Haiku is not lightweight!

[phw]

@cb88

That’s the complete opposite of every post you ever posted about graphics acceleration since forever.

Actually it isn’t ¯\_(ツ)_/¯

HaikuStick

2D acceleration? Even if some of our drivers do that (and we’re talking 1990 hardware there), the app_server does not make any use of it. Did you ever notice? Did the rendering seem slow? I think not. Back then, the 2D acceleration had been benchmarked, and it turned out that with our design (double buffer, with app_server rendering most things in system RAM and not video RAM), 2D acceleration turned out to be slower because it resulted in: sending data to the video card (a full frame) sendin…

[GSoC 2017] 3D Hardware Acceleration - Weekly Report 4

Acceleration was intentionally disabled. Modern CPUs are so fast that the 2D “acceleration” on most cards was actually slower, while not allowing antialiased drawing. So we just stopped using it because it was silly. 3D acceleration is a different thing, of course. I would also prefer something more “native”, using the existing accelerant interface. But this is a very huge tasks. It takes Linux a separate team of full-time devs for each supported device family. And we have maybe one or two par…

Performance of future Haiku versions

If an accelerated driver would make the OS slower, we wouldn’t be using such driver. This is indeed what happened with 2D acceleration. We implemented it, benched it, and found out it was slower, so we disabled it.

What improvements do you want to see for Beta3?

We have disabled 2D acceleration a long time ago because it makes things slower on “modern” hardware (anything using PCI or AGP). And for PCI-express most cards don’t even support it, because you can do things better and faster with the 3D acceleration (2D is just a special case of 3D now). Only the hardware accelerated cursor is still supported, I do plan to have a look at that but it isn’t that great of a feature so I worked on other things so far.

etc. etc.