Re: [coreboot] [RFC] marketing coreboot after recent support of AMD board

[Scott Duplichan]

Paul Menzel wrote...

]In my opinion it is very important though that the coreboot support of
]that board is comparable to the one by the vendor BIOS

This is where it is easy to underestimate the work that goes into a
BIOS. A tier 1 BIOS can take 10+ man years of BIOS development time,
even after paying an IBV (independent BIOS vendor such as AMI or
Insyde) huge $$$ for their reference BIOS. ASRock probably did not
spend 10 BIOS man years, but maybe one or two. Their AMI Aptio UEFI
BIOS has many setup features that not found in coreboot+seabios.
Some are generic and will eventually find their way into the coreboot
or seabios code base. Others, such as over clocking controls, take a
lot of customization work. Schematics and NDA documents are also
needed. I didn't even try the PCI express slot. It is unlikely to 
work without some additional BIOS code. The AMD reference code will
go a long way towards simplifying S3 power state support, but adding
it could still be a large job. Supporting hardware monitor features
takes a lot of work. The vendor BIOS has all of this and lots more.
For some of us, these features are mostly unimportant and coreboot+
seabios is adaquate today.

] before
]announcements are made, so that one can show a video or comparison
]charts of the boot time and so on. That would make people talk about
]coreboot and create demand. So hopefully Scott will be able to get
]Windows and GNU/Linux booting soon.

Getting Windows and other operating systems to boot is not a huge
job. But it does take many hours of debug and unfortunately I have
to catch up on some other work.

]Unfortunately I still cannot find the board ASRock E350M1 in stock
]anywhere in Germany.

I think we will find it is easy to get coreboot going on any board
that uses this processor and south bridge. Hopefully additional 
E350 coreboot ports will appear soon.

]One question, to make this message even longer, is the problems Scott is
]having to boot Windows just due to the board port (lack of
]documentation(?)) or does the generic new AMD "chipset" code also needs
]more work, so that it would be better to wait for another board where
]then coreboot could be ported to more quickly after Scott found the
]"shortcomings" of the chipset code?

The chipset code is in fine shape. AMD gave us everything, and then some.
To get Windows working takes nothing more than debug time. WinPE 3.0 
boots, so it is pretty close.

Thanks,
Scott

]Thanks,
]
]Paul