Ah I'm confused because i thought the question sounded like "AD run with the spiked exe released for AD has particles, but if you replace with newer version of spiked exe no particles appear".
That's how I took the original situation, but if I'm totally wrong then disregard. Probably my crap reading comprehension.
oh LMAO, sorry just read the original post #196 and he's talking about them not appearing in vanilla.
So yeah ignore me.
I'm playing the Xmas Jam 2018 with Quakespasm-Spiked-ADMod.exe and i'm getting this error with both Scampie's and Strideh's maps:
Host_Error: Illegible Server Message 119, previous was svc_serverinfo
Anyone knows what's happening? :/
Spile will know better what that is but try updating to the latest build:
The qss-admod was a one-off build I made with a few patches to Spike's engine when AD came out but it hasn't been bug fixed since.
Thank you very much, sir!! Now it's working =D
I didn't knew that you guys have been updating Quakespasm-Spiked =D
It's me again. Sorry for the double post...
One thing that i noticed when i was testing xmas jam with another source-ports is that the start map have snow falling from the sky in quakespasm and markV, but not in quakespasm-spiked...
i opened the start.map file in trenchbroom, and it seems the snow is made by something called "misc_particletemplate"... quakespasm-spiked doesn't understand this feature?
set pr_checkextension 0 if you want mods (like AD) to treat qss like they would quakespasm.
But when i do this i don't have those cools effects (smoke, sparks, somke, etc) from quakespasm-spiked anymore :P
i was also trying to make a *.ent file for QMaster's map "Gift of the Boxlord" (to make snow in the map) but i noticed that the *.bsp already had a line where it says:
but there is no snow falling, even when i write r_particledesc fte_weather or r_particledesc weather on the console :(
How to disable the custom particle effects? I've tried all cvars I could think of, and nothing. Specifically, I'd like to disable the custom TE_GUNSHOT particles, because lots of 8-bit alpha overdraw makes the engine too slow when up close.
Also, I wonder if Spike could fix the crash when FSAA is enabled on Intel HD Graphics 3000 IGPs.
Ok, I've just re-read the thread and will try a couple ways to disable the custom particles.
I cannot fix your intel drivers for you.
that said, I do wonder if its caused by the use of glCopyTexSubImage. Either way its not something I can test/debug/fix for you.
particle configs can get loaded a few ways:
non-default r_particledesc cvar setting
worldspawn _texpart_* keys
qc particleeffectnum() calls that include a full-stop separator.
chances are the mod is doing it explicitly, which implies you need to mod your mod to disable it.
(assuming this is AD you're using, you'd need to open effectinfo.txt and strip out the TE_ effects that are generally at the start of the file - sock used a different prefix for extended effects, and if you strip his new effects then there's no fallback and you'll see less particles there. Or you can just set pr_checkextension 0 to prevent the mod from realising that it can use those particle effects thus preventing it from forcing them to load.).
When I set r_particledesc to "haze.cfg" on a map with torches it hardlocks on r10. I've booted up into start and e1m1 with this active and it has froze every time.
Many maps have different textures with same names. To load correct HD texture for current map it placed into textures\map_name subfolder. QSS even in this case loads both textures: from each textures\map_name\ and textures\ folders. And if texture in textures\ folder has _luma. file it looks especially bright.
It looks like this (on Forgotten Sepulcher):
I found elegant decision - just put in textures\map_name\ subfolder entirely black _luma. file for current texture. It solve problem.
It is a bug or its a feature?
That's definitely a bug. If loading from textures/mapname/texturename, the engine should ignore textures/texturename_luma.
Now I don't remember if that bug also happens in my own engine...
RMQ Winter 2011
If anyone remembers the old RMQ project,the e2m1rq map in the Winter 2011 release is a bit crazy wit hthat opening cutscene.
In Mark V that cutscene plays at normal speed,showing the grunts and marine fight and enforcers teleoprting at normal (and intended) speed.
QSS plays the cutscene at hyper-speed.
The second cutscene plays after maybe 5 seconds in Mark V
(the one with they shall not contain thee)
and QSS plays the cutscene a little later (8 secs?) but at the right speed.
RMQ Winter 2011
will not fix - its a bug in the rmq qc, not qss (or fte for that matter).
you can work around it by setting pr_checkextensions to 0, giving you the same result as (unmodified)qs or markv.
specifically, the qc has different behaviour between:
a) no extensions (read: qs expected)
b) extensions+DP_SV_CLIENTCAMERA (read: dp expected)
c) extensions without DP_SV_CLIENTCAMERA (read: they didn't bother to test that pathway properly, and its bugged).
The DP_SV_CLIENTCAMERA extension is designed as a simple wrapper around svc_setview (except that it also switches the pvs too). rmq doesn't even use it, so I don't know why its bothering to check for it.
How to use external replacement conback, gfx.wad and menu images?
The source seems to support it, at least for menus, but I'm having no luck.
I don't know where else to report this but there's a bug with one of the xmasjam2018 maps. In spipper's map "A Christmas Romp," entering the map and saving causes QSS to crash immediately. I don't have the same problem in regular QS.
This a patched in one of the last commits in QS (it's a buffer overflow with keyvalue length)
Does QSS have particle limits or any sort of limits when it comes to fancy effects? For mods that take advantage of it like AD, it's my preferred port of choice but there are moments when it slows down to under 1 FPS. Granted my computer has some pretty old parts in it but I feel like the Quake engine shouldn't be that difficult to run even with some fancy effects.
I had performance problems in Leptis Magna and also in ad_azad during the final boss fight. During the final boss's death animation, I was worried that the game might crash. Usual when I get to this point in a map, I have to save and reload in regular QS.
you mean like the r_part_maxparticles or r_particle_tracelimit cvars?
if its the rain/snow/etc effects, you can just set r_part_rain_quantity 0 (or really any value lower than its default).
I don't know what's actually causing the slowdown and why it's so drastic but I'll try changing those cvars. None of the maps in which I'm having performance problems have any weather effects so it's something else causing issues.
Try hiding the world with r_drawworld 0 and see what is rendering and what the fps is. IIRC all torches in the map render in AD+QSS
You're right in that all the torches get rendered in AD. I think everything with a particle effect on it essentially gets rendered from torches to armor.
Using the boss fight in ad_azad as a benchmark, QSS grinds to a freeze with the default settings. With r_part_maxparticles set at 16384 down from the default 65536, the fight is actually pretty playable. It still drops well below 60fps though. I wonder if it's just my CPU - using a first generation i7 at 2.67 GHz.
I tried ad_azad in FTE with the Quakespasm preset and the game doesn't slow down at all for me in ad_azad (general FTE slowdown issues aside). I get a consistently smooth frame rate all throughout the boss fight whereas it freezes for me at certain parts in QSS. I don't get it.