Unvanquished ➤ Development activity
Events from:
|
Unvanquished | Friday, 27 January 2023 |
DolceTriade commented on pull request Unvanquished/Unvanquished#2368
DolceTriade
commented on
Unvanquished/Unvanquished#2368
· |
|
necessarily-equal commented on pull request Unvanquished/Unvanquished#2336
Should I go ahead and merge this? |
|
necessarily-equal commented on pull request Unvanquished/Unvanquished#2368
Should I go ahead and merge this? |
|
bmorel commented on issue DaemonEngine/Daemon#788
which seems to be what you are doing That's the whole point of this ticket: I'm not doing anything yet. I want upstream to choose a solution befo… |
|
DolceTriade commented on issue DaemonEngine/Daemon#788
I'm not sure how beneficial it would be to bring glm into daemon... I think implementing some direct proxy between vec and glm at the public interf… |
|
bmorel commented on issue DaemonEngine/Daemon#788
That's what is currently done in unvanquished. This suppose that glm::vec3 have the exact same memory layout and byte size as vec3_t, notably. It's… |
|
DolceTriade commented on issue DaemonEngine/Daemon#788
Is using |
|
Daemon Engine | Thursday, 26 January 2023 |
|
bmorel opened an issue in DaemonEngine/Daemon
stop using `float[2]`, `float[3]`... in the public interface
#788
I have a modified unvanquished codebase which almost got rid of the C |
|
bmorel opened an issue in DaemonEngine/Daemon
drop the `Vec` API
#787
Before gamelogic started switching to glm, it was using a homemade vector API. This is used in few places in daemon's codebase I think, but one mig… |
|
bmorel opened an issue in DaemonEngine/Daemon
erroneous const-ness for some functions
#786
While migrating my codebase toward glm, I identified 3 functions that have wrong const-ness in daemon's public API:
|
|
illwieckz commented on issue Unvanquished/Unvanquished#2186
I created the |
|
slipher commented on pull request DaemonEngine/Daemon#785
You still need to guard |
|
illwieckz created a repository UnvanquishedAssets/res-leveleditor_src.dpkdir
A simple package only meant to be loaded by level editors to load level editor stuff. Updated Jan 26 |
|
illwieckz opened an issue in DaemonEngine/crunch
the help of crunch says "-grayscalsampling" without e
#31
Reported by julius on Xonotic mapping channel: the help of crunch says "-grayscalsampling" when the correct parm seems to be "-grayscalesampling" |
|
illwieckz opened a pull request in UnvanquishedAssets/tex-common_src.dpkdir
basic color image and materials
#4
This is an old pull request of mine I forgot to submit before. First purpose is to provide a set of basic colors. Such basic colors are heavily us…
+408
-28
|
|
slipher commented on pull request DaemonEngine/Daemon#785
It surely breaks Windows. The CI image just has a ton of crap preinstalled on it. |
|
|
Unvanquished | Wednesday, 25 January 2023 |
|
illwieckz closed an issue in Unvanquished/Unvanquished
Map loading screen doesn't disappear after `vid_restart`
#2375
I first noticed it some days ago but I was running an arm build with the very experimental panfrost driver from Mesa git main branch. But I just re…
12 comments
|
|
illwieckz commented on issue Unvanquished/Unvanquished#2375
Said to be fixed by #2382 |
|
illwieckz commented on pull request DaemonEngine/crunch#30
It looks like |
|
illwieckz commented on pull request DaemonEngine/Urcheon#60
I added a commit to use the |
|
illwieckz commented on pull request Unvanquished/Unvanquished#2383
See https://github.com/DaemonEngine/daemonmap/pull/13/files for better cell height adjustment code that was known to work with anthill. |
|
illwieckz pushed to illwieckz/rgb-tga in DaemonEngine/Urcheon
|
|
illwieckz commented on pull request DaemonEngine/crunch#30
with Metro's |
|
illwieckz opened a pull request in DaemonEngine/crunch
crunch: add `-noNormalDetection` to avoid normal map detection
#30
Add
+14
-7
|
|
illwieckz pushed to illwieckz/nonormaltype in DaemonEngine/crunch
|
|
illwieckz commented on pull request DaemonEngine/Urcheon#60
With Metro map textures stripping the opaque alpha channel before calling crunch saves 2% of file size.
Unfortunately |
|
DolceTriade commented on pull request DaemonEngine/Daemon#785
Honestly I thought find_package was calling pkg-config under the hood. |
|
necessarily-equal commented on pull request DaemonEngine/Daemon#785
pkg-config is love, pkg-config is life. I don't know the implications for other platforms, but at least it seems CI works. I know we rely on pkg-co… |
|
DolceTriade commented on pull request DaemonEngine/Daemon#785
Does this have implications on other platforms? Can we do find_package first and then pkg-config after? |
|
DolceTriade deleted branch ishq/vrbug at Unvanquished/Unvanquished
|
|
DolceTriade pushed to 0.54.0/sync in Unvanquished/Unvanquished
|
|
necessarily-equal opened a pull request in DaemonEngine/Daemon
Use pkg-config for curses
#785
This reportedly fixes a compilation error on Gentoo, similar to jonathanklee/ngp#30.
