cpfmarketplace. You guys are thinking and comparing too much, just buy it! Just the WW / NW variants PLUS form factor alone is well worth it. By chance the good solid built is a welcomed plus point as well....it won't beat the quality of Shadow TC6 that i received from Ric/CNqualitygoods, but that's yet for another use and appreciation. When you hold this light, this feels like a mini TK70.
You probably would not get more kick if you get the Big Bruiser from Elektrolumens, which costs 250 before shipping and has the same form factor. LOL! :D
In CPF there are guys who get lights due to this and not absolute lumens. But i can understand about getting too hung up with lumens on paper....but like i said in reality, the diff in a big house between 1800 and 2500 lumens is not really appreciable...too close. Now when i fire up my 4000 lumens Ebay "65W", 1800L vs 4000L, now we are talking.
To be fair, it's like almost twice the price. The fandyfire 3xml is <50 at dx. 2 of those is pretty close to your 4k lum HID. :bigsmile:
I'd never touch those drop-ins with no threads again. Try getting any dealer to pay you return shipping for the DOAs in that batch.
And Lightake will send you another Wii crazy bunny toy.
Just for fun.....Edit : 46.50 x 2 = 93 , vs HID's $122 here and $138 in USA.
Thanks 2100 for your comments. I understand what you're saying, and I'm sure you're right that I would probably not notice the difference, at least not as much as I'd notice the diff between the tints. What bugs me is that the NW 4-mode is "leaving something on the table". If it were driven at 3 A, I would be satisfied even if the output were the same as the Skyray or TF 3T6. But since it's driven at 2 A, while the 3-mode is driven at 3 A or 4 A, I would feel gypped. :)
Also, one of my minor targets was to be able to beat a Fenix TK45, and a Maelstrom S12. The CW 3-mode definitely does that, but the NW 4-mode probably wouldn't. How can I brag if I can't beat my friend's lights??? :) (and the S12 is quite a bit smaller than this Dry, so even worse!) In fact, whatever light I buy has to clobber my current brightest light, which is around 650 lm, or else I wouldn't bother.
I can't buy a 3-mode Dry because it has no real low mode. I could maybe try a CW 4-mode, but who knows if the CW 4-mode would drive at 3 A or 2 A?
Oh and lastly: "just buy it"... easy for you rich guys to say!
BTW, does the extension on the trustfire 3xml fit the skyray?
Oh BTW, 2100, if you have a way to communicate directly with Ric, can you ask him why the 4-mode DRY drives at 2 A instead of the 3 A he stated on cpfmarketplace?
Last favour to ask (for now): 2100, if you ever feel like doing beamshots, I'd love to see your two DRYs' beamshots side by side.
petebaby,
TK45 ..... Hey NW already has that 15% disadvantage to begin with. And like i said, for throw, the hotspot size matters way more. TK45 is more throwy, that 10-15% reduction in hotspot size (from Chicago X's pix) is going to require like 40%-60% more lumens to compensate just by "lumens brute strength". As in, 1800 lumens you need to push to 2500-2800L and that is for both lights with CW, and you factor in another 15% disadvantage of NW, sorry it is really too much.
To see which light is brighter via throw on a 200m target by using lumens is a totally losing battle.
Oh yeah, you do have a point. While the 3-mode 4A with the NW is cool, it lacks a real low. But then again at medium, that is going to run for nearly 4hrs. :)
I'll try. Seriously it is really academic on paper. Why does Chicago X's warm white drive at 3A and mine at 2.8A while it is specced at 4A? heh.... I'll just take things as it is.
I'm actually not interested in throw at all. I have a fairly throwy light (5x XRE) already, and find it not as useful as a floody light would be. From the beamshots I've seen, the Skyray and Dry look fairly floody, which is great. What I want is that good flood, but so much of it that it throws fairly well too. I don't need any more than 100 m.
Yep, they are the same hosts. In fact I am using the SR's tailcap on TF body.... nothing in particular....just for convenience because i had it lying around.
Good flood with throw = HID 65W (the Ebay/Aliexpress type). Really, can't beat that. Pipebomb though... its 3S3P. Not for novices.
I actually want a low mode (not SUPER low, but around 100 lm would be fine) not for the run-time, but because whatever light I get is going to be one of my two EDC lights. I've already used a 300-400 lm light to read a menu in a dark restaurant, and people were annoyed. :) Yes, one of those people was my spouse.
Heheh .... tough requirements.
Anyway, i mentioned above for my Xeno E03 WW is 38 lux reflected and 44 lux NW reflected. In fact i must mention that lux meters are calibrated for 2583k, so the advantage already goes to the WW. Cree specs indicates 240lm/200lm so that is 1.2X, 20% diff.
In fact i must mention that lux meters are calibrated for 2583k,
I don't think that's true. They should be calibrated similarly to human eyes in photopic vision which maximize response to like 5kK.
You can try comparing the graph here: hhttp://en.wikipedia.org/wiki/Photopic_lux to various approximations of sunlight or BB radiation.
The ISO standardisation is 2856k. The actual ISO file costs CHF 58 to download though. LOL!
Check out CPF's testing, there is a big enough sample base. The incans are very close, but LEDs are somewhat off most of the time. Most claim to be close to the CIE photopic curve.
http://www.candlepowerforums.com/vb/showthread.php?94232-Light-Meter-Benchmark-Testing-–-CPF-style
Guys like ma_sha1 mentioned that LEDs somewhat lack the yellow component, even compared to say 5000-6000K HIDs. That partially explains why a 125k SR90 can hang around with a nearly 200k HID. I use a compensation factor of about 1.2-1.3 (as you can see from the figures), others may factor in higher values. This is just talking about PBCP. Look at the R, G, B, totally way off.
Else there really is no way of explaining the differences between HID vs LED phenomenon in lux readings.
If they're cal'ed to 2.8kK, then they're not representative of human eyeballs (photopic), by definition.
The definition is the CIE standard observer function: http://en.wikipedia.org/wiki/File:CIE_1931_XYZ_Color_Matching_Functions.svg
Interestingly, if you look at the cree CW output curves, they match this quite well (peaks at 450 and 500-600). IOW they're more efficient because they're designed to be.