nanogui: Thread: nano-X speed

Hi,
I'm using nano-X under X and I find it quite slow. For example a
GrMoveWindow for a 600x400 window - with no other overlapping window - takes
about 2sec. 

Does it sounds normal, or is it because of the X backend ? Should I expect
better performance on a framebuffer ?

I should say that the X server is Exceed running on Windows 2000, on the
other side the linux machine is a PIII 800.

-- 
Fabrice Gautier, 
####@####.####

: I'm using nano-X under X and I find it quite slow. For example a
: GrMoveWindow for a 600x400 window - with no other overlapping window - takes
: about 2sec. 
: 
: Does it sounds normal, or is it because of the X backend ? Should I expect
: better performance on a framebuffer ?

The primary reason for the speed problem is the result of the Microwindows
X11 blit implementation.  It currently reads the pixels one at a time
and copies them into a PSD struct, only to be copied back out.  What
is needed is an X11 screen-to-screen blit that doesn't copy in the
screen driver.  The reason this ended up this way is a bit complicated,
but should probably be redone.

Another quick fix would be to comment out the "quick" window-move
code in nanox/srvfunc.c::GrMoveWindow, and have the windows
redraw themselves...  This will work faster on X11, but is worth it.

Regards,

Greg

I'm using nano-X under Frmebuffer and I find it quite slow.its getting
problem with my embedded device.how can speed up display on my Embedded
Device.

Regards,
Mohan.

mohanreddy.b ha scritto:
> I'm using nano-X under Frmebuffer and I find it quite slow.its getting
> problem with my embedded device.how can speed up display on my Embedded
> Device.

Hi,
Have you tried to write directly onto /dev/fb0 (or wherever your 
framebuffer is mapped) and see if it is a framebuffer driver problem or 
a user space problem? How (and how frequently) does your framebuffer 
driver read /dev/fb0 and write to the video controller (if any)?
Some more details on your architecture may be useful.

Regards,
-- 
Carlo Zinato
Viscount International S.p.A.
R&D Labs, Ancona, Italy
Tel. & fax 0039 071 2905127/2905142
####@####.####

No virus found in this outgoing message.
Checked by AVG. 
Version: 8.0.100 / Virus Database: 270.4.1/1515 - Release Date: 23/06/2008 19.16