I was checking for GCC >= 4.7, while Clangs pretends to be GCC 4.2. Use a
feature test macro instead. The comment I made in r4161 regarding GCC vs.
Clang code was wrong. Now, Clang generates slightly faster code for these cases
(solid and masked 4-pixel wide vlines).
git-svn-id: https://svn.eduke32.com/eduke32@4182 1a8010ca-5511-0410-912e-c29ae57300e0
Which was especially noticeable in the Lunatic build. Also, use
glPushAttrib/glPopAttrib there.
git-svn-id: https://svn.eduke32.com/eduke32@4176 1a8010ca-5511-0410-912e-c29ae57300e0
This "simply" requires an appropriate factor in the projection matrix.
Also, get rid of a special-case factor for >=1.6 aspects (making HUD models
wider then, I think).
git-svn-id: https://svn.eduke32.com/eduke32@4172 1a8010ca-5511-0410-912e-c29ae57300e0
Known as getactor/sprite[].xpanning from CON, values are 0-255. It should be
assumed that this is only valid for wall-aligned sprites (currently, face
sprites also pan).
A test is provided in lunatic/test/animatesprites.con. It should be run in E1L2
and only with the Lunatic build.
git-svn-id: https://svn.eduke32.com/eduke32@4170 1a8010ca-5511-0410-912e-c29ae57300e0
By detecting them in the SDL_KEYDOWN events. This list is getting a bit
ridiculous...
git-svn-id: https://svn.eduke32.com/eduke32@4163 1a8010ca-5511-0410-912e-c29ae57300e0
Compiling a 32-bit NOASM build resulted in code containing a MOVAPS instruction
that accessed a memory location not aligned to 16 bytes (MinGW, GCC 4.8).
git-svn-id: https://svn.eduke32.com/eduke32@4162 1a8010ca-5511-0410-912e-c29ae57300e0
The functions mvlineasm1, mvlineasm4 and tvlineasm2 can now be set to clamp
the vertical texture coordinate (vplc), preventing the unsightly stray lines
on the bottom of non-y-flipped sprites. (The first part of this effort, r3483,
handled their top).
However, this is only enabled for the mvlineasm ones: the vectorized variants
suffered almost no slowdown (even though a PADDUSD SSE instruction would be a
nice thing to have), while it was pretty significant for the sequential
translucent ones.
Summarizing, this leaves two cases where stray lines may appear in the non-ASM
build (the saturation is NYI for a.nasm):
- at the bottom of y-flipped sprites
- at the bottom of translucent sprites (can be toggled by #define)
Another observation is that recent GCC generates much faster code for this
stuff than Clang from SVN.
git-svn-id: https://svn.eduke32.com/eduke32@4161 1a8010ca-5511-0410-912e-c29ae57300e0
For a full 1680x1050 screen drawing a solid/masked wall, the FPS increases
from 118 to 133 and from 114 to 116 (respectively) for me.
Guarded by the macro USE_VECTOR_EXT in the source.
git-svn-id: https://svn.eduke32.com/eduke32@4160 1a8010ca-5511-0410-912e-c29ae57300e0
This fixes some init and update anomalies: before,
- init would always be the graphical mode (0), but writing
merely "osdtextmode" would switch it
- osdtextmode would not get saved to the cfg
git-svn-id: https://svn.eduke32.com/eduke32@4139 1a8010ca-5511-0410-912e-c29ae57300e0
These were too much "looking into the implementation of a module" for my
taste, hence OSD_GetShadePal() earlier.
git-svn-id: https://svn.eduke32.com/eduke32@4138 1a8010ca-5511-0410-912e-c29ae57300e0
This renders tweaks like r4122 unnecessary.
Also, note a "bug" with the program generating that C code.
git-svn-id: https://svn.eduke32.com/eduke32@4135 1a8010ca-5511-0410-912e-c29ae57300e0
I updated the help windows to prefer these variants because they are superior to the -xSquished versions.
Also, factor out the command-line processing code for the above, plus con/def modules and clipmaps.
git-svn-id: https://svn.eduke32.com/eduke32@4128 1a8010ca-5511-0410-912e-c29ae57300e0
So that there are no error messages like "M32 file `a.m32.m32' not found."
