Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

BUG: ezquake issues with certain tf maps #551

Closed
lordee opened this issue May 7, 2021 · 5 comments
Closed

BUG: ezquake issues with certain tf maps #551

lordee opened this issue May 7, 2021 · 5 comments

Comments

@lordee
Copy link

lordee commented May 7, 2021

ezQuake version:
tested on multiple: 3.2 (patched for larger lightmap limit), 3.6

OS/device including version:
confirmed issues on windows and linux

Describe the bug
pineapple: cannot see player models in demo
cathedral: cannot see walls in demo (this map has a lot of func_walls and possibly structural/normal brushes with no textures on them, which appear transparent in ezquake)

both of these maps have a large number of entities - pineapple has some clock thing on the wall made of i don't know how many entities and cathedral makes use of a large number of func_walls

mvd was recorded on fteserver
cathedral.zip
pineapple.zip

@meag
Copy link
Contributor

meag commented Jul 9, 2021

Hi - I'm missing the models to play these, what mod is it?

@meag
Copy link
Contributor

meag commented Jul 9, 2021

Scrap that - got it working after some fortress-one downloads...

meag added a commit to meag/ezquake-source that referenced this issue Jul 9, 2021
QW-Group#551

No idea why the demos are like this, but ez/FTE have
  hacky-looking code that conflicts with each other

Differences in when incoming_sequence is advanced, so
  ezQuake reads first full packet into 0 but then first
  delta packet into 2, referencing 1 and hence broken
FTE advances incoming_sequence on first mvd packets
  even though they have 0ms delta (so the first packet
  is read into position 1 and the first delta packet is
  2 referencing 1 etc.

Both look slightly dodgy but have been around for ages,
  not touching it... adding another hack of copying the
  first entity packet in mvd into both positions so
  whichever way round it should work
meag added a commit to meag/ezquake-source that referenced this issue Jul 9, 2021
GLM_DrawWorldExecuteCalls() still switches between
  the two, but toggling r_drawflat etc left the alpha
  surfaces in the old mode, as the flags were never
  tested so it wasn't recompiled
Found during QW-Group#551, 'track 2meht5r' at the start
meag added a commit to meag/ezquake-source that referenced this issue Jul 9, 2021
Found during QW-Group#551
"track <name with char over 128>" raised debugging
  warning
@meag
Copy link
Contributor

meag commented Jul 9, 2021

More changes to come unfortunately - classic renderer + alpha surfaces was never completed, and is currently worse than 3.2

@meag
Copy link
Contributor

meag commented Jul 11, 2021

Didn't link this very well in the commit text...

Cathedral
f0607af - FTE's first entity packet was read into frame 1 and the first delta packet referenced it, ezquake read the first packet into 0... applied workaround where first packet is read into both. FTE & ezQuake's code to make the difference so this seems to be a long running problem, not sure why it hasn't come up before.

Also some fixes for the alpha-tested surfaces... our handling of these is still pretty poor though (8cf3fed, aa04fd5, 6f5854c)

@meag
Copy link
Contributor

meag commented Jul 11, 2021

Pineapple
Players not visible - might need to check with Spike for the true solution to this (it's an FTE extension), but can't see how the extended model limits handle the player.mdl index being > 256. When playing there's the high bit set in the player skin but that's not present in mvd. Have applied hack where if the 256 + is the player model, and the current player model is a brush model then we assume it's broken and update. Am also looking for the skin flag now but it's not set in the demo.

Please reopen if there are any problems.

@meag meag closed this as completed Jul 11, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants