Regarding underfilling the HP 100's to put less stress on the reg, I meant less stress relative to filling the HP 100 to the full 3,500 psi. Obviously 3,100 psi is going to be more stress on the reg than 2,800 psi in the overfilled LP tank. I'm just looking for a good compromise between maximum air volume and least amount of headaches with things like difficulty with getting the fill pressure you want (I've heard it can be difficult to find a place that will give you a full 3,500 psi), least potential for equipment problems, etc., etc.
In reading one of the other threads a consideration I forgot to include is trim. I'm 5'-11". It seems many people my height prefer the longer tanks in this respect, but then they probably aren't diving a short DR Trek wing either.
I found a lot of good information about tank configurations on the New Jersey Scuba Diver website. While there seems to be a lot of 'opinions' regarding SCUBA gear setups as everyone seems to want to preach their own mantras and personal preferences, this article seems to be a fairly level-headed piece of information.
http://www.njscuba.net/gear/eqpt_05_tanks.html It acknowledges DIR guidelines but also points out the most important thing is that tank configuration really has to do with the type of diving.
With me, some days I may have a long walk for a shore dive and may not be going below 60 feet so I just want to dive a single tank other times I may be doing a boat dive in the San Juans and may be diving a lot deeper and don't want to be air limited as far as the length of my dive. I know optimally I should have a wing for singles, a wing for doubles, a doubles setup with manifold and a couple other tanks for singles, but I don't dive often enough to justify the cost of all that.
The statement that really hit home with me on the New Jersey Scuba site was "What I am recommending is the ability to flexibly configure your equipment for whatever type of dive you are doing right now - whether it be deep, shallow, long, short, boat, shore, currents, waves, etc. The discussion of temporary doubles bands is motivated by the fact that manifolded doubles are anything but flexible ( and are also ridiculously expensive ) and that it is relatively rare to actually need double tanks for a particular dive. "
In most cases I'm either going to be diving a single or when I want more bottom time, I'm going to be diving doubles and I like the one-and-a-half scenario where I'll have a third tank I'll swap in for my second dive. I really like the flexibility of having independent doubles instead of using the manifold although I know the manifold configuration is more popular with most people here. I'm not doing wreck penetration dives or cave dives or tri-mix or anything that technical. Mostly rec dives with an ocassional deep dive. Plus with the independent setup, if I don't want to spend a bunch of money on tanks to begin with, I could just go rent some ALU 80's and put up with the less optimal boyancy characteristics.
I guess I've spawned this discussion into two issues, best tanks for my applications, and manifold versus independent but I purposely didn't bring the manifold vs independent issue up initially as I figured it's one of those subjects that has been beaten to death here many times before.