In cases where performing occlusion queries for multiple red walls
leading to the same sector, we would sometimes leak a number of
query objects. This was apparently harmless on NVIDIA drivers, but
would leak tons of memory on ATI.
git-svn-id: https://svn.eduke32.com/eduke32@1993 1a8010ca-5511-0410-912e-c29ae57300e0
The code was using the Polymer-cached version of the sector cstat to
determine the translucency of its diffuse material, meaning it would
never get it right on the first try; that's what you get for having
completely different variables named 's' and 'sec'.
git-svn-id: https://svn.eduke32.com/eduke32@1991 1a8010ca-5511-0410-912e-c29ae57300e0
We normally enable backface culling always, but turn it off for
two-sided sprites; we need to make sure it's re-enabled after that
as sector masks are drawn after.
git-svn-id: https://svn.eduke32.com/eduke32@1990 1a8010ca-5511-0410-912e-c29ae57300e0
Some of the ROR code was operating under the assumption that drawasks
always correspond to one call of polymer_displarooms(), but that's not
necessarily the case; external view draws all sectors from the top
polymer_drawrooms(), which was causing crashes. Just ignore the sector
mask queues in external view mode, we already don't draw masked walls.
git-svn-id: https://svn.eduke32.com/eduke32@1982 1a8010ca-5511-0410-912e-c29ae57300e0
Check all the bunch siblings if we get a floor/ceiling hit and walk
them if their own planes hit.
git-svn-id: https://svn.eduke32.com/eduke32@1981 1a8010ca-5511-0410-912e-c29ae57300e0
When walking the sectors in displayrooms, detect floor/sector masks to
place them in a mask queue. The queues are allocated on the heap, but
a stack pointer is maintained by the displayrooms to allow it to play
nice with mirror recursion.
Right now the masks are drawn after all of drawmasks is finished; that
means you can see sprites and wall masks through sector masks, but not
the opposite. There is a slight attempt at blending far to near, but
nothing too conclusive yet; that means that in some cases some sector
masks won't be visible through other sector masks, but it should work
for trivial cases.
The next step would be to copy all the drawmasks logic into Polymer
and have it (attempt to) properly interleave sector masks where they
belong in the blending order.
git-svn-id: https://svn.eduke32.com/eduke32@1980 1a8010ca-5511-0410-912e-c29ae57300e0
Previously they would get counted both as 1-way walls for immediate
drawing and as mask walls for delayed drawing.
git-svn-id: https://svn.eduke32.com/eduke32@1979 1a8010ca-5511-0410-912e-c29ae57300e0
Assigns the right alpha value to floor/ceiling planes if their
translucency bits are set and make sure that changing the cstat now
triggers an update of the material for editing purposes.
This doesn't do anything by itself, but will be needed to correctly
draw floor/ceiling masks.
git-svn-id: https://svn.eduke32.com/eduke32@1977 1a8010ca-5511-0410-912e-c29ae57300e0
When walking the sectors, put more drawing work between queuing and
reaping the occlusion queries to potentially make them more efficient.
git-svn-id: https://svn.eduke32.com/eduke32@1976 1a8010ca-5511-0410-912e-c29ae57300e0
Enumerate the vertical siblings of each walked sector and check them
against the current frustum. No occlusion queries yet, but that should
correct any missing sectors until a flaw in that logic is discovered.
git-svn-id: https://svn.eduke32.com/eduke32@1975 1a8010ca-5511-0410-912e-c29ae57300e0
Reorganize polymer.c a bit to match the order of prototypes from the
header, move macros from the middle into the header.
git-svn-id: https://svn.eduke32.com/eduke32@1974 1a8010ca-5511-0410-912e-c29ae57300e0
- fix an earlier 'oops': make highpals load correctly again
- maphack lights shouldn't be double-loaded now
- slight tweak to the loading screen fadein/out timing
- two bound checks
- make some variables static in engine.c
git-svn-id: https://svn.eduke32.com/eduke32@1969 1a8010ca-5511-0410-912e-c29ae57300e0
which switches have been hidden in the original and user maps. One is to make
it face a wall, and another is to 'embed' it in e.g. the floor, like the
monitor with the burning fuse in E4L1. Both kinds show up when the switches
are rendered as models, revealing the secrets that the mapper sought to hide.