Also, in build.c's 'save as' code, add a bound check that would probably
always pass in practice, but looks a bit safer and may fail in very cornerly
cases.
git-svn-id: https://svn.eduke32.com/eduke32@4124 1a8010ca-5511-0410-912e-c29ae57300e0
Instead of directly in build.c:app_main(). Previously, it was possible that an
argument to an option, e.g. "WGR2" with "-game_dir WGR2", would get interpreted
as the initial map to load. This in turn would attempt to execute WGR2.cfg as
sequence of OSD commands -- where that file is an EDuke32 config file instead!
Now: maps are passed without any options, e.g. "mapster32 [opts...] debug.map".
The map file name is not added to the "additional parameters" for map testing.
git-svn-id: https://svn.eduke32.com/eduke32@4121 1a8010ca-5511-0410-912e-c29ae57300e0
Because the one implemented on top of X shows a special mark for tabs instead of
whitespace.
Also, in wm_msgbox(), use vsnprintf.
git-svn-id: https://svn.eduke32.com/eduke32@4120 1a8010ca-5511-0410-912e-c29ae57300e0
This fixes ESC in certain situations such as the editor map selection menu.
git-svn-id: https://svn.eduke32.com/eduke32@4117 1a8010ca-5511-0410-912e-c29ae57300e0
The leak happened because a struct was hashed that had uninitialized
bytes in padding inserted by the compiler. The hash string in now constructed
as concatenation of three CRC32s as 8-byte hex strings, i.e. the individual
CRC32s are padded with leading zeros.
Note to users: because of the hash change, it's sensible to delete the
'textures' and 'textures.cache' files.
git-svn-id: https://svn.eduke32.com/eduke32@4096 1a8010ca-5511-0410-912e-c29ae57300e0
Variables at global scope which share a name tend to actually be the same variable. Not good when you pass such a variable to function that immediately returns if the passed parameter (the variable) equals said variable (itself).
git-svn-id: https://svn.eduke32.com/eduke32@4090 1a8010ca-5511-0410-912e-c29ae57300e0
NOTE TO TEAM MEMBERS: Please review this, as I am not sure what is actually intended here.
git-svn-id: https://svn.eduke32.com/eduke32@4087 1a8010ca-5511-0410-912e-c29ae57300e0
- Don't print osd.szCSDVersion when it is empty, preventing an extra space between the OS name and version number. (An example of what this field holds is "Service Pack 1".)
- Under SDL, allow the "console" edit field to repaint after "OK" is clicked. Important when a long def file is loading.
git-svn-id: https://svn.eduke32.com/eduke32@4086 1a8010ca-5511-0410-912e-c29ae57300e0
Unfortunately, SDL2 does not (yet?) provide a Yes/No box for wm_ynbox().
git-svn-id: https://svn.eduke32.com/eduke32@4085 1a8010ca-5511-0410-912e-c29ae57300e0
Most often, this had happened when casting comparison functions for qsort()
like these: "int yax_cmpbunches(const int16_t *b1, const int16_t *b2)"
to a function pointer type expecting "const void *". Alas, this is undefined
behavior: see
http://blog.frama-c.com/index.php?post/2013/08/24/Function-pointers-in-C
and posts linked from it.
At least two cases have not been fixed:
- The savegame system maintains pointers which are either to data or a function in
the generic "void *". This ought to be made into a union.
- Probably also:
#define OSD_ALIAS (int32_t (*)(const osdfuncparm_t*))0x1337
git-svn-id: https://svn.eduke32.com/eduke32@4068 1a8010ca-5511-0410-912e-c29ae57300e0
- Currently: only tvlineasm1 and tvlineasm2, but incomplete (no reverse
translucency, nonpow2 textures will crash)
- For System V AMD64 calling conventions; requires YASM
git-svn-id: https://svn.eduke32.com/eduke32@4066 1a8010ca-5511-0410-912e-c29ae57300e0
This reverts commit c7e51147f1869d42fc8365f748bb8f6bd4c6fc2f. [r3835]
Conflicts:
eduke32/build/src/defs.c
[Dunno what caused this, but I cleaned this out. Also, the change from
build.c was not reverted.]
