You're not just a PC with a PowerPC CPU and some VGA chip trying to make the claim of being an Amiga.
I understand that PowerPC has become very popular in networking infrastructures, at least partially because network data is big-endian and x86 would spend a lot of time byteswapping rather than processing and moving data along.
I understand some reasons that PowerPC is no longer ideal for a desktop or other personal computer general purpose machine, such as the price we'll pay for such unique and small market boards.
But I'd like to know some reasons why it is technically insufficient. I can't find the right post to quote, but I remember seeing Dave saying this somewhere. It's been over a decade since my undergrad intro to computer architecture courses, and I've not yet got to them in my Masters program, to have a really good understanding of what is good about x86 (and maybe ARM now too) that is lacking in PowerPC, particularly with QorIQ AMP e6500 cores out now.
The things many of us end users look at is speed (e6500 claims 2.5GHz, below some current, soon to come dekstop x86_64 things nearing 4GHz), memory (AMP has DDR3 controllers like AMD Bulldozer/FX does), e6500 virtual cores sound similar to my understanding of Bulldozer/FX "module", PowerPC is now multicore like x86. Altivec has returned for comparison to SIMD, though I don't know how the "update" compares to latest SIMD or what that comparison means to us. We should be able to use an AMD SB950 southbridge (which as I understand is an SB850 with different writing on the lid), as it should not be any different than hooking up an SB600 as seen on X1000. We should be able to add USB3 using the same PCI-Express chips as the PCs currently use, until it's added to southbridges.
Yes, this is comparing the not available yet AMP series to x86, at least somewhat to also not yet or just very recently released Bulldozer chips, and perhaps this is different than comparing PA Semi's chip, or the 8641/8640 or other PowerQuicc chips to x86.
Ignoring that we have to spread engineering etc. NRE costs over a thousand or so boards compared to x86 spreading same costs over hundreds of thousands of units, and other market size details tha make it difficult to get a PowerPC board onto our desks (I wish onto our laps!) What are technical things we should hope for in PowerPC if we must be restricted to that? Ignoring the market size benefit to x86, what technical features there are superior to our equivalents, which ones to we lack completely, etc.
I don't disagree that x86_64 should be considered by the AmigaOS powers that be today, I just would like to have a very technical understanding of why that is, aside from the marketing issues that are more obvious.