My proposal, implemented in this commit, is to apply a heuristic for such
switches at premap and make them invisible (set cstat bit 32768). The
conditions are re-checked during the game in case there is a switch coming
out of the floor, for example. A new spriteext bit is used for this feature.
git-svn-id: https://svn.eduke32.com/eduke32@1968 1a8010ca-5511-0410-912e-c29ae57300e0
The two bugs are:
- RPG projectiles don't pass through water
- can't emerge from slime in bobsp2
* Coalesce Polymer lights that are at the same position for FIRE-like sprites so that only one light is actually generated. This is good for usermaps that have multi-sprite fires, like bobsp2 again.
git-svn-id: https://svn.eduke32.com/eduke32@1967 1a8010ca-5511-0410-912e-c29ae57300e0
This fixes a ridiculous interaction problem between non-black fog and
additive lighting passes. Plagman is stupid.
git-svn-id: https://svn.eduke32.com/eduke32@1965 1a8010ca-5511-0410-912e-c29ae57300e0
* fix 'squishing' when taking off shrunk (the reason was TROR game code using updatesectorz) and a potential sector[-1] access
git-svn-id: https://svn.eduke32.com/eduke32@1960 1a8010ca-5511-0410-912e-c29ae57300e0
by using setspritez() instead of setsprite() for the sectnum updating.
* Change 'yax-nextwall(0/1)' to 'upwall/downwall' in the corruption checker, makes it much nicer to read.
git-svn-id: https://svn.eduke32.com/eduke32@1959 1a8010ca-5511-0410-912e-c29ae57300e0
* make Ctrl-Alt-A (formerly 'toggle inner gray walls') now remove all gray walls entirely in 2d mode
* make r_usenewaspect enabled by default
git-svn-id: https://svn.eduke32.com/eduke32@1957 1a8010ca-5511-0410-912e-c29ae57300e0
* fix breaking out of 'switch' blocks and compilation of the 'default' case
* Have a way of assigning 'special function' menu [' F] entries to script states. Writing a string literal after the state name will register the state under that name. Menu names are limited to 24 characters and it's possible to have up to 16 of them.
* new branching command 'ifinteractive', true if a state runs from the menu mentioned above.
* new command: getnumberfromuser <<retvar>> "query_string" <maxnum> <flags>
See 'state collect_teleporting_sectors' in a.m32 for a combined usage of the new functionality.
git-svn-id: https://svn.eduke32.com/eduke32@1955 1a8010ca-5511-0410-912e-c29ae57300e0
* helper feature: when shade preview is enabled, the ceilings and floors of highlighted sectors are shown in pal 6. This makes it possible to highlight the TROR-joining candidate sectors in 3D mode using e.g. 'sethighlightsector searchsector 1' while aiming at the respective ceilings and floors, and get visual feedback.
* when failing TROR-joining early ('no consistent joining combinations found'), print to the console why it failed.
* more Makefile.deps updating
git-svn-id: https://svn.eduke32.com/eduke32@1952 1a8010ca-5511-0410-912e-c29ae57300e0
* New shade/visibility calculation code, which is activated with 'r_usenewshading' (on by default), and is closer to the classic look. Also tweak the FOGSCALE macro to have approximately the same fog distance with all renderers.
* Mapster32: END modifier to RShift. If it's pressed when RShift is released, sprites which are in grayed out sectors are also selected; Make changing shade affect all highlighted sprites in 3D mode (when aiming at one of them).
* some debug code to watch out for suspicious glGenTexture/glDeleteTextures calls, not active.
git-svn-id: https://svn.eduke32.com/eduke32@1943 1a8010ca-5511-0410-912e-c29ae57300e0
First, it is now possible to disable the memcache (which is there to cache the texcache) by setting the new cvar 'r_memcache' to 0. Do this if you're constrained on memory or getting crashes when e.g. doing vidrestart often. Also, the memcache will disable itself (and free its storage, if it's there) the first time it fails to allocate.
Fix a strcpy with identical arguments in game.c and a couple of uninitialized mem accesses related to Polymer lights.
git-svn-id: https://svn.eduke32.com/eduke32@1941 1a8010ca-5511-0410-912e-c29ae57300e0
The main feature is the addition of -mx and -mh command-line parameters to EDuke32 and Mapster32. These parameters include a con and def "module" respectively. This translates into essentially including the file from the bottom of the compiled script.
