Rockbox mail archiveSubject: Re: Porting time estimate
Re: Porting time estimate
From: Dave Chapman <dave_at_dchapman.com>
Date: Sat, 01 Sep 2007 03:36:59 +0100
Brendon Costa wrote:
> With that sort of information, how long do others think it might take me
> to work on a port of rockbox if I say spend one or two hours a day on it?
As an example the Gigabeat F/X port took about a year from start to a
functioning Rockbox port, and the developers had datasheets for most of
the components, including the CPU. This was with about 3 or 4
developers working in their spare time on it.
I think the Sansa E200 port took about 8 months from the time of the
first contact from Sandisk offering a couple of free players to
developers, and the time the port got sound playback. The CPU/SoC in
the E200 (PP5024) is very similar to the other PortalPlayer SoCs
supported by Rockbox, although a lot of reverse-engineering had to be
done as well.
Unless you can find a datasheet for the STMP3600, you're going to have a
lot of reverse engineering to do. But on the bright side, the ARM core
is already a common target for Rockbox, so it won't be a port from scratch.
My advice would be to try and get other people interested - either
existing Rockbox developers who may be interested in working on new
targets, or (more likely to succeed) other skilled Sansa Express owners.
My observation of other ports is that the key to attracting other
developers is to publish all your work on a website as early as
possible, and keep it up to date. It's worth the investment in your time.
And of course, read this wiki page (if you haven't found it already):
Received on 2007-09-01