Am Di 25. März 2008 schrieb Andy Green: > - gpg control packet > Somebody in the thread at some point said: > > > **pixelclock** has to go down in landscape? Is this for the different address > > sequence scheme in RAM for each of the 2 modes, or does glamo some internal > > transformation by blitting, thus eating 2/3 of the bandwidth or what? > > Yes, I guess the problem for this chip comes from the fact it still > issues a portrait raster to the LCM, so it has to compose each portrait > line from pixels that are totally disjoint in memory. > > In portrait, adjacent pixels are in ascending order in memory, so after > issuing the address to the DRAM and paying the penalty for random > access, prefetch or block fetch modes from the memory work well since > you are going to use that data for the next pixels.
Really bad: if data in video memory is 15bit-packed, and thus not word aligned (hypothetically! i don't believe in this), glamo has to fetch 15bytes per 16 pixel in portrait, and 31bytes per 16 pixel in landscape. :-/ jOERG
