I just installed a new Radeon HD 5450 with 1 GB on my 12.2 machine. Previously I was using the integrated HD 4250 on my motherboard. In preparation for the new card, I reverted to the radeon driver and installed the card. Just for fun, I went to the hw acceleration stress test page at mozilla.org, which is below, ran the test, and maxed out the test at 60+ FPS. I then ran atiupgrade and installed fglrx version 13.1. I again ran the stress test and received a score of only 17 fps. According to about:support, hw acceleration was basic 0/1 for both drivers, although my settings in preferences advanced were to use hw acceleration. With my 4250, the test showed 21 fps with fglrx. On the other hand, when I run Flightgear which is 3d graphics intensive, the frame rate for fglrx is 3 times faster than it is for the radeon driver. GLXgears shows a huge difference in favor of fglrx. Unfortunately, the difference in favor of the radeon driver in Firefox is very noticeable, especially for graphics intensive pages. My xorg.log files show no errors and show acceleration as active for each of the drivers.
Does this mean I have to sacrifice desktop performance to get game performance by using fglrx or sacrifice acceptable game performance to get good desktop performance by using radeon? Or am I doing something wrong? It seems to me that there shouldn’t be such a huge performance hit in using what is supposed to be a higher performance driver.
Have others experienced something similar with these drivers?
My results with both drivers were with no hardware acceleration. With hardware acceleration on and the radeon driver, the speed went down to 17 fps. With fglrx it went to about 5. So I’m talking about a different issue than the one addressed in the other thread.