Yes the problems with the comet cursor and Mozilla seem to be gone.
Also tried in my ECS 2.0 VM with an ancient Firefox 4.0.1 . No issue with the comet cursor with GENGRADD at 3840x2160x16bpp in Virtualbox . I guess that's one problem I never saw.
Yes the problems with the comet cursor and Mozilla seem to be gone.
Also tried in my ECS 2.0 VM with an ancient Firefox 4.0.1 . No issue with the comet cursor with GENGRADD at 3840x2160x16bpp in Virtualbox . I guess that's one problem I never saw.
IIRC, it was to do with DIVE and/or hardware cursors and generally DIVE is disabled with current video drivers. Enough people had the problem that it was recommended to not use the comet cursor.
I'll test with a supported chipset (T42) and see if I can reproduce.
Hmm, had a hell of a time getting DIVE working, seems someone has changed the pref, layers.accelerated to be backwards, when at the default of true, stderr outputs that it is set to false and disables DIVE. Setting it to false allows DIVE and the comet cursor didn't seem to hurt. Anyways stupid change, they should have just changed the default if they wanted to disable DIVE due to the comet cursor issue, and I don't know why else to disable video acceleration unless it's a bug. I'll look at the code later.
On eCS 2.2b2 with Firefox 10.12 and the SNAP I downloaded from Lenovo, when the mouse cursor changes, eg to the pointing finger, the whole system blocks for up to a minute, and then comes alive again. Be irritating inputting text.
Also discovered DIVE can only be enabled in FF with 16M colours.