Sorry if I'm stirring up a topic which has kind of died on this thread, but I wanted to take a shot at the great DIR vs. Computers debate. Keep in mind that this is all my $.02 as an OW diver with exactly 4 dives completed. And in the interests of full disclosure I've already bought myself a reg setup with DIR hoses from fifthd and I'm gradually being swayed over to the dark side of the force -- OTOH, I bought myself a Suunto Vyper from LP as well...
1. Deco/tech diving approximation methods
I just happened to follow a link right before I read through this whole thread to a page which describes some methods of approximating decompression profiles, very similar to the "rule of 120":
http://thedecostop.com/forums/showthread.php?threadid=3198
There's a few more rules, but none of them require you to try to do RGBM calculations in your head. I'm sure that with enough experience with producing decompression profiles from a program like d-plan that you can come up with similar approximation rules for any way that you'd like to modify a dive profile -- even for pretty advanced tech diving.
Really this stuff isn't like trying to plan how to launch a rocket to the moon in your head. There actually is a lot of room for error on the conservative side of the decompression plan which you can use to your advantage.
2. The usefulness of approxmation methods
One thing that my undergrad studies in physics and astronomy taught me was that approximation methods were essential. The first thing that you always did when presented with a complex problem was to simplify it by using some kind of approximation method to figure out roughtly what the answer should be. A typical method would be to treat the problem as being perfectly linear. Often you could find a lower or upper bound on a solution. All these techniques were invaluable for catching the case where you'd dropped a minus sign somewhere in your arithmetic and wound up with an answer which was a few orders of magnitude off of your first approximation.
Similarly in scuba diving I could see how approximation methods could be essential for tech diving. If you've only run your plan and a couple of backups that you take down on slates, you are more limited than the person who can adapt their plan on the fly to deal with any contigency that arises. Those contigency plans using approximations may not be ideal, but iif they are conservative approximations they will prevent you from getting bent, which is the only goal which really matters.
3. The desirability of intuition
Another thing which I took out of my physics and astronomy courses is that it is good to have an intuition about what your answer should be. Without even doing any kind of approximation calculation you should be able to know what your answer is going to be to roughly the nearest order of magnitude. This is a skill that comes with practice and experience and is probably similar to the skill that chess players have where they can look at a chess board and tell if it was the product of a legal game or not.
In scuba diving we all probably have a little bit of this ability already. If I present you with a no deco dive on air down to 120 feet with 60 minutes of bottom time a little bell should be going off in your head telling you that something is wrong. I think a strong argument can be made that honing this ability is important.
4. The lack of any existing deco computers
I think its been pointed out by a lot of others that there are no wet compuers which will do the kind of computations you need for the demanding tech diving that the GUE folks do. Since we're talking about the GUE opinion on computers it makes sense to talk about why they've got that impression based on their experience. I'll address my experience later, but as far as GUE goes there simply isn't anything which works for them.
5. The difficulty of decent UI design
I already find the UI of the Suunto Vyper to be a little bit cluttered. I know it is going to take me some time to get used to it. I appreciate, however, the skill in the UI design which was required to get all the information that the Vyper has to offer into that package in a way which minimizes task loaded underwater and gives you the information that you need to know.
I can't imagine trying to design a UI for a tech diving computer, particularly one which could respond seamlessly to any change in plan which could occur underwater. I'm sure that it is technically pretty easy to take GUEs d-plan software and cram it into a watch-sized pacakge. That isn't the biggest problem though. The biggest problem is that if some emergency has come up at 200 feet you don't want to be pushing buttons for 5 minutes on your computer to try to modify your plan. If anyone is going to come up with a computer like this, the major challenge is going to be in the UI design, not in the bubble model that it uses.
6. Reliance on computers is bad
This is true, and an argument which is beaten to death. So far I've been making arguments slanted somewhat against computers, and here's where I actually want to make an argument in defense of computers.
Just because you have a computer it doesn't mean that you're reliant on it. All of my physics and astronomy classes, along with my job taking care of a huge network of computers, has taught me not to trust computers. It has taught me about the desirability of intuition and approximation methods, and the desirability of understanding what the computer is doing and how it could be going wrong.
I think there may be a belief, and certainly there is a conveyed impression, that computers actually cause people to become reliant upon them and that isn't the case. The people have to choose to become reliant upon the computer.
7. My own thoughts
i'm not unhappy at all about buying a $300 "battery box". Since I'm inexperienced I appreciate having the computer there giving me indications of what my ascent rate is, or what my P02 is, and having alarms for if I go too fast or go through my NDL. And if I get into an emergency, while it'd be better to understand deco diving I'm definitely going to appreciate having the "bend-o-matic" telling me tthat I'm probably in trouble and what I should do about it. Yeah, knowing all that stuff intuitively would be much better, but all the studying in the world isn't going to make me an experienced diver right at the start.
it also gives me a check against making bad mistakes. I could drop an hour of a surface interval and miscompute my residual nitrogen and wind up doing a repetitive dive that gets me bent. A computer can help since if I wind up coming up with an answer which radically disagrees with the computer I know that I had better search for why the discrepancy happened.
A compuer also doesn't get nitrogen narcosis. Of course if you are narced then your computer also gets radically less useful. This is another argument in favor of very simple UI design in computers and why creating a UI for a tech computer would be so demanding.
So, I'm happy having bought a computer. I actually do trust it at this point better than i trust my own brain. I can certainly understand why the GUE guys would trust their own brains better than any dive computer though. One day i hope I'm there with them.