@olegil

Thank you for explains. I mean a base code for all the devices: i.e. video card, audio card, whitout optimization. An example is the vesa drivers, that works with the majority of video card, but are not optimized.

_________________
retired

@AmigaBlitter

Ok, that is something different. But such a driver doesn't need to not be CPU optimised. It's just that it's made to support as much of the devices of that type as possible.

There's nothing stopping you from making a VESA driver with full CPU optimisation. In fact, if you had to do a VESA driver that is the smartest thing you can do, since VESA is very slow already.

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@olegil

With VESA what you have to do is to have a second buffer (shadow buffer) in RAM and then for write accesses do it both in VRAM and in the shadow buffer. Reads can then be done solely from the shadow buffer. So no slow VRAM reads needed.

This results in good speed as can be seen in XFree86/Xorg X11 vesa driver.

@amitv

Abra-CADABRA!!!