#Team Comtress 2 Megathread - It's finally here
1 messages · Page 8 of 1
it doesnt get announced nor is there an easy way to know information about the enemy
information is easily available in dota and thats kind of the whole gameplay
dota favors being aware of information and making the right decisions for it
you cant plan your entire game in tf2 unlike in dota
youre not really locked in on a class or any of their weapons, being able to freely swap makes things difficult
but yea i totally agree with that lol
btw did you see how i implemented main menu music in hud code all to resize the main menu dimmer when you click the casual button
I don't remember if I showed you
why did that get deleted
wtf
who is deleting that
bruh
BRUH
i hatr thisd erver....
vgui is the best thing ever because it gives me something super wasteful and pointless to do in my free time instead of actually learning something useful like java
Are you a hud dev?
yeah
i havent made any but ive contributed to a few and found a lot of tech in editing vgui
I think the biggest draw of vgui is that it’s extremely easy to understand on a basic level. If you have a basic understanding of grids and English vocabulary you can do 98% of edits you’d ever want.
The only real stumbling block is that hudlayout.res is a parent of like 50% of things, and if you don’t realize that then you just won’t understand why your stuff only moves a bit before getting cut off. Which is hard to explain concisely, but relatively digestible to pick up on.
The biggest drawback of vgui is the remain 2% of hud edits are impossible via intended means which is why so much of advance hud editing is basically hacking stuff together. If we were just content saying things like “getting rid of dashboard dimmer is impossible” then vgui would be extremely simple.
I think it’s kinda proposing scripting with convars be replaced with just writing your own code because convars aren’t applicable and have a learning curve for people that already know C++.
It’s not incorrect but I feel like for the intended purpose, the current implementation is fine. It’s just when people start doing things that weren’t intended is where stuff gets complicated and hacky.
Not necessarily saying you shouldn’t replace VGUI, just that I think interpreting the hacky stuff like 10 technical hud developers do as “having to use stupid hacks to make a menu” is a bit off from what vgui was intended for IMO.
I probably wouldn’t go through the effort of porting any of my huds to a new system. If I wanted to practice building interfaces in JavaScript or something I’d probably just get around to making my portfolio website.
But if hud editing had always made use of JavaScript or whatever more useful languages Panorama used, then I would have learned those things and made my huds using them and it would have been beneficial to my education and career.
you can also program websites in english, there are plenty of frameworks that convert markdown to html
here's the deal
horizontal health bars
I mean yeh, but when I make a portfolio website I want it to be practice for something useful.
Health bars in any direction. They’re in VGUI, same with ammo bars. They’re just still in the HL2 implementation so everything is based off 100 iirc, which is a useless number in TF2
You could make VGUI not hacky by just adding a parameter to allow for hud devs to actually make use of HL2 elements, or just implementing those HL2 elements for TF2.
I don’t think an entirely new system is needed to make VGUI not hacky.
But that doesn’t change the fact that it’s still be proprietary, I think that’d be the main purpose behind switching from VGUI. But then in making your UI code not industry proprietary, you’re making it proprietary relative to TF2, so it’s give and take. You either alienate yourself from the general UI implementation ecosystem by sticking to VGUI or you alienate yourself from the existing TF2 ecosystem by switching away from VGUI
that's not the point
IMO, unless the mod becomes big enough to rival base TF2, abandoning VGUI would basically have a similar effect as disabling hud mods, people just wouldn’t make HUDs for a mod.
Just look at how many custom HUDs exist for TFC, I think there’s 1 or 2, and that’s with HUDs being able to be ported directly.
I’m making my own points here.
- Like I think the aspiration of switching to a nonproprietary system is valid.
- Viewing VGUI as hacky is generally not suuuuper fair IMO. Though I feel like I’m also one of the hackier hud devs so maybe I’m the wrong person to speak on this.
- HUD devs are lazy and are going to stick to TF2 and mods that can be directly ported or at best ported with some tweaks. But the farther you get from direct portability, the more traction the mod would need for any hud dev to bother.