git-svn-id: https://svn.eduke32.com/eduke32@4061 1a8010ca-5511-0410-912e-c29ae57300e0
This reduces the proportion of time the trig calculations take of the whole
preprocessing from 50% to about 13%.
git-svn-id: https://svn.eduke32.com/eduke32@4060 1a8010ca-5511-0410-912e-c29ae57300e0
The offsets (that will translate the axes for .pitch/.roll rotation) were
calculated improperly with screen sizes other than 1280 x 1024 and were
dependent on other related variables. The fix proceeded empirically, i.e. mostly
trial, error, and some educated guesswork. (It's telling that the magic constant
1280 is needed.) A test for Lunatic is added to test.lua that will rotate
SEENINES.
Bug pointed out by Mblackwell.
git-svn-id: https://svn.eduke32.com/eduke32@4058 1a8010ca-5511-0410-912e-c29ae57300e0
- engine.c: comment out some unused decls with RELEASE=0 DEBUGANYWAY=1
- Lunatic: update dump.lua and v.lua
- Lunatic: test/test_geom.lua: note a perf regression (wrt older LuaJIT,
or I mis-configured / built something wrong)
git-svn-id: https://svn.eduke32.com/eduke32@4048 1a8010ca-5511-0410-912e-c29ae57300e0
(Auto-insertion of an inner sector from a loop happens either because the loop
is created in an extended sector, or because it has been punched through a
ceiling/floor into the upper/lower neighbor sector.)
git-svn-id: https://svn.eduke32.com/eduke32@4033 1a8010ca-5511-0410-912e-c29ae57300e0
Note: the wall line will not light up periodically, but point insertion will
be possible nontheless. A bit inconsistent, but I didn't want to do a full
analysis of which side effects allowing a sprite highlight and a line highlight
to coexist at the same time would have.
git-svn-id: https://svn.eduke32.com/eduke32@4026 1a8010ca-5511-0410-912e-c29ae57300e0
This makes CLANG=1 RELEASE=0 DEBUGANYWAY=2 OPTLEVEL=1 builds definitely pretty
playable (only Polymost tested) while having a decent amount of sanitization.
git-svn-id: https://svn.eduke32.com/eduke32@4020 1a8010ca-5511-0410-912e-c29ae57300e0
Redundant for the same reason as the preceding one: the pointer has already
been dereferenced at the point the check is made. Also, all 5 calls of that
function pass a valid pointer (provided the pointers from which the expressions
are derived are valid). Reported by Stack, too.
git-svn-id: https://svn.eduke32.com/eduke32@4017 1a8010ca-5511-0410-912e-c29ae57300e0
Again, in case a bunch is discarded. The tags would appear as really tagging
the walls instead of being "hidden".
git-svn-id: https://svn.eduke32.com/eduke32@4012 1a8010ca-5511-0410-912e-c29ae57300e0
That is, if a bunch is discarded when copying a sector to the clipboard.
(This happens if not all sectors that are part of the bunch are copied.)
git-svn-id: https://svn.eduke32.com/eduke32@4011 1a8010ca-5511-0410-912e-c29ae57300e0
Don't touch TerminX's r_usetileshades==1 or it being the default.
git-svn-id: https://svn.eduke32.com/eduke32@4010 1a8010ca-5511-0410-912e-c29ae57300e0
- For setting 1, don't draw them for red walls whose sector floors have equal
height.
- Make setting 1 the default, because I consider drawing them for such walls
(and white walls) noise (as opposed to information). The "verbose" setting 2
is still available.
git-svn-id: https://svn.eduke32.com/eduke32@4007 1a8010ca-5511-0410-912e-c29ae57300e0
DONT_BUILD, because there have been no significant changes of the non-Lua build.
git-svn-id: https://svn.eduke32.com/eduke32@4006 1a8010ca-5511-0410-912e-c29ae57300e0
- libs and headers are expected in platform/Windows/* (not there yet)
- prefix Lua #includes with luajit-2.0/
- build bytecode objects with absolute path name (used for debug info)
git-svn-id: https://svn.eduke32.com/eduke32@4000 1a8010ca-5511-0410-912e-c29ae57300e0
- Use the proper tile sequence when multiple pskies are present in a map.
