[after r3601] Multiplayer Desync

[edward850]

I added the missing call to the equivalent point to where it is in the software renderer - but I'm currently on a computer where I can't compile the engine so someone else will have to do any checks.

I'm getting "identifier not found", even after including p_local.h.

[Graf Zahl]

Figures...

The one time I can't check I get the case wrong...

[edward850]

I didn't think of that either. And yes, that fixed it.

[Graf Zahl]

On second thought about this whole thing, I am even more convinced that the current way prediction is handled is wrong. Right now it's done for each single view that gets rendered. And all gets started at FRenderer::RenderView. So wouldn't it be best just wrap the call to that with P_PredictPlayer and P_UnPredictPlayer?
That way prediction is cleanly moved to the play code and the renderer kept clean of the intricacies of the play simulation.

The only other place where the renderer is called is for savegame pictures and there it hardly matters whether players are predicted or not.

Of course that's one for Randy to decide so for now I'll leave this thread open.

[randi]

Okay, did that.

[Matt]

I'm still getting a problem in r3881/r1456, but apparently only while loading skyboxes, glows or dynamic lights (but, apparently, only when a dlight is visible from the very start).

[Edward-san]

Please don't bump threads in closed bugs/suggestions. If you want to refer to this thread, link it. I suggest to move these two posts on a new thread in 'Bugs' forum.

[Graf Zahl]

Even then, it's not a ZDoom thing.