Gilles solution will be OK for testing . . if. . we can find reliable ways to
BREAK any existing firmware. . . so that we can use the
Apple Firmware Restore CD . . . in case something goes badly.
A bit of a convoluted, time consuming process and something that I would not attempt as I need my cMP for daily work.
Someone with
TWO 4,1 or 5,1 cMP's might be OK with it. "
Codejingle" comes to mind, he seems to have ba##s of Titanium . . . plus an 'insatiable hunger to know'.
Actually. Gilles appears to work for an authorised Apple dealer which probably has a few 4,1/ 5,1 cMPs lying around to play with.
That is the ideal testing environment. I'm jealous.
I have a strong feeling that native NVMe m.2 /SATA booting will become a reality in the near future.
Perhaps after High Sierra gets to GM stage.
With 26,000 views so far, there seems to be a lot of interest in this project.
Gilles has proved that Apple
could have enabled NVMe M.2 / SATA booting for at least 5,1 cMPs.
Does it strike anyone that in Apple's history, NVMe is the
only protocol that we can run apps from yet cannot boot from ( 2017 Macs aside - not sure exactly which models ) .
Seems intentional. Not nice on Apple's part considering Mac Pro original high retail prices.
If Gilles project becomes widespread successful it would become a huge slap in the face for T.C. demanding a response.
A common complaint about cMPs is our slow SATA II internal speeds. Two NVMe M.2 SATA's would enable me at least to remove ALL internal SATA II slot drives = reduce heat + increase data transfer speed. Gilles Twitter pics show all empty SATA internal bays - which would release the SATA II power connectors for GPU's or whatever.
Seems T.C. wants the cMP to fade away. I doubt that he has ever seriously used one nor does he appreciate the enormous defection of the Mac professional users to Windows due to the shameless neglect in Mac Pro development.
Just my honest opinion. Correct me if I'm wrong.