Mechanically the prometheus just doesn't work in the desktop Amiga's. I mean come on:
http://www.relec.ch/prometheus/images/A4D01.jpgIt didn't mechanically work well in the towered Amiga's either.
http://www.relec.ch/prometheus/images/A4T01.jpghttp://www.relec.ch/prometheus/images/A3T01.jpgI am sure I read that the prometheus creator didn't particularly like the final design.
Matay probably knew they were only going to get one shot at this. High development and manufacture costs meant that the PCB they designed had to work mechanically in all Zorro III Amiga's so as to be compatible and hopefully lead to more sales. I think the big problem was that a lot of the A4000 desktop users had already towered their machines using a multitude of mechanically different cases
and different Zorro busboards. They had to come up with a mechanical design that would work for all models.
I feel that this decision was a compromise too far, the Prometheus design resulted in being mechanically poor in all cases. There were things Matay could have done (but I dont think they did) to mitigate the problems:
o Matay could have tried to sell the PCI riser card you demonstrate in your example A4000D. I am almost sure this was a user discovery and a homebrew mod.
o Matay could have designed and sold metal brackets to make it easier to fit PCI cards safely and sturdily. I dont think that ever happened.
It was the right product (technologically speaking) and right time. Just a poor roll out and too many corners cut to lower costs. Matay should have targetted one case (probably official commodore A4000D) and created the best possible mechanical design. Then use the profits from those sales to pay for PCB reworks (and metal work, replacement back panels etc) for other cases.
This (to me) is just common sense... there must be more to the story we dont know... I wonder what it is?? Anyone?