+2
-2
|
|
slipher commented on issue Unvanquished/Unvanquished#2381
I think if things rely on the mapname being set in the cgame, it is a bug. Agreed. Well, I do use it in the cgame for cg_navgenOnLoad, but that c… |
illwieckz pushed to illwieckz/rgb-tga in DaemonEngine/Urcheon
illwieckz
pushed to
DaemonEngine/Urcheon
·
1 commit to
illwieckz/rgb-tga
|
|
illwieckz pushed to illwieckz/rgb-tga in DaemonEngine/Urcheon
illwieckz
pushed to
DaemonEngine/Urcheon
·
1 commit to
illwieckz/rgb-tga
|
|
illwieckz commented on pull request DaemonEngine/Urcheon#60
We may want to also implement this optimization in crunch, but I want Urcheon to take care of this in all cases… |
|
illwieckz commented on pull request DaemonEngine/Urcheon#60
I noticed crunch was producing formats currently unsupported by Dæmon engine with some images:
|
|
illwieckz opened a pull request in DaemonEngine/Urcheon
urcheon: strip alpha channel if fully opaque when converting images with crunch
#60
Crunch doesn't detect opaque alpha channels, so it would convert RGBA tga with opaque alpha channel to DXT5 crn (and may even convert to swizzle fo…
+12
-1
|
|
DolceTriade commented on issue Unvanquished/Unvanquished#2378
Hm, idk what's going on. Even if I set the value by clicking on it, it doesn't stick. As far as I can tell, the datasources are getting their value… |
|
DolceTriade commented on issue Unvanquished/Unvanquished#2378
Ah, this affects specifically |
|
DolceTriade commented on issue Unvanquished/Unvanquished#2378
f5d89ca I think was the fix... Maybe this is another issue |
|
DolceTriade commented on issue Unvanquished/Unvanquished#2378
I could have sworn I fixed this... |
|
DolceTriade pushed to ishq/vrbug in Unvanquished/Unvanquished
DolceTriade
pushed to
Unvanquished/Unvanquished
·
2 commits to
ishq/vrbug
|
|
Unvanquished Assets | Tuesday, 24 January 2023 |
|
illwieckz commented on pull request UnvanquishedAssets/unvanquished_src.dpkdir#172
Before releasing 0.54, we may want to rename |
|
illwieckz opened a pull request in UnvanquishedAssets/unvanquished_src.dpkdir
add option to enable/disable light styles in presets and options
#172
+11
-1
|
|
illwieckz pushed to 0.54.0/sync in Unvanquished/Unvanquished
illwieckz
pushed to
Unvanquished/Unvanquished
·
2 commits to
0.54.0/sync
|
|
illwieckz pushed to 0.54.0/sync in DaemonEngine/Daemon
|
|
illwieckz pushed to master in Unvanquished/Unvanquished
|
|
DolceTriade commented on issue Unvanquished/Unvanquished#2375
I'll try to find some time to dig deeper into this today or tomorrow. |
|
DolceTriade commented on issue Unvanquished/Unvanquished#2381
I think if things rely on the |
|
illwieckz commented on issue Unvanquished/Unvanquished#2375
It would be good to know what kind of issue that change was meant to fix, to know if reverting that change to fix the current issue would bring bac… |
|
illwieckz closed an issue in Unvanquished/Unvanquished
Map loading screen displays incomplete information when connecting to a server
#2376
I probably put the code to fetch the data too early in map load code to get it while connecting to a server that is not itself. Loading a map with … |
|
illwieckz pushed to master in Unvanquished/Unvanquished
|
|
illwieckz merged a pull request in Unvanquished/Unvanquished
cgame: read map name from server config string instead of cvar, fix #2376
#2380
cgame: read map name from server config string instead of cvar, fix #2376 cvar is only reliable when hosting map with the client server config str…
+7
-1
4 comments
|
|
illwieckz commented on pull request Unvanquished/Unvanquished#2380
I see no other way to implement this since |
|
illwieckz commented on pull request Unvanquished/Unvanquished#2380
Agree, if we need to read the cvar earlier, we should set it earlier. @DolceTriade The cvar is never set at all when joining a server. I don't kn… |
|
illwieckz opened an issue in Unvanquished/Unvanquished
`mapname` cvar is never set when joining a server, is it a bug or not?
