Er... several hundred feet? 216 feet (66 m) is the deepest you can safely go on normoxic air without incurring possibly fatal symptoms of Oxygen Toxicity; but an unacclimatised diver is almost certain to suffer from Nitrogen Narcosis at half that depth or less, increasing in severity as he descends. You will certainly need to have trained as a technical diver with qualifications in diving on mixed gases (trimix) to go any deeper.
By the wording of your question, you do not sound like an experienced diver, and I would strongly caution you against attempting deep diving below 100 feet (30 m) until you have gained sufficient experience, knowledge and training to attempt it.
However, that is not directly relevant to your question about lighting. I'll move your thread to the Dive Lighting section, where you will see other threads that may be of interest to you.
there are a lot of good answers to your question but not all answers may suite your needs.
Can you give more information about for what kind of dives you want to use this light?
Also, there are quite different forms of dive lights. is there anything that you had in mind?
there exist small or bigger handheld torches and light heads with a separate battery cannister connected with a cable so you don't have the weight of the batteries in you hand and can increase runtime due to larger battery packs.
depending on in what kind of waters you dive, you may want a narrow or a wider beam.
do you plan to buy a torch or build your own?
perhaps you can have a look on the websites of some dive light manufacturers like green-force, metal-sub, mbsub, tilly-tec, halcyon, underwater kinetics, dive rite and many others just to get an idea of what exists on the market and compare this with what you have in mind.
I'm going to guess that you are a new diver, or are about to be, and you want to buy a good light that will serve you well in the years and dives to come. As DM51 warned, PLEASE do not dive like that without proper training.
Unfortunately, there's not a simple answer to your question. There are hundreds of different lights from over a dozen different manufacturers, all made specifically for diving. I carry four lights on a dive trip, and select from among them based on the dive I'm doing. But you don't have to start there. A single general purpose light can cover many different kinds of diving, and will last many years.
Here are some general pointers.
1. Don't buy anything but LED. HID, zenon, krypton, halogen, are dinosaurs not worthy of consideration except in very special circumstances.
2. Unless you expect to dive exclusively in clear water, get something with a tight beam. Shine it at a wall from six feet away. If the central bright spot is bigger than a dinner plate, it's not a tight beam. A really tight beam is the size of a tea saucer from 6 feet. For use exclusively in clear waters, you may want a wider beam.
3. A little sidespill is good for general purpose. This is the area of dim light outside the central spot. It can cover quite a wide range. Most lights have plenty of sidespill, but there are a few that don't.
4. Pay attention to batteries. If you travel, make sure the light will take the standards - 2/3A (aka 123A), AA, C, or D. If you dive a lot, rechargables make sense. In many cases rechargables can be used interchangably with non-rechargables, but check to be sure. In some lights the rechargable batteries come pre-assembled into a pack. This is fine, and is actually a little more reliable if the pack is well made. But those batteries will eventually fail, and will be expensive to replace. Also make sure you won't get caught needing to recharge (or replace) your only pack when it's time to dive!
5. Learn how to maintain it. Dive lights flood. It's a fact of life. But the likelihood can be greatly reduced by proper maintenance. Battery care is also part of maintenance, and is especially important for rechargables.
6. This one is controversial, and I'll catch a lot of flack for saying it, but I would stay away from lithium ion batteries for a dive light. They don't fail often unless provoked, but when they do, it's pretty spectacular. If it's in a sealed container tied to your body, it's even more so. By the way, flooding the light would be considered 'provoking a failure'.
7. If your light is critical to safe diving, carry a backup. It doesn't have to be as big or bright as your primary, it just has to get you out of trouble. Good quality and good mainenance are especially important if this is the case.
8. Variable focus is a cool feature, but you pay for it, either in higher cost or lower reliability, or maybe both. I don't think it's necessary for most dives, but others will disagree.
If you drop your torch and it's still glowing "several hundred feet" below you it's no good to you...because if you are tempted to retrieve it, chances are you won't make it back to the surface.
Stick to enginnering a torch for the depths you will be diving.
DIW has some excellent points, but I respectfully disagree with a couple of them.
HID's are not 'dinosaurs'... in fact 35W HID's are yet to be clearly outshone by current LED technology. Comparisons aside, a 35W HID can easily end up being your brightest torch with the tightest beam (as mine is). There seems to be a heavy bias towards LED on this blogsite for various sound DIY reasons; it's elegant, modular, innovative and cheaper... however for pure lumens output in a collimated (tight spotlight) throw I will stand by HID as still trumps. LED pumps out light in one direction across about 140 degrees, give or take. By comparison HID is a 360 degree glow-spot that can be gathered and collimated by a reflector more efficiently. Having said that, to assuage the LED disciples out there, I can think of plenty of other situations where LED is better than HID...plus I exclusively use LED as my main torch for day diving. It's all up to you and what you want to use it for.
I will also be the first of many to discount DIW's advice not to use Li-ions. Li-ion's are fine for torches. Lithium-based is the most energy-dense battery chemistry currently available to you. Not only that, some of the high power designs you see here won't practically work in a dive situation without them. In some designs Li-ion allows the torch to be so compact that you don't need a separate cable and canister, with all the advantages that implies. As for safety...I've flooded my share of Li-ion torches and the most spectacular thing that happens is my frustration at the end of the dive. As for the torch, all that happens is... they go out.
Thanks for your responses. As you can tell, I am not yet a diver, having gone on only one guided dive about five years ago. But it's something I'm considering getting into, and the prospect of building my own diving torch dovetails nicely with my other interests, as I'm an electrical engineer by trade. So I'm going to look at some commercial torch prices, see how they solved some problems (like waterproof switching, for example; having not yet looked at commercial torches I'm leaning towards an internal hall effect switch and external magnet), and then decide whether to buy a torch, make one, or both.
So thanks for this info. Don't worry, I won't be killing myself, as there's a dive shop in town that has complete courses for beginners like myself (Steele's here in Oakland).
Any dive light you get should be rated to at least 100 meters.
Sure most of us may never get to that depth but that is not the point. You want equipment that exceeds your requirements.
HID is a good choice for some needs. Greater than 35 watt then HID is king ( for efficiency). Its also easy to focus vs LED.
But a 35w HID puts out more light than most divers will ever need.
For me LED is what I choose and it exceeds my needs for all my dives.
Plus its more robust, cheaper and upgradeable.
Li Ion. I use nothing else. I dont use NiMh or Alkaline at all any more.
I will add to this that efficiency isnt really all that important unless you really do require long run times & small canisters.
a standard battery canister can take 2 rows of 7 18650 LI Ion cells at 2.5 amp hour each.
14 x 2.5a x 3.7 volt = 129.5 amp hour. That enough to drive a 35watt light for 3 hours at full output (3.7 hours on paper but probably not in practice).
Make the cannister 65mm longer and add 50% more capacity.
'Efficiency' is important if you want to build a light without a separate canister.
You might want to build a single container...housing the light, switch, drivers, and batteries. No cable.
The 'can' or canister design is a popular, traditional design. Notwithstanding its popularity it's pretty much compulsory if using dated technology. As lights have become more efficient (lumens/watt), and batteries have become more compact our designs are no longer mandated to the can design.
You can (pardon the pun) design either.
Plenty of criteria to consider; cost, brightness, beam angle, beam colour, bulk, weight, buoyancy, run-time, robustness, switch type, modes (HI, LOW etc, and if possible), likely depth rating. Plus, what is your skillset where machining, electrical work, lighting is...and what access you have to parts and machinery.
Some people put their heart and soul into this sort of thing, while others knock something together that reflects a different set of priorites.