This is the first step towards component-based entities.
There's still some transform-related stuff in the struct that needs to
be moved, but it's all entirely client related (rather than renderer)
and will probably go into a "client" component. Also, the current
components are directly included structs rather than references as I
didn't want to deal with the object management at this stage.
As part of the process (because transforms use simd) this also starts
the process of moving QF to using simd for vectors and matrices. There's
now a mess of simd and sisd code mixed together, but it works
surprisingly well together.
This is a big step towards a cleaner api. The struct reference in
model_t really should be a pointer, but bsp submodel(?) loading messed
that up, though that's just a matter of taking more care in the loading
code. It seems sensible to make that a separate step.
It now takes a context pointer (opaque data) that holds the buffers it
uses for the temporary strings. If the context pointer is null, a static
context is used (making those uses of va NOT thread-safe). Most calls to
va use the static context, but all such calls have been formatted
consistently so they are easy to find when it comes time to do a full
audit.
There's still some cleanup to do, but everything seems to be working
nicely: `make -j` works, `make distcheck` passes. There is probably
plenty of bitrot in the package directories (RPM, debian), though.
The vc project files have been removed since those versions are way out
of date and quakeforge is pretty much dependent on gcc now anyway.
Most of the old Makefile.am files are now Makemodule.am. This should
allow for new Makefile.am files that allow local building (to be added
on an as-needed bases). The current remaining Makefile.am files are for
standalone sub-projects.a
The installable bins are currently built in the top-level build
directory. This may change if the clutter gets to be too much.
While this does make a noticeable difference in build times, the main
reason for the switch was to take care of the growing dependency issues:
now it's possible to build tools for code generation (eg, using qfcc and
ruamoko programs for code-gen).
Only as scalars, I still need to think about what to do for vectors and
quaternions due to param size issues. Also, doubles are not yet
guaranteed to be correctly aligned.
I added Sys_RegisterShutdown years ago and never really did anything
with it: now any system that needs to be shutdown can ensure it gets
shutdown on program exit, and in the correct order (ie, reverse to init
order).
I've decided that setting pr.max_edicts and pr.zone_size as part of the
local progs initialization rather than in PR_LoadProgsFile makes more
sense. For one, it is unlikely for the limits to change every time progs is
reloaded. Also, they seem to be a property of the VM rather than the progs.
However, there is nothing stopping the caller from updating max_edicts and
zone_size every call.
These are the ones where I could easily make scan-build happy. They do seem
to be potential holes where invalid data in one place could result in use
of uninitialized values.
While scan-build wasn't what I was looking for, it has proven useful
anyway: many of the sizeof errors were just noise, but a few were actual
bugs (allocating too much or too little memory).
The search for these files will stop in the vpath that contains the .bsp
file to which they belong. This will prevent problems with
id1/maps/start.lit being used for shadows/maps/start.bsp.
This is based on ccr's patch, but a little more thorough (he missed some
potential problems) and probably more readable (the original code wasn't
the greatest (still isn't, but...))..
I'd forgotten that ED_ConvertToPlist mangled light into light_lev and
single component angle values into a vector. This fixes much of the
breakage in qflight (but not the light levels)
The data needed for the checks has been gone from QF for a very long time
(since Jan 2000), and good riddance to it, really: I suspect it's in
violation of id's copyright on the game data (ok, it might be fair use, but
still...).
Something is funny with Ubuntu such that -ldl needs to be specifically
added even though QFutil's .la specifies it. I don't know if it's a libtool
issue or not, but this does work.
More will probably be necessary, but this was sufficient to get prover to
the point where qfcc segged building qwaq (0.7.2).
hipdemo1.dem has a single leading space. They probably did "%2d" for the
track number, as there is no space before the minus sign in the other
hipnotic demos.
Now the user can create and destroy IMTs at will, though currently
destroying IMTs is currently all or nothing (imt_drop_all).
An IMT is created via imt_create which takes the keydest name (key_game
etc), the name of the IMT (must be unique for all IMTs) and optionally the
name of the IMT to which the key binding search will fall back if there is
no binding in the current IMT, but must be already defined and on the same
keydest. This means that IMTs now have user determined fallback paths. The
requirements for the fallback IMT prevent loops and other weird behaviour.
