You dont know anything about BOM prices or PCB design, so using phrases like "next to nothing" is just insulting to those who have been trying so hard to bring MiniMig to the masses.
Lets see, I have sourced, ordered and paid for all the components on the MiniMig and am well aware of the unit costs. I was not for one second trying to say that everybody else is doing anything wrong, I was and am talking about a mass produced simplified design, as a reason for which the MiniMig is not a waste of time.
If what you said was true, everyone would be doing it, they are not, so you are WRONG!
This is a young project and people are heading towards the direction of reducing the component count of the board, as another person has already mentioned, someone has already put the 68k into the FPGA as an IP core. The thing is, you probably know as well as I do that what would cost "next to nothing" in mass production would cost the earth or be practically impossible for the numbers we're talking about or, even worse, hand-soldered single units. So, I am not exactly disagreeing with you here, I just do not believe that the fact that the MiniMig is currently made in single units and is expensive means that it's useless or that repairing A500 boards is a better idea.
The main reason no-one has used a BGA chip is because they are impossible to solder without expensive BGA IR flow equipment that no "homebrew" team will ever have.
Yes, but I was talking about a cheaper, smaller and simpler version. You aren't exactly going to order 200 boards and parts and hand solder them, are you? Obviously, you wouldn't wanna use BGA even for 200, but my point was, there is nothing on the MiniMig that is prohibitively expensive and the design has potential for significant cost saving for, say, an Amiga-in-a-Joystick.
Anyway, the reason I mentioned BGA was not because I had a brainfart and wanted to throw a buzzword. I'm well aware that it would require a more complex PCB. The reason I mentioned BGA was that all 1M-gate Spartan 3 chips are BGA. The thing is, as more hardware would be integrated on the FPGA, the larger number of I/O pins wouldn't really be needed. The ideal candidate for something like that in the Spartan series would be a 3E, which is biased towards more logic on package versus I/O pins but the largest 3E is 500K gates, which I think is not enough for a 68k IP core plus the MiniMig chipset. Depending on the number of pins available versus the number of pins used if the 68k and perhaps the controller gets moved onto the FPGA, you may not have to use a very significant number of BGA IO pins and get away with using a 2 layer board but I may well be very very wrong, I have not checked the Spartan 3 BGA pin-out to see what fan-outs could be used.
Before you start cost reducing something that doesnt need cost reducing and redesigning a PCB (something almost certainly beyond your capability)
And what makes you think that that's almost certainly beyond my capability? I have designed PCBs and the MiniMig PCB is not exactly the most complex design in the world. I probably would not be the right person to spearhead such an effort, I'm not that experienced with PCB design but again, I'm talking about possibilities, not how "I'd do it better".
lets try to get some of the MiniMig v1.1 boards made en mass and price savings that way!
I fail to see your point here. You were complaining about the fact that it's expensive and saying that you could fix A500s for less, right? I basically gave you reasons for which working on the MiniMig *is* a good idea and have been talking about the possibilities.
Again, my point is, mass production brings costs down but when mass producing, simplifying the design can bring some quite major savings.
If we could find a company/individual ready to stump up the time to talk to the manufacturer and the £2000 or so required to get a batch of 200+ units made we could reduce the costs to the end customer. The profits from this could be invested to make more boards and future cost downs.
That I agree on 100%.
Apologies if I sounded like a smart-ass.Edit: Added a few things as I went along.