I fixed the classic buggy behavior of the BROKEHYDROPLANT and REACTOR2 sprites.
I also fixed a small, long-standing bug where FRAMEEFFECT1 blurs are not affected by sector floorpal. You can see one example of this by shrinking the Enforcer on the upper inside of the toppled building in E3L11: Freeway.
I tweaked the Makefile so that it would automatically regenerate the keep.me files in the $(OBJ) and $(EOBJ) directories after they are deleted for cleaning.
One final change is a slight positioning cleanup of both programs' --help dialog boxes.
git-svn-id: https://svn.eduke32.com/eduke32@1937 1a8010ca-5511-0410-912e-c29ae57300e0
Usage: For an ANM file <somefile>.anm/ANM, EDuke32 looks for <somefile>.ivf, which is the VP8 stream transported by an IVF container. It can be extracted from a WebM file with e.g.
mkvextract tracks <filename>.webm 1:<filename>.ivf
(part of Mkvtoolnix, the Matroska toolset)
Libvpx is required, and the 'YUV'-->RGB conversion is implemented using a fragment shader, so it's for OpenGL modes only. Also, this commit doesn't enable the code.
Unfinished: sound, aspect correction for fullscreen w/ non-square pixels, ... ?
---
Make MAXNODESPERLINE in engine_priv.h actually a macro that depends on MAXYSAVES and MAXDIM instead of using the obsolete precomputed value. I think this might have been the cause for the latest patched-up overhead view crash.
git-svn-id: https://svn.eduke32.com/eduke32@1933 1a8010ca-5511-0410-912e-c29ae57300e0
* Read-only CON access to sector bunchnums by adding the labels '.ceilingbunch' and '.floorbunch' to the CON sector structure
git-svn-id: https://svn.eduke32.com/eduke32@1931 1a8010ca-5511-0410-912e-c29ae57300e0
* slightly tweaked status line, now also displaying the number of bunches
* fix a couple of issues unearthed by valgrind and a TROR nextwall corruption when joining sectors
git-svn-id: https://svn.eduke32.com/eduke32@1930 1a8010ca-5511-0410-912e-c29ae57300e0
* it is now possible to paste (and have an outer red wall automatically) highlighted sectors into a sector that is extended on one side and non-sloped on the other. For this, the highlighted sectors must be in one connected component.
* Make it possible to not display inner gray walls, toggled with Ctrl-Alt-A, but only when no manual grayout (Ctrl-R) is in effect.
* bugfix: don't clear original TROR-nextwall link when duplicating extended highlighted sector
* m32script: protect wall members relevant to TROR
* save autogray (Ctrl-A) and showinnergray (see above) to mapster32.cfg
git-svn-id: https://svn.eduke32.com/eduke32@1928 1a8010ca-5511-0410-912e-c29ae57300e0
- sethighlightsector <sectnum> <on?>
- updatehighlightsector
- collectsectors <<array_to_collect_sectnums>> <initial_sector> <<num_collected_sectors>> <sector_filtering_state>
The latter does a breadth-first search starting from an initial sector and collects nextsectors only when the filtering state, given a sectnum as RETURN input, writes a nonzero value into RETURN. As a usage example, a.m32 includes the state 'collect_teleporting_sectors', that collects all sectors containing an SE7 and highlights them afterwards. This way, it should be possible to retrofit old maps with TROR by distributing small scripts that do most of the work (right now, joining has to be done by hand, though).
* corruption checker: for the 'nextwall inconsistent with nextsector' corruption, suggest an alternative fix by searching fitting nextwalls and changing the nextwall of the corrupt wall (as opposed to the nextsector). It will display with a leading '?' in the listing, and can be demanded by suffixing 'corruptcheck tryfix' with it. For example,
corruptcheck tryfix 9-21 ?
would fix some corruptions in Billy Boy's cranium.map without introducing drawing errors.
* fix demo playback (tueidj figured this out)
git-svn-id: https://svn.eduke32.com/eduke32@1927 1a8010ca-5511-0410-912e-c29ae57300e0
* When dragging highlighted sprites, do a setsprite() after each position update. This way, they won't end up on the wrong level
* TROR support for SE 31 and 32 in-game, example provided in test map
* some uncommited stuff for TROR: SE 6/14
git-svn-id: https://svn.eduke32.com/eduke32@1924 1a8010ca-5511-0410-912e-c29ae57300e0
* Make the minimum resolution possible 640x480. Sometimes you'd accidentally enter something like 'vidmode 800 60' and crashes would ensue...