Actual key binding via in_bind is unaffected. This is why the IMT name must
be unique across all IMTs.
The "imt" command works with the key_game keydest, but imt_keydest is
provided for specifying the active IMT for a specific keydest.
At startup, default IMTs are setup to emulate the previous static IMTs so
old configs will continue to work (mostly). New config files will be
written with commands to drop all of the current IMTs and build new ones,
with the bindings and active IMT set as well.
This fixes the status bar refresh issues in sw. The problem was that with
two viddef's hanging around, things got a little confused and recalc_refdef
wasn't getting into the renderer.
This fixes the flickering status bar for those renderers that use
scr_copyeverything correctly. sw and sw32. glsl flukes out by not touching
scr_copyeverything and gl is wrong by always clearing it to 0 and never
setting it later.
The cshift is now drawn before centerprint, console and the loading plaque.
Unfortunately, due to something odd going on, the sbar stuff is drawn by
the console rather than by Sbar_Draw :P.
Need to up the precision by one due to the difference between g and e, but
much prettier. Might need to rename that function :P I wish I'd thought to
check if g would work, but thanks to divVerent for the suggestion.
This fixes the infinite loop in Sock's shadows mode. Thanks to Spike for
helping me with what should be happening. However, we're still uncertain
about just how the thinktime/sv.time logic should behave.
It turns out gcc has a way to force functions to inline even when it thinks
doing so would not be a good idea (call to a modest sized function unlikely).
F6 is fantastic, until you hit it by mistake after dieing when you meant
to hit F9 (I've done that way too often). quick.sav is still the last file
written via F6 (so F9 is unaffected), but now the previous quick.sav
becomes quick1.sav. Up to 5 (currently) backups will be kept: quick1 is
the newest, quick5 the oldest. A menu for accessing the backups has been
added as a sub-menu of the load menu.
This fixes the broken intermissions. There might be a better fix, but this
will do for now because the one alternative solution (getting
V_CalcIntermissionRefdef to poke directly at entity states) is not
something I particularly want to do.
Note that this is only a soft-limit and prevents excessive beams when only
one is expected. If more beams are desirned, just use 0 for the entity
number.
Really, when cl_nodelta is in effect (eg, .qwd demo recording and thus
playback). QW now uses the new shared entity state block as I'd intended.
Thanks to the cleanup of ghost entities (ie, entities that have been
removed but continue to be rendered), glsl overkill has gone from 157 to
163 fps :)
It was a hack to help with cleaning up the renderer, but is now in the way
of merging the clients.
This happens to fix the position/angle lerping, though angles behave a
little oddly.
CL_StopPlayback was incorrectly calling Key_SetKeyDest with key_game. As
CL_SetState calls Key_SetKeyDest with the appropriate parameter,
CL_StopPlayback calling Key_SetKeyDest is actually redundate, so just
change key_game_target back to IMT_0.
NQ's client-side onground now works similarly to QW's: -1 = not on ground,
0+ = on ground (plane num?) but only 0 is used in NQ as the necessary
information is not sent over the network.
The protocol is not affected: the incoming colors byte is split into the
two fields when read from the packet. This simplifies a lot of code
elsewhere.
When the stufftext buffering was implemented, we forgot about
svc_serverdata...
While nq might normally not have the issue, I thought it best to implement
in nq, too.
nq now rejects anything that doesn't look like a .dem, and qw not only
rejects anything that looks like neither a .qwd nor a .mvd, but even
detects which one is being played by the contents rather than the file name
(foo.mvd.gz would be mis-detected as a qwd!!).
The change to nq SV_RunThink to match that in qw finally bit my bum. I
suspect it's really a bug in the honey progs (using ltime on a non-pusher),
but QF locking up is not very friendly. I want to keep the option of having
a think loop in the same frame for progs that expect it (using
self.nextthink = time), but also NQ's original method of running think
every frame. Hopefully no progs expect "self.nextthink = time" to run
every frame rather than loop within the current frame.
Now it doesn't matter if you get 22 fps or 72, you jump the same height,
which actually happens to be slightly higher than the previous 72fps jump.
