The glamo chip and its future
Carsten Haitzler (The Rasterman)
raster at openmoko.org
Sat Jun 28 17:05:43 CEST 2008
On Sat, 28 Jun 2008 12:57:47 +0200 Michael Stather <lists at michaelstather.com>
> Couldn't the glamo just be driven with a highter bandwidth, or is this a
> chipset limitation?
no it can't. we already are seeing performance far beyond anything the glamo's
designers have ever seen. in theory we could get more, but the chipset (glamo
+ samsung 2442 combo) won't be getting that for us. :(
> And if the chip just isn't capable of performing better, what about the
> idea to have plain framebuffer access only through the processor but
> include a CLPD/FPGA in the design which may be loaded "on the fly" with
> a firmware for a simple 3D chip. I was involved in a project of a game
no possible. too late. phone is shipping. lcd is wired up to the glamo. on-chip
lcd controller is not connected. fpga's are power hungry. not going to happen
on a phone. also tyhe glamo is our interface to the sd-card in the freerunner.
> console, with 2D acceleration only, but IMHO simple 3D acceleration
> isn't that hard. And this would have the advantage that the 3D core is
> fully open and can even be improved with optimisations and more features.
> I think through its openness openmoko could also become quite popular
> for gamers, since there exist many good open-source games which could be
> ported easily to the platform if there was hardware acceleration (OpenGL
> ES or even better plain OpenGL). Thinking of consumers, they want fancy
> 3D effects and games, and since all options nowadays are either j2ME
> commercial games or smartphones which have all a rather small community
> or are closed like the iPhone, this could becomne another real advantage
> of the openmoko platform.
> Openmoko community mailing list
> community at lists.openmoko.org
Carsten Haitzler (The Rasterman) <raster at openmoko.org>
More information about the community