#2381
I'm not sure this is a bug, but |
|
|
Unvanquished | Monday, 23 January 2023 |
|
illwieckz commented on issue Unvanquished/Unvanquished#2265
Improved cache invalidation (https://github.com/slipher/Unvanquished/commits/navgen53) Do we need that for the release? |
|
illwieckz commented on issue Unvanquished/Unvanquished#2343
This issue is less visible after DaemonEngine/Daemon#784 was merged, but we can still reproduce it with weapons like the pulse rifle: |
|
illwieckz commented on issue Unvanquished/Unvanquished#2351
It is believed to be fixed by DaemonEngine/Daemon#784 , please reopen if you reproduce it after merging DaemonEngine/Daemon#784 . |
|
illwieckz closed an issue in Unvanquished/Unvanquished
Dynamic lights too bright on Nvidia
#2351
On Nvidia that the dynamic lights are way too bright. Here's a single blaster shot: GPU: NVIDIA GeForce GTX 1050 Ti
2 comments
|
|
illwieckz merged a pull request in UnvanquishedAssets/res-weapons_src.dpkdir
particles, trails: dim dinamic lights a bit (divide by two) since all of them are now rendered
#13
particles, trails: dim dinamic lights a bit (divide by two) since all of them are now rendered See DaemonEngine/Daemon#784
+9
-9
1 comment
|
|
illwieckz merged a pull request in UnvanquishedAssets/res-buildables_src.dpkdir
particles, trails: dim dinamic lights a bit (divide by two) since all of them are now rendered
#10
particles, trails: dim dinamic lights a bit (divide by two) since all of them are now rendered See DaemonEngine/Daemon#784
+1
-1
|
|
slipher merged a pull request in DaemonEngine/Daemon
Render all dynamic lights, not a random 25%
#784
The dynamic lights are evenly divided into four "layers". Due to the typo of 'return' instead of 'break' (unnoticed since the original tiled render…
+1
-7
7 comments
|
|
slipher pushed to 0.54.0/sync in DaemonEngine/Daemon
|
|
illwieckz commented on pull request DaemonEngine/Daemon#784
The same code is present at the end of the function. There is no longer an early return so it does not need to be duplicated. Oh, so that even me… |
|
slipher commented on pull request DaemonEngine/Daemon#784
@slipher what and why is the other code removed? It looks like some debug code but I don't know more about it… The same code is present at the en… |
|
slipher commented on pull request Unvanquished/Unvanquished#2377
The idea is to put the most used commands at the beginning rather than alphabetize |
|
slipher commented on pull request Unvanquished/Unvanquished#2377
Should this be ADMBP? |
|
slipher commented on issue Unvanquished/Unvanquished#2265
The main server-side navmesh generation was merged: #2353. What remains to have this released is:
Enable cvars for navmesh generation by default ( |
|
slipher commented on pull request Unvanquished/Unvanquished#2369
Let's go with replacing ATTACK_RANGE with ROCKETPOD_RANGE everywhere, it sucks to have multiple names for the same constant anyway. |
|
|
Daemon Engine | Sunday, 22 January 2023 |
|
slipher pushed to master in DaemonEngine/Daemon
|
|
illwieckz commented on issue Unvanquished/Unvanquished#2375
Why is it stuck in a while loop in the first place? I have no idea, the trace isn't very talkative about it when interrupt the game… |
|
DolceTriade commented on issue Unvanquished/Unvanquished#2375
Hm, not entirely sure how this fixes it... Why is it stuck in a while loop in the first place? |
|
DolceTriade commented on pull request Unvanquished/Unvanquished#2380
Agree, if we need to read the cvar earlier, we should set it earlier. |
|
|
Unvanquished | Saturday, 21 January 2023 |
|
illwieckz commented on pull request Unvanquished/Unvanquished#2380
For reference the code to display the levelshot also uses |
|
illwieckz pushed to illwieckz/map-loading-map-name in Unvanquished/Unvanquished
illwieckz
pushed to
Unvanquished/Unvanquished
·
1 commit to
illwieckz/map-loading-map-name
|
|
illwieckz commented on pull request Unvanquished/Unvanquished#2380
Note: The fact the cvar value is not set when the client is not also a server may be a bug, what do you think about it? |
|
illwieckz commented on pull request Unvanquished/Unvanquished#2380
The map loading screen is painted before |
|
illwieckz pushed to illwieckz/map-loading-map-name in Unvanquished/Unvanquished
illwieckz
pushed to
Unvanquished/Unvanquished
·
1 commit to
illwieckz/map-loading-map-name
|
|
illwieckz commented on issue Unvanquished/Unvanquished#2375
@DolceTriade Doing this fixes the bug, it's a partial revert of e7f47fa. diff --git a/src/cgame/cg_rocket.cpp b/src/cgame/cg_rocket.cpp index d8e71… |
|
illwieckz commented on issue Unvanquished/Unvanquished#2375
Restoring the wonky switch doesn't fix the bug. |
|
illwieckz commented on issue Unvanquished/Unvanquished#2375
@DolceTriade maybe the “refactor the switch statement to be more readable instead of relying on wonky fallthrough” from 7f47faab2cfa6c87df8aaf17685… |
made with Pluto planet software