I think there a host of good idea's. I think Odin should (once he figures out the ps3 insides and gets an idea of possibilities) make two lines of ps3. He mentioned he wanted a low cost one, so maybe a budget controller and another controller (with a higher cost) which offers customization / more buttons (such as sticks available for macro's) / more tac switches or something like that. The Raven and Thor's Hammer (I think it was) was a good example of this. I'm not very tech savy but I think this would be a good idea. This would probably as you will reach a wide market if you offer a cheap ps3 controller with (better than ever other ps3 mod I've ever seen) features, at a decent cost, while you could say build each order individually for a more advanced controller aimed at more "Hardcore" ps3 gamers (at first at-least).
Now onto actual controller idea's:
I say two tac switches is enough, maybe three one for a page switch of sorts or a mode on/off switch. A modification program with something similar to the 360 macro library would also be good as you can get the community involved in the macro making instead of you making a good amount of pre-sets. This would also reduce the need for post-release support as people can just search/make your own macro's.
My final idea is quite a good one. From reading your specs (of the new viking controller) the chip hardware is the major cost with you even saying "It costs 4X as much as the older chip". With the ps3 controller, what I like doing (since my slim only has 2 usb slots and I usually have a headset and my phone or something charging [yes this is my inspiration]) is have my controller's usb plugged into my pc while I use the ps3 so it charges and I play. You could use a basic chip (I am clueless as to how it would work but it might be cost effective?) and have a program shipped with it or available for download through the website. This program does all the macro's and stores all information (*cough*does all the processes*cough*) on your own PC, so you won't have to buy an expensive chip with a good processor and memory for macro's. This has several advantages (hypothetical?). I see it being cheaper, easier to update/edit macro's on the fly, and having less limits. This is obviously negative in some aspects, such as it's not very portable, so you can't take it to your friends house.
What I see as best?
You could offer two controllers, one (cheaper chip) which needs to be plugged into a pc and uses that. And one portable controller (obviously more expensive) which can store and run independently of your controller (and the inside chip). This solves the "price-tag" issue, but doesn't offer you (much) extra work as it will basically just be the exact same software. It will probably not be done this-way but (at least) in my head this seems cool.
