Right, after prolonged interest from people here in NZ it'd appear that I'm doing this - I've also got a grudge to settle!
I'm thinking of a simple 2 PIF setup as demonstrated here:
Ich bin Michael Schultz aus Werneck und erzähle euch auf dieser Webseite ein paar Geschichten aus meinem Leben.
www.retrovideogames.de
This has a lot to do with the costs of components here in New Zealand, as well as the price and availability of N64s that have succumb to board-rot or damage over the years. Generally only the main traces or power components have failed, and the PIC chips themselves seem pretty bulletproof. So that's the concept at least.
So a friend of mine sent me down his N64 that had the region free mod from Otakus in it with the message "heard what you were doing, will this help?" and if anything it's raised more questions than I had expected. For instance they appear to be muxing the entire "left hand side" (which is on the right hand side in this photo) of the PIC chip, and have employed a very obsificated method of routing the board - which suggests that what's going on here is actually really simple, or that the region switch process requires feeding different lines to the PIF as it's preformed. I'd guess it's both actually.
They've also got some very clever solutions that I'd not have thought of, such as using IDE connectors for the wires, and a status LED to show the user what's going on.
I finally got the time to look at the lines going into the PIF using my logic analyser and found that when it's the matching region there's constant chatter between the PIF and the CIC, and when it's not the lines go low and there's no chatter - which makes checking for the correct region very easy: simply look for chatter, and if there's none it needs to use the other region chip.
This is where I could do with some assistance! The Otakus solution has scraped all the chips, so if anyone would be willing to guess what the 4 smaller chips (being used for the mux) are I'd be interested to know - they're being used to switch between both the data cannels between the two chips as well as controlling the power to the chips themselves. I strongly suspect that the chips don't draw much current, and they've got a 3.3v power supply and logic. I'm mainly asking this because the MC14551B isn't available for assembly with JLC and anything I can do to make my life easier is strongly valued. I'd also be wise to find an alternative to the MC14551B, but at worst I can just buy some from China.
And the plan from here out? I need to develop a prototype and testing PCB that's setup in such a way that I can mount the components and easily test things until I get it right. While I could sit down and make myself a mess of wires I'm getting to the point where I'm really not all that capable of doing things like that anymore, which is fine, you just have to throw some time and money at it to resolve the problem.