And Hoikas said, "I cannot tell a difference in the skinning times on my
not-crappy computer with only a tenth of a millisecond resolution." So we
now have hundredth of a millisecond in the profiling display. These are
the kinds of problems you have when developing with a machine that doesn't
suck. It would be nice if players didn't try to run the game on machines
from before the exctinction event that ended the Permian era.
This should help people realize that things are happening... Even if
whatever server they're connected to takes a very long time to process a
join request. At some point in the future, the linking process should be
de-serialized.
Seriously? Offsetof has been around for a long time, and eap added a
compiler_assert for his crazy netcode. There's no excuse for that to be
happening at runtime.
- Age User Name and Age Instance Name are equal in the case of some child
and public ages. If they are, only return one of them. Don't return
"Ae'gura Ae'gura"
- Changed to return plString instead of a pointer to a temporary buffer
that immediately goes out of scope
Turns out, it was an artifact of us suspending the simulation during links
and partly because of Cyan's late adding of the avatar controller to the
sim. Now, we add the avatar as soon as the age data is loaded. This causes
the camera stack to be populated with whatever garbage PhysX decides on,
then xJourneyClothsGen2 is free to set the real stack after we get all the
SDL from the server.
Verified to fix Teledahn oddness and not display a regression in Kemo.
This happens on DirtSand shards when two avatars are loading the same age
at once. LoadCloneMsgs are dispatched to the clients before they have
loaded the original keys. The server has no way of knowing that however,
so this is the correct place to fix the problem.
This should keep the avatar from getting stuck on some stupid stuff such
as on the pellet machine. A real fix would be to tweak the data, but we
cannot do that.