Yet another excellent reply by the OP which says nothing. No, it's NOT OK. It's ignorant. My guess is that feeding the fire is what scratchr really seeks here. Time to remove the fuel. :-)

 Or slideshows, to show off stuff you've made.

Ok, and the current limiting resistor is where again? High power LED or not, excessive current firstly, kills any chance of having usable light for usable time, and second, will instantly degrade, or kill any LED regardless of how "high power" it is if the junction temperature exceeds max rating for ANY amount of time.

For the record, I am an EE and design circuits which are reliable for long periods of time using proper design principles.

This is not an instructable. It's an experiment from someone with little electronic knowledge. Not trying to be mean, but connecting an LED w/o current limit then justifying it by saying "I used a high power LED" just showcases the ignorance. The author's curt, non-replies further highlight this point.

What you have here is a slow-speed flash of sorts. Really bright for an instant, then with no charge left, a declining dimming light source.

This also exceeds the current ratings and probably the voltage rating of the supercap as there is no voltage regulation used. Also damaging and potentially explosive.

Sorry.

This I suppose should be flagged as unacceptable as personally I strongly advise nobody to build this and expect much from it. If you have trash parts to play with, then go ahead and play. But "rechargeable flashlight" this isn't!

Adding current (LED and Supercap charge) and voltage limiting (Supercap charge) will easily and cheaply add a lot of potential.

And NOWHERE have you mentioned the critical parts omitted: current and power ratings.

Voltage alone means squat here ESPECIALY when dealing with non-linear semiconductors like LEDs which JFYI, have a NEGATIVE TC. It's assumed that's foreign to you but is really one more major nail in this experiment. It's called positive feedback and VERY few cases, and I've used PF in exactly VERY FEW CASES, is a very powerful technique. Passing totaly unregulated current through diodes, limited strictly by loop resistances does not qualify by any means, "regulated" and not even "limited" an any sort of sense which will ensure any form of lifespan of the devices.

Your max ratings while "within limits" take nothing else into consideration and as such, are meaningless by themselves.

"Tiny not efficient" in this case boils down to tiny, inefficient, and deadly for the components used. I'll give you the first two.

Bottom line: You get an A for experimental effort, which many of us begin with. But sorry, you get an F for publishing it as a misleading project which, if you read the comments, many other people would love to build based soley upon your (sorry) uneducated claims.

The good thing is that nothing here is dangerous enough to really cause harm, except perhaps to one's cash...

Again, not trying to be mean. But electronics relies ona lot more than "well the voltage is less than the max so it's OK to do". Hardly. Three fundamnetal design quantities: voltage, current, and perhaps most importantly, power dissipation. Cover those, and you'll be way ahead.

Have fun, be safe.

you should have pictures of each of the building stages and a schematic, not just pics of the finished product

I'm going to keep an eye out for your next article on this - it looks interesting so it'll be good to see the details. 

you did a great job im sorry to see that so many people are being jerks