Caanoo / WIZ Caanoo Framebuffer Question

[SiENcE]

Hey,

does someone has a clue why a dump from /dev/fb0 is 24bit? I thought the Caanoo has 16bit. This is what i get ("cat /dev/db0 >output.raw"). It's 24bpp and BGR color encoded (not RGB)!

Maybe it's normal, but i was expecting only 16bit.

thx.

 

[Rikku2000]

Hmm i try now too ad it into quake1 but the result is 0.
code is: system("cat /dev/db0 >output.raw");

and when i use it over an gpe file my output raw have 0 KB

 

[crow_riot]

Hmm i try now too ad it into quake1 but the result is 0.
code is: system("cat /dev/db0 >output.raw");

and when i use it over an gpe file my output raw have 0 KB

there's a typo, i think you should use "fb0" instead of "db0" ...

 

[Rikku2000]

thanks but quake dont want build the raw...

 

[SiENcE]

hi rikku,

try to save the output on your sd card. ">/mnt/sd/output.out" or something.

back to topic: Can someone answer my question please?

cheers
SiENcE

 

[Zhek]

The Caanoo supports both 16 and 24 bit depth modes. I have tested them both and they are mentioned somewhere in the documentation.

 

[SiENcE]

Mh. So the Main Menu is 24bit?

Strange.

 

[Pickle]

I remember someone at gph mentioning it would would 24 bit

 

[SiENcE]

thx Pickle.

So its better to use 24bit instead of 16bit because of performance loose while it's converting?

 

[Reesy]

The pollux chip supports 8,16,24,32 bit RGB modes, the docs say that internally everything is 32bit (well actually the docs describe the format as A4R8G8B8 which would be 28bit which make no sense so its probably a typo and should be A8R8G8B8 ie 32bit).

The hardware should be doing all of the conversion work behind the scenes, so it should be more efficient to use 16bit since it requires less writes to fill the framebuffer.