John Martin wrote: > Martin Bochnig wrote: > >> >> >> PGX24 is also m64. >> And therefore "NO", according to what Alan Coopersmith has said. > > PGX24 (ATI Rage Pro) and m64 (ATI Rage-XL/Mach64) have enough > hardware differences that it warranted separate Solaris drivers. > They are not the same. > >> >> But the default Xorg Ati drivers work well enough. >> Ati == Ati. >> Which benefit do you expect from a SUNW provided driver? >> > The biggest issue with using Xorg drivers is some of them rely on > the x86 VBIOS on the card to do the low level initialization and setup. > The FCode on the Sun pgx24 and m64 does low level initialization but it > may setup the card in a mode unexpected by the Xorg driver. A lot > of testing needs to be done to verify the behavior of Xorg drivers > on SPARC.
That's exactly true. ... and it was also the reason for me, to ebay all SUNW Ultra and Blade workstation models ever produced plus all the different frame buffers (except SB1500/SB2500/U25 and U45 workstations). It is indeed necessary to systematically go through a complete as possible testing matrix. (i.e. Some frame buffer works well in a sb1k/sb2k ---> Then put a RSC in, it will load it's sf280r obp image from then on and no ati frame buffers will work under Xorg anymore !!). Thanks for mentioning that point again. Genesi btw has found in interesting solution for their ieee1275 based ppc systems to work around the card initialization problem.
