@ amiga-3k:
If the price is right, a MiniITX sized Amiga would appeal to me, even if it were just and only OCS/ECS Amiga (Minimig). A littlebit more advanced (Jens' Clone-A MiniITX) would appeal more to me. I can surely do with some space-saving on my desk :-)
Why is is that everyone *THINKS* that Herr Schoenfeldt's A500 implementation is more advanced than MiniMig when this is absolutely not true? I have this itching feeling that Herr Schoenfeldt is the type of salesman that would be able to sell you sand in the dessert. He is great at presenting his products, I must admit.
@Hans:
Actually, Dennis' approach has been to replicate the A500 in a behavioural sense based on reference manuals, software emulator code and observing behaviour of the actual hardware. AFAIK he didn't specifically divide it up into the different chips. Hence, it's not guaranteed that you could replace say the A500 Agnus with a Minimig Agnus equivalent. He didn't have to ensure that the timing of all signals connected to/from Agnus matched the timing of the original chip. The Minimig may be cycle-correct in its overall behaviour, but this doesn't mean that the internal signals have the same timing as the original A500. You'd have to reverse-engineer the signal timings for the individual chip and modify the design to take this into account.
Hans,
It doesn't matter to me how certain you think you are about how Dennis went about cloning the A500 and creating MiniMig. What does actually really matter to me is that you have your facts all mixed up. Dennis has in fact been cloning the A500 chip-by-chip. If you do not believe me, please feel free to read the 850+ posts in the "Amiga in an FPGA: MiniMig" forum topic OR even contact Dennis on the subject.
As for the accuracy of the actual signal timings; I agree that MiniMig's synthesized A500 chip set might be a little off compared to the real thing. This is no big deal in the sense that it is something that can be fixed in a real short window of time. Although I my self have not done any in-depth FPGA programming as of yet, I am certain that it is something that even I could FIX in a matter of hours with a timing diagram of the original A500 handy.
Oh, why is it that the signal timing has to be absolutely correct again? Aside from the bugs, MiniMig runs everything A500, doesn't it? It is a true A500 clone after all! Did I actually hear you say "drop-in chip replacements"? Have you any idea what something like that would cost if production is not on an impressive scale? Wouldn't you rather spend your money on a reasonably priced modern implementation of the A500 than on a relatively expensive chip in the hope that none of the other chips need replacing? I know I would.
Minimig works, although he's not done with bug-fixing yet.
Jens Schoenfeldt isn't even done with his chip set yet and he has three programmers and a logic analyzer. Ha-Ha. Besides, I am sure that once he too has quite some bugfixing to do. What a joy.
jen-ss (Sander)