However, there's still only *one* psky chosen at map load time.
- Handle three horizfrac cases:
* 0, psky always at same level wrt screen
* 65536, psky horiz follows camera horiz (e.g. E4L9)
* otherwise, fall back to default hard-coded parallax implemented as an
angle fraction
git-svn-id: https://svn.eduke32.com/eduke32@3977 1a8010ca-5511-0410-912e-c29ae57300e0
- Rename sky_t members: yscale -> horizfrac, bits -> lognumtiles.
- Add default sky (8 tiles, horizfrac=32768 (i.e. 1/2 the scene horiz), offsets
all zero) and CLOUDYOCEAN sky (8 tiles, horizfrac=65536, offsets all zero)
to multipsky[].
- Get rid of "psky_t g_psky", merely maintaining a g_pskyidx instead. Set it up
at map load time so as to keep the behavior of the legacy per-map psky:
the last sector index with a matching psky ceiling wins.
- In mapstate_t, save g_pskyidx too, not (former) pskybits and pskyoffs[].
- Make on-map-load global psky setup consistent for the game and editor by
factoring it out into common.c: G_SetupGlobalPsky().
- Remove a couple of useless initializations, add some static assertions.
This commit is more likely to introduce subtle differences in behavior.
Specifically, getpsky() now always returns the default sky properties instead of
the global sky ones (but with all-zero offsets) when no match for a suiting
multi-psky is found. This is only likely to affect the yscale/horizfrac of
non-multi-pskies when a global non-default multi-psky has been set up.
Bump BYTEVERSION again.
git-svn-id: https://svn.eduke32.com/eduke32@3976 1a8010ca-5511-0410-912e-c29ae57300e0
- Consolidate psky* arrays into a "typedef struct psky_t" "g_psky" and
"multipsky[]".
- Factor out getting parallaxed sky properties into getpsky().
- Condense initial multi-psky setup by memcpy'ing from multipsky[].
- New function: MultiPsky_TileToIdx().
- Add new define PSKYOFF_MAX and related consistency-checking assertions.
- Lower MAXPSKYTILES to 8 to reflect current usage (was 256).
- Game: make multi-pskies consider dynamically-remapped MOONSKY1, BIGORBIT1
and LA. (Not very useful as the editor will still only act for the static
values -- 80, 84 and 89, respectively.)
An attempt has been made to preserve behavior even in strange cases, so this
commit is unlikely to introduce regressions. Because of point 6, BYTEVERSION
had to be bumped.
git-svn-id: https://svn.eduke32.com/eduke32@3975 1a8010ca-5511-0410-912e-c29ae57300e0
- Allow xoffset and yoffset to modified independently of a tile definition, much like texhitscan and nofullbright.
(Both still default to zero when a tile is specified, to keep current behavior, and because it makes sense.)
- Add actual detection of the "nofullbright" keyword, which appears to have been overlooked in r3230.
- Internal: Eliminate the need for one int32_t by condensing two variables into "flags".
git-svn-id: https://svn.eduke32.com/eduke32@3973 1a8010ca-5511-0410-912e-c29ae57300e0
Also, very slightly tweak a factor toward a "brighter" (farther visible)
scene overall to account for the "circular" nature of the fragment distance
(as opposed to an "ortho" distance in Build).
git-svn-id: https://svn.eduke32.com/eduke32@3962 1a8010ca-5511-0410-912e-c29ae57300e0
So there's no "jump" from vis!=0 to vis=0 making the texture appear brighter.
git-svn-id: https://svn.eduke32.com/eduke32@3959 1a8010ca-5511-0410-912e-c29ae57300e0
That is, for shade of objects, use shade tables. For visibility, use GL fog.
Mixing these two does not produce satisfactory results in areas where both
shade and visibility darkening are high, such as around the E1L1 restroom.
git-svn-id: https://svn.eduke32.com/eduke32@3941 1a8010ca-5511-0410-912e-c29ae57300e0
Having to tag these few variables and functions as ATTRIBUTE((used))
seems sensible, but what follows is just strange. OSD_Printf? headspritesect?
git-svn-id: https://svn.eduke32.com/eduke32@3932 1a8010ca-5511-0410-912e-c29ae57300e0
