I think there will be three categories.
As an example: A light currently has a 5k turbo and 1k sustained.
Fancy cooling is added and it will become 5k turbo and 3k sustained.
No matter how fancy the cooling the heat has to go somewhere. There are limits to that.
(all numbers are only for explanation)
I think the light will get pretty big and sensative to drops if you use cpu fans etc. And boost drivers are needed if you want single cell xhp35 lights.
When do you need a 20000 lumen turbo light for more then 45 seconds?
There is a place for both lights. Especially for smaller lights where size and weight is important (EDC) the turbo feature is a nice thing, because otherwise you would not have that much output (for a short while), because its just physically not possible.
I think unsustainable turbo modes have a place, especially on carry lights. Walking around with my D4, I have several hours of regulated output which is plenty bright for most tasks. In fact, I used my D4 for hours last weekend while working on my car. But it's also nice that I can have that momentary 5,000 lumen burst when I want it. It's good for when the creepy dude pops out from behind the gas station dumpster asking for money and scares you, after 10pm. Suddenly he's blind!
i think it is fine for them to have momentary turbo…
just tell folks the parameters
i do not want the size, unreliability or cost of fans, liquid cooling, radiators, high resolution temperature graphs on a screen - any of that
i’m fine with ‘it gets hot fast and might step down after a while’
(colder ambient temps mean longer turbo - unless the battery gets cold too 
)
wle
Even olight or acebeam ditched the idea to make their biggest lights with built in fans for greater runtime even tho the prototypes had them, its not the easiest thing do a massive light and just throw in a few fans i dont know what it takes to give us better runtimes max? the hosts must get alot bigger then those we have now…
I love having turbo modes available - any cooling system which can sustain X lumens can also handle anywhere between 2X and 10X lumens for a while until it heats up. This is even true of a water cooler and radiator setup. Dumping a bunch of extra power until the cooling system warms up is basically free light when I just need a short burst to look into the distance, and I don’t have to use those modes if I don’t want to. However, I do wish marketing made clear what modes were actually sustainable and emphasized those modes more heavily. That is a failing of FL1 more than anything.
In the particular scenarios described in the OP, though… Maybe you should look at reviews before spending $600 on flashlights that don’t perform the way you hoped. I can’t think of many industries or products where you can just take the manufacturer’s claims at face value without independent verification. There’s a reason runtime graphs are part of a good flashlight review.
That looks like another expencive hobby.
If weight isnt a consern you can use a light with bigger mass. I dont know how hot the blf gt70 gets but my normal gt stays cold on turbo.
wow. knowing how hard this is this dude is unbelievable, especially at night and such a long run
You can get what you want, but you’ll carry a 10 pound light in one hand and a briefcase-sized battery unit in another.
I don’t understand the need for max power with unlimited run time - unless “unlimited” means for the duration of the battery charge, which won’t be a lot running on max. Just like high performance sports cars. A Porsche GT2 RS does 670BHP at 7000rpm. But no one (except Indy500 and F1 racers) run their engines up at the redline constantly - you’ll kill your engine. Plus it’s not necessarily the most efficient way to drive the car. We start and stop, speed up and slow down, and for me, my flashlight use is just that. Crank it up when I need to, then drop it down to a sufficient level. Same thing when I drive - I only crank up the high bean when necessary, and it’s almost never necessary to keep them on all the time.
For max power with long run times, a portable scene light would be great, but they’re not necessarily convenient for handheld use for long periods.
Just give me direct access to the highest sustainable level, rather than making me wait for turbo to step down to get to the level that’s sustainable for an hour.
This just makes sense to me, but most UIs don’t allow for it.
Regarding turbo mode:
A small pocket light like the Emisar D4 might run sustained around 500 lumens, but can turbo for 4300 lumens.
Sure I could just get a light that can sustain its max output of 500 lumens. But in my opinion, that would be considerably less useful than also having available a turbo mode.
The D4 can already run at 500 lumens for an extended period so I wouldn’t really be gaining anything by buying a light whose peak output is a sustained 500 lumens. All that would happen is I would lose the very useful added functionality of being able to burst extremely high lumens for brief periods.
Regarding active cooling:
I admit, I would like to have active cooling in my lights to provide higher sustained output. A small fan would be pretty nice. The problems:
This is one reason I like Anduril UI over the default UI in the Emisar D4.
In the D4, double-click takes you to turbo which then rapidly ramps down. With Anduril, you can set double-click to any level you want. So the highest sustained level or maybe the highest level that won’t burn your pocket. You can still get to turbo with a double-click when the light is on.
Still hoping to get Anduril onto my D4 lights. Hopefully this week if I can get the software and hardware to work.
It’s good to know that Anduril allows that.
You know, in high school, I learned that in weight training, there was such a thing as “reps” and such a thing as “maxing out” and that those were two totally different things. Why is it hard for people to understand that “unsustainable” Turbo mode is useful and desired by many, while there are usually PLENTY of lower modes you can use if you want a “sustained” output level? Just turn the light down to a lower level! How hard is that?! 
Did you read this thread?
If you explain it using “Cars” I will understand better.
CARS DavidEF
I built a Convoy C8 with an XHP 70.2, DD driver, and running two 18350 batteries. I would say that of the 5 brightness levels, I use mostly 1 and 2. I rarely use 5 because it’s actually too bright for near-range sight, but it does illuminate a wide view like daylight, and I don’t need to lug a 4 pound light with me. Of course it will get hot fast, but not long enough before I drop the brightness down. I probably could have just dumbed down the output or used a smaller emitter and get sustained output for many hours, but I also like to have that “turbo” mode available when necessary. And this thing is so blight it likely would temporarily blind a would-be assailant during an evening walk. At least burn his face - the beam on high is incredibly hot!
Brushless motors can be waterproof; you can even fly a drone underwater if you seal off all the other non-waterproof components. I’d agree there probably is or will be better passive solutions available. It’s very easy to drive an LED to spec and have a flashlight not even get warm during operation on the highest setting, but that seems to contradict an ethos of BLF, of getting the best bang-for-the-buck lumens wise. I know the opinion of a few is that max output not sustainable for an extended period of time is useless. But running under max output for however long the batteries will go until they die, which is not long, in my view is equally useless. Not like the old days when you’d get a flashlight with incandescent bulbs and alkaline batteries, and even if the batteries are almost drained, they still put out some light - which can be pretty important depending on what you do with your light.
…okay, so imagine you own a Formula One racer, a Toyota Camry, and a Fiat 500… 