* fix a warning
git-svn-id: https://svn.eduke32.com/eduke32@1921 1a8010ca-5511-0410-912e-c29ae57300e0
Implementation note: the so instrumented sprites use actor[].t_data[7] to hold a magic number (0x18190000 + pivot spr idx), and t_data[8] and [9] for the initial coords relative to the pivot point.
git-svn-id: https://svn.eduke32.com/eduke32@1913 1a8010ca-5511-0410-912e-c29ae57300e0
* fix memory corruption when loading a Polymer savegame using another renderer and then change to Polymer
* fix possible crash in OSD_Exec() and uninitialized mem access in game config reader
* move makeasmwriteable() to baselayer.c
git-svn-id: https://svn.eduke32.com/eduke32@1910 1a8010ca-5511-0410-912e-c29ae57300e0
"The most noticeable change is the addition of the "includedefault" CON and DEF command, which will attempt to include eduke.con (or nam.con, or ww2gi.con), then game.con, or duke3d.def, or nam.def, or ww2gi.def. This is useful for TCs like my add-ons, where for my pseudo-mutators I currently say "include EDUKE.CON", but I also have to juggle this terrible order of paths, so that I can have an EDUKE.CON file in my HRP which says "include GAME.CON" to allow the mainline game to actually run, but also allow DukePlus to load its EDUKE.CON file (since it uses that and not an -x switch), and also allow any custom EDUKE.CON files in the root to be used."
git-svn-id: https://svn.eduke32.com/eduke32@1909 1a8010ca-5511-0410-912e-c29ae57300e0
* Voxel tweaks: horizontally scale wall-aligned ones by 5/4, make them ignore per-tile yoffset in classic (i.e. emulate Polymost; I think this is more sensible since they're not clipped to floors/ceilings anyway), make Polymost know the voxel scale
* Always cull back-facing, one-sided, wall-aligned sprites (classic/Polymost), irrespective of whether it's a sprite, voxel or model. This can lead to falsely not drawing them in certain circumstances, but IMO that's preferable to visible hidden switches etc.
* Change defaults for r_novoxmips to 1 and lazytileselector to 0
git-svn-id: https://svn.eduke32.com/eduke32@1908 1a8010ca-5511-0410-912e-c29ae57300e0
* When TROR-joining and sectors reachable through former red walls could need displacement, ask the user; if something is wrong with the walls of the two components to join, jump to that place and print coordinates of offending wall and its point2 in the OSD
* Ctrl-TAB in 2D mode now toggles filling of the currently aimed-at sector; it's not very in sync with the rest of the controls though (e.g. some commands will potentially affect each sector that contains the crosshair)
* Fill sectors slightly different so that highlighted sectors above each other are easier to distinguish
* In the editor, always clear the screen to a 'rainbow' of the palette before each frame. This makes no-draw ('HOM') and accidental translucency glitches stand out more clearly
* a few misc. tweaks, consistency checks, and fixes
git-svn-id: https://svn.eduke32.com/eduke32@1901 1a8010ca-5511-0410-912e-c29ae57300e0
* Mapster32 now makes sandwiches, too: select all floors of a bunch and 'extend them below'. This will put a new sector between the highlighted ones and the their lower neighbors and displace the z's accordingly. Make sure you have enough headroom.
* To facilitate the above, RAlt now has two more modifier keys (checked at release time, as usual): for every sector that would be selected, END will select all sectors whose floor bunchnum equals the first (this is useful for sandwiching), and HOME will select all sectors whose ceiling bunchnum equals it (not very useful, but provided for completeness).
* Sloping extended sectors is now checked more rigorously
-- misc:
* When deleting highlighted sectors with DEL, pressing LShift will now always delete them irrespective whether the mouse pointer is over one of them. This is useful for the cleanup of corrupted maps.
* 2D drawing is slightly tweaked so that active walls always display over inactive ones
git-svn-id: https://svn.eduke32.com/eduke32@1900 1a8010ca-5511-0410-912e-c29ae57300e0
game: when entering/leaving water or slime, delay changing palette by one game tic to make it look right
git-svn-id: https://svn.eduke32.com/eduke32@1899 1a8010ca-5511-0410-912e-c29ae57300e0
