Rockbox mail archiveSubject: Re: Neo35 car player
Re: Neo35 car player
From: BlueChip <cs_bluechip_at_webtribe.net>
Date: Mon, 19 Apr 2004 09:17:06 +0100
At 08:52 19/04/04, you wrote:
>On Mon, 19 Apr 2004, tim vogt wrote:
> > Do any of the Rockbox core crew know when OpenNeo took their starting
> > snapshot of Rockbox?
>Quite a long time ago (I think you could ask them, or possibly attempt to
>figure it out by reading the Open Neo sources/changelog). We've seen diverted
>quite a lot. I personally think the Open Neo guys actually on purpose changed
>the code to divert more than they had to since a lot of therir changes merely
>changed comments from /* to //, modified white-spaces and similar no-ops.
LOL - I knew kids who used to do that at school - thinking that because
they didn't understand C, then therefore the teachers wouldn't either :)
>They did however extend the ATA code to work with the bigger disks the Neo
>players often have. We have not incorporated those changes.
I guess it will be of no use to us until laptop drives get bigger.
>After their fork, we've done HUGE changes to Rockbox, the mp3 code restructure
>perhaps being the one with the biggest impact. Today, I don't think you can do
>many patches on any "side" that would apply cleanly on the other.
>We know that the Open Neo guys track our patches closely since they get/apply
>most of our core bug fixes, and while they now actually provide a cvs commit
>mailing list, hardly anything ever is committed in that CVS so we are
>prohibited to do the same. That's why I think they still benefit from us,
>without ever contributing anything back. I've mentioned this numerous times to
>them, but they just think I'm a whiner. Which is true too.
I also think it is disrespectful when people use your code without credit
...I too am considered a whiner. Maybe if you changed your name they would
offer you more respect?
<sorry, couldn't resist that ;)>
> > If someone (e.g., me) were to attempt to port Rockbox to the Neo35,
> would it
> > make sense to look at the diffs between the Rockbox version they started
> > with and OpenNeo, and then try to merge the relevant diffs into the current
> > version of Rockbox? Or is that too simplistic?
>Since we already have support for Neo in the build environment, the first
>thing you'd do would be to make the development work for Neo35.
>This actually builds a version for Neo35 already today, but it won't work!
>You should focus on the very early startup code next. I know the Neo guys had
>to do some stuff to make the Neo startup that we don't do on the Archos, and
>walking through the startup of the Open Neo code and make sure that Rockbox
>does the same for Neo would be basicly what is require to make Rockbox boot on
>Once it boots, it'll be a lot easier to continue working. Then you'd basicly
>make sure that all code is made to work with 4 lines of text etc.
Yes I guess the keyboard and buttons will be the most significant difference.
Would it be fair to say that all Rockbox does is:
# init chips
# read hdd
# read buttons
# write to display
# write to MAS
..If so you should be able to quickly identify where most of the problem
may lie :)
I note from the photos that NEO has the 3507 (player) chip
(*NOT* the 3587 (recorder) chip.)
Given only 256K of memory, you may want to forget or tweak plugins not to
steal 32K of it - most plugins will be useless anyway!
The great thing is a socketed ROM - so no risk of killing the unit :)
Received on 2004-04-19