[maemo-developers] [maemo-developers] Cairo performance comparison, 770 / N800 / PXA-320

[From: klaus at rotters.de]

Am 13 Jan 2007 um 21:00 hat Kalle Vahlman geschrieben:
> For the cairo audience there's the question of the tessellation
> process, can it really be so fast on the PXA-320 or is there a bug
> somewhere that twists the results? What could be so good in PXA-320
> (or not-good on the other devices) that the results are so drastic?

I didn't know that the N800 has a PXA-320 uC by Mravell. So I googled a 
bit to find out that it has several interessting points:

See here: http://www.marvell.com/products/cellular/application/pxa320.jsp

* a 32 bit memory interface (the TI OMAP uC in 770 has just 16 bit) IMHO 
very important
* a 256 kB L2 cache (the TI OMAP just has a L1, IMHO)
* a build-in 768 kB frame buffer with 2D accellerator. On the 770 there was 
just the 16 bit data path to SD-RAM sharing the access to the video/frame 
buffer chip. So I think there is no longer a dedicated framebuffer chip in the 
N800.

So it becomes more clear why this release of IT2007 could not run on the 
N770. The underlying hardware seems to be very different.

Has anyone more detailed information about the hardware used in the N800?

-Klaus
-- 
 Klaus Rotter * klaus <at> rotters <dot> de * www.rotters.de