ZDoom

I do believe it's about time this gets merged.

I must also point out that my DamageMobj refactor was merged, yet it was imperfect. And when a few bug reports came in, I simply fixed them myself, as is expected of me.

I think this submission can also constitute having such treatment done to it. If any bugs occur, Marisa can fix them.

Really wish this'd get merged already. I keep running into situations where I absolutely need it.

If someone else can do it?
Sorry, I don't have much time to do real programming work on GZDoom and that's unlikely to change until the end of August.

So... any chance this can get a final review?

I'm afraid that's another bug unrelated to this. Though it may be related to the one of projectiles/traces phasing through them.

Looks like this is coming along nicely now. Marisa has made those requested changes too, I'm stoked! :mrgreen:

Speaking of custom bounce behavior, Marisa, I've been in desperate need for finding out how to make projectiles not just explode on sloped floors. Where can I find the code for this?

[quote="Rachael"]What would probably really help this along, though, is having some bare-minimum test samples to ensure everything works correctly.[/quote]

In my case I've been making sure it all works through the [url=https://github.com/OrdinaryMagician/flak_m/tree/experimental]experimental branch of doom tournament[/url]. The biorifle and flak cannon need direct 3d floor data access for projectile alignment and custom bounce behavior, respectively.

In the PR's current state I've verified that all works correctly on maps with sloped 3d floors, despite a small bug/glitch that is unrelated to this PR which I'll report later (projectiles and hitscans may sometimes "phase through" sloped 3d floors if they're too thin).

I don't know enough about this part of GZDoom or Doom mapping to evaluate it, nor do I have an interest in learning it - sorry. :)

If _mental_ or dpJudas take a look at it and if they don't anticipate any problems, then I think this can be merged.

What would probably really help this along, though, is having some bare-minimum test samples to ensure everything works correctly.

Of course, Graf can also just merge or reject this on a whim, as well, if the mood ever struck him or he wanted to.

So, I guess it is happening, then. There is now a demand for it from more people than just me.

Huh... I've encountered a point where I'll actually need this to continue what I'm working on.

Totally not a priority right now.

Shamelessly bumping this. I haven't heard anything in a while.

Just wanted to ask if this is still going to be a problem for vulkan or anything. The data exposed is entirely read-only just in case.