They use the same kind of drivers so there would be no difference in brightness, it just depends on the level of driver you choose.
yes the xiaozhi was the one i was talking about. i thought the xiaozhi was running at 2.5amps?
so you prefer the s3 over the xiaozhi because of the better tint. could you tell which one is brighter and floodier? i only want one more tube light and cant buy both.
From a dealer website:
Xiaozhi:
C20:
I’m debating with the same “how many AMC7135’s” question for Convoy S2 or S6.
Can anyone tell if there’s any notable efficiency (or other) differences when running 8xAMC7135 at 50% vs 4xAMC7135 at 100% mode?
If I’m not mistaken, only the high level (100%) is non PWM, other levels are.
Since PWM would reduce runtime then,
4xAMC7135 at 100% would be longer runtime vs. 8xAMC7135 at 50% (PWM).
The 8x drivers will give you over an hour continuous on high mode, if you have gloves that let you hold the thing that long (won't hurt the light, might hurt your hand). Are you worried you won't have the self-control to switch the thing to a lower mode when you need to?
With the 2.8A driver if you need it dimmer/longer runtime it's easy to switch to a lower mode. With the 1.4A driver if you need it brighter, you... well, you can't. Unless you can solder in the dark.
Basic non-disco, 3 modes, AMC7135*5 in my S2, with XM-L T6 3C 
No, I’m worried that the 8x light would be used 99% of the time on the 50% mode, and I’d be better off with 4x model instead. Surely these will overcook the LED, if the max mode is used for long periods of time on 8x. 
Having access to a really high mode would be cool, but is it worth it if the “main” mode that gets used most is less efficient?
I like having the 2.8a option but keep it on memory low.
tbh, I’ll be switching many of my 2.8a drivers to 3a qlites in the fullness of time, just to gain the moonlight mode.
my regular edc tube uses a different current controlled driver and will keep that, no pwm is good around moving machinery.
As boaz said too, forget the cool tints, go with 4c or 5c, you will find it much nicer to use. And if you want to gun a smallish light at 2.8a, take a look at the roche f8, the cnqg d5 is the f8 analogue host kit and I carry and use it every day. It is a very very nice light, fits fine, to a point of not noticing it in the screwdriver pocket of my contractors pants and handles 3.4a fine.
This will be getting both a qlite with stacked chips and a high cri xm-l2 as soon as my order comes in since it is without question my most used light.
Nah, I torture test all mine to the point of silliness before they're put into use, by dropping in a freshly charged cell, switching it on high, and tailstanding it until it starts blinking a low voltage warning. Yes it gets hot but it has to survive doing that before I'll consider it ready for use, not one of them has had a failure yet.
But, I use copper MCPCBs in every one, and use 22 AWG wire, and thermal grease on the pill threads, etc. A pre-built one might not be as robust, but I haven't used one like that personally so I can't say for sure. That doesn't mean they're likely to fail, just that I can't say with certainty. I'd have to try it and see it happen, and try enough of them to be sure I was getting an accurate sample size. But in general, this size light isn't automatically unsuited to those kind of drive currents.
People build & use 3 amp P60 lights all the time, and though they have more mass the thermal path isn't as good as the tube lights. The overall mass doesn't matter once it's heated up to a max stable temp, what matters is the outer surface area where the heat is transferred to the air, and a P60 only has, what, 8-12% more surface area than a little tube light? The max temp at the end of a torture test will be very nearly the same for both designs.
I understand your concerns about runtime & efficiency, but really, over an hour straight on high mode, over 2 hours and probably approaching 2.5 hours on medium, with a 3 amp version... that's a long time for so much light. If you need more than that carry a spare cell just in case.
The 2.1A version would be the one to get IMO. I have the 2.8A version and it works well but I tend to use medium as High heats up the torch too quickly which is probably reducing it’s efficiency. 2.1A would be a nice sweet spot for the torch as it would give you nice levels of light and not warm up.
I just ordered a 2.1A S2 after owning a 2.8A Xiaozhi. I was looking for a lower medium, and feel the difference in output from 2.1A to 2.8A is neglible. Should arrive any day now.
-Garry
Thanks ComfyC, I’ve got a 2.8A S3 and S5 headed this way. They will get lots of air on the bike, good to know you torture’em for us in advance so we don’t have to worry!
For sure. I have a number of hosts I built with that configuration and they definitely get too hot to use for a prolonged time. But I still find them very useful for many situations where I need maximum light for short periods of time. So I think the best situation is have a number of lights to address a number of situations, maybe have a 2.8 and a 2.1 for example.