Hi all,
I’ve just noticed that the 16->32bit conversion during software surface
blits uses substancial amounts of CPU.
My test case was to blit a 32bit RGBA surface to a 16bit XWindows screen
or via a second 16bit software surface and then to the screen. In each
case the CPU usage went up from 1% to 20% (as reported by top). I check
that it is the conversion (the blit) - just converting it without actual
screen display still chews through CPU cycles.
Specs: Linux, XF4.0.2, SDL 1.2, 16bpp X, Matrox card, Athlon 700.
Image: 320x200x32->320x200x16 @ 15Hz
Is that to be expected?
Ciao
Andreas–
| Andreas Schiffler aschiffler at home.com |
| Senior Systems Engineer - Deskplayer Inc., Buffalo |
| 4707 Eastwood Cres., Niagara Falls, Ont L2E 1B4, Canada |
| +1-905-371-3652 (private) - +1-905-371-8834 (work/fax) |
Yes.
Hardware conversion is generally not available in 2D APIs, AFAIK.
//David
.- M A I A -------------------------------------------------.
| Multimedia Application Integration Architecture |
| A Free/Open Source Plugin API for Professional Multimedia |
----------------------> http://www.linuxaudiodev.com/maia -' .- David Olofson -------------------------------------------. | Audio Hacker - Open Source Advocate - Singer - Songwriter |--------------------------------------> david at linuxdj.com -'On Tuesday 27 March 2001 22:45, Andreas Schiffler wrote:
Hi all,
I’ve just noticed that the 16->32bit conversion during software surface
blits uses substancial amounts of CPU.
My test case was to blit a 32bit RGBA surface to a 16bit XWindows screen
or via a second 16bit software surface and then to the screen. In each
case the CPU usage went up from 1% to 20% (as reported by top). I check
that it is the conversion (the blit) - just converting it without actual
screen display still chews through CPU cycles.
Specs: Linux, XF4.0.2, SDL 1.2, 16bpp X, Matrox card, Athlon 700.
Image: 320x200x32->320x200x16 @ 15Hz
Is that to be expected?