On 22 Oct, Terje Sletteb=F8 wrote in message <005a01c6f604$e7ec4140$9a02a8c0@pc>: > >From: "David J. Ruck" <druck@xxxxxxxxxxxx> >=20 > > All OS modules and applications that do any manipulation of image dat= a > > ex=3D pect the order of RGB tbits o be as specified in the PRMs, and > > will not work otherwise. >=20 > Thanks for your reply. I'm a little "rusty" having been away from the > platform 10 years or so, does anyone know where in the PRM this order i= s > stated? I didn't find it with some searching. Page 5-88, I think. > Anyway, I did a quick test on my Iyonix with the FX5200 graphics card, > using BASIC to write directly to the video memory, and it appears that > the graphics card, in 32-bit mode, has the word layout &BBGGRR, or migh= t > it be that this write is trapped and reversed, in software? No, that's correct: it's &00BBGGRR. > > > What I'm getting at is that, if possible, we should get a solution > > > that=3D minimises or avoids run-time overhead, if possible. > >=20 > > That solution is already in place, except for little used 32K modes. >=20 > Ok, so I'm trying to find out exactly what this solution does, i.e. if > it's transparent for the programmer, even working directly with the > video memory, or if it only works with higher-level abstractions. OTTOMH, I would imagine that it is transparent as far as messing with the= screen memory goes. That's as low level as RISC OS apps go, unless they want to be exclusively Iyonix-based. --=20 Steve Fryatt - Leeds, England http://www.stevefryatt.org.uk/ --- To alter your preferences or leave the group, visit //www.freelists.org/list/iyonix-support Other info via //www.freelists.org/webpage/iyonix-support