3

Without actually testing the light, I’ve found a quick way to ballpark the current drawn by a light is to take a look at the runtime figures given by a particular cell.
You can usually find this stuff written up on a sheet if you have the light, or included in some reviews. (This doesn’t help as much if a manufacturer falsifies their data, but it’s better than nothing.)
Then, you divide the capacity of the cell they tested with by the runtime to get some kind of current value.

For example, the Thorfire VG15S tested with Thorfire’s 3000mAh cell has stats of:
table(table#posts).
|Level|Runtime|Estimated Current (3000mAh/Runtime)|
|Firefly|360h|0.0083A|
|Low|50h|0.06A|
|Mid|2h45m|1.091A|
|High|1h25m|2.118A|
|Turbo|45m|4A|

Cheaper cells aren’t going to give quite as much capacity at higher current draw, which is probably why Turbo works out to 4A (I think I’ve heard more of a 3A value).
But it can be pretty on-target if the testing cell can handle the current, like the Sofirn SP33 stats: runtime of 52m on Turbo with their 5000mAh-5500mAh cell, working out to about 5.77A-6.35A. (Which is close enough to the 6A draw I’ve heard quoted for it before.)