Effectively, you jump the height you would if you got infinite fps ;)
The renderer now gets initialized and things sort of work (qw-client will
idle, though nothing is displayed). However, as the viddef stuff is broken,
it segs on trying to run the overkill demo.
Still, nothing will work: no plugins are loaded and they're all broken
anyway.
glx, sgl, glslx etc are going away, just the basics will be built: fbdev
(probably go away eventually), sdl, x11 and hopefully someday win. That's
actually the only reason anything links.
Just about to do a release, and I realized windows users wouldn't have any
way of checking out the new renderer. I'll add wglsl when I get a chance to
do some testing.
Most subsystems that depend on other subsystems now call the init functions
themselves. This makes for much cleaner client initialization (more work
needs to be done for the server).
The renderer should now be free of any direct access to client code. Even
3d rendering is now done via a function pointer.
The cshift code is done as a 2d screen function.
This brings NQ's physics a little closer to QW's. After studying both the
original WinQuake source and the progs source, this change should be
harmless, making very little difference. However, it does allow an
entity's think function to be called multiple times in the same frame (for
when the entity needs more time to think, but it would cause runaway loop
errors). Maybe need a "runaway think" check.
This has several benifits:
o The silly issue with alias model pitches being backwards is kept out
of the renderer (it's a quakec thing: entites do their pitch
backwards, but originally, only alias models were rotated. Hipnotic
did brush entity rotations in the correct direction).
o Angle to frame vector conversions are done only when the entity's
angles vector changes, rather than every frame. This avoids a lot of
unnecessary trig function calls.
o Once transformed, an entity's frame vectors are always available.
However, the vectors are left handed rather than right handed (ie,
forward/left/up instead of forward/right/up): just a matter of
watching the sign. This avoids even more trig calls (flag models in
qw).
o This paves the way for merging brush entity surface rendering with the
world model surface rendering (the actual goal of this patch).
o This also paves the way for using quaternions to represent entity
orientation, as that would be a protocol change.
There are some problems with menus and the console messing up the key_dest
state (they assume console/menu or game, nothing else), but otherwise
things seem to work.
I got rather tired of there being multiple definitions of mostly compatible
plane types (and I need a common type anyway). dplane_t still exists for
now because I want to be careful when messing with the actual bsp format.
Currently only "id", "hipnotic" and "rogue" are supported (anything else is
treated as "id"). Has no effect in quakeworld (good thing too: changing
gamedirs is a little broken).
If a tossed entity has ONGROUND set, it won't move, but if it has an upward
velocity, it can't possibly be ONGROUND, so clear the flag if the entity is
moving upwards.
This also makes functionality consistent across the platforms, such as
adding support for -dedicated to sdl based nq, and various timing
calculations are now consistent.
It seems qsockaddr's assumptions aren't necessarily portable, as OpenBSD
seems to be doing weird things with qsa_family. Even if that's not the
case, this is cleaner.
See http://forums.inside3d.com/viewtopic.php?t=1675 for inspiration.
As this might break the gameplay of certain maps, the new sv_jump_any
cvar can be used to disable this feature (bringing back the old behavior).
in_dga -> 0 (until X is fixed)
gl_multitexture -> 1 (why was this 0? not enough support back then?)
cl_usleep -> 1 (seems to be ok)
host_mem_size -> 40 (even 32 isn't enough these days)
rate -> 10000 (we're not in the modem era any more)
cl_mem_size -> 32 (16 is not enough, 32 sounds better than 24)
This is based on the code presented on
http://forums.inside3d.com/viewtopic.php?t=2376 (and updates linked from
that thread), but with modifications for hopefully supporting rotating
trains (and not duplicating big fuctions).
For now, only nq. qw will come after some more testing.
This fixes the hang after a few (3 or 4) play/timedemo runs of bigass1. Now
23 runs (and counting) seems to work well, though there seems to be a
memory leak somewhere.
nq's SV_CheckVelocity will be a little different (spherical rather than
cubical), but sv_maxvelocity defaults high enough for it to not matter to
most players. It might even improve play.
The remaining differences need some careful thought before the physics
merge is done.