Doing that is still far less flexible than something like HTML + JS, since ultimately you're stuck with whatever behaviour the existing panels give you.
peaches it will be okay
if push comes to shove we can just force a random guy to make a ReactVGUI library so I can make all of my websites in VGUI
also vgui is kinda horrible to work with on the c++ side 
ive decided that i actually would like to use html + css + js instead of vgui because when you're using vgui you're playing by valve's rules and you have to do things the specific way valve wants you to do them, there's no tooling, there's no flexibility. you have to use valve's data format you can't run your own code or do like data gen or whatever. when you're working with a web stack you can use what ever tools that you want.
I just hope that all of the files are editable unlike what minecraft did when they moved away from their old proprietary ui on bedrock they made everything unable to be modified by resource packs
I mean yeh, but the main thing is why would I bother?
I think the project is cool but at the end of the day I’m not going to spend hundreds of hours writing a hud for a mod that I’ll play out of interest when I am bored of the main branch of TF2.
the point isn't to serve you
the point is to make it like more maintainable and easier to work with for mastercoms and for new hud developers who want to make their hud just for team comtress 2
mastercoms is a web developer it would be easiest to make their mod if they used familiar technologies rather than buggy archaic broken ass vgui
vgui is super hacky
there is no scripting which really sucks you have to do all events and shit in the game code
more people will be able to make a hud mod than ever before it will be a lot more accessible
Im not saying TC2 should serve me, I’m getting some HUD dev representation in this conversation and just expressing my views as someone who chose to invest in learning hud development for the main branch of the game and has ported my hud to TC2 for personal use.
TC2 doesn’t need to serve me but I think if there’s a discussion on transitioning away from VGUI for hud development then there needs to be a discussion on who it serves. If it’s strictly for maintaining the UI of the TC2 default hud then yeh whatever mcoms find easiest is the best.
If it’s for building a base of community made huds then frankly I think TC2 benefits from nearly direct compatibility with main branch TF2. Because at the end of the day people are going to make huds for what they play, and no offense to the project, but I don’t see it overshadowing TF2 in popularity, nor do I think that’s the aim of the project. So you already have a large base of community huds, that currently can be ported over with like 15 minutes of work, which is more than reasonable for the current scope of the project. As the project grows and establishes itself from base TF2 I think it’ll be reasonable to have some more time investment from porting over, but if it’s a totally unique thing then I just don’t think that’ll happen.
I don’t think this is true
Web development is the largest area in programming most programmers would be able to pick up hud developing INSTANTLY
you don't need programming experience to write html
and you're telling me there's more people in the huds.tf questions than web developers?
"no offense to the project, but I don’t see it overshadowing TF2 in popularity, nor do I think that’s the aim of the project. "
I am unsure about that, a version of TF2 that allows you to use your regular TF2 cosmetics whilst having significantly better gameplay through massive QoL changes is a pretty compelling argument.
I think VGUI at a basic level is more digestible than HTML
No but that have overlap with the TF2 community and the custom hud community and the desire to make or edit their own hud?
yeah but with web tech you don't have to directly touch html, you can use any number of libraries to make it easily data driven just like vgui
And honestly I don’t think a web developer is going to be intimidated by VGUI
Would that be similar to CriticalFlaw's HUD maker? Front end UI to help you modify pre-existing HUDs without touching raw code?
Again. Someone asking how you move the health number down isn’t going to set up libraries.
how would that be much harder with html
no but something like that would be able to be added to the huds themselves or a new version of the hud maker would be able to be made with much more capabilities
So a hell of a lot friendlier to less experienced coders.
*So a hell of a
lot friendlier to less
experienced coders.*
No it’s probably just mean instead of having to write your own code you could do stuff like```
Create JPanel.List “…”
List.Xpos “…”
That’s just some psuedo code. But basically you could use a library to dumb down JavaScript or HTML to the level of VGUI.
HTML is super easy, it's way easier than VGUI
plus you can look stuff up way easier
also
im not removing vgui
for now, this is going to be the main menu only
idk if ill ever replace all of the game HUD, that seems a bit pointless
also, it's just going to be a single VGUI panel
that renders the main menu UI
it wont break any huds
they can just have their VGUI main menu override
but if they want to opt in, they can use the new single panel
it will be so dope and swagger thanks mastercoms
will it use steam's cef can it use steam's cef libraries so you dont have to update it constantly to prevent huds from hacking youre computor
either way im TOTALLY making a hud that embeds a monero miner
automatic hud developer donation
if u need help from me to port vgui im probably not your guy i dont know web development very well
that wasn't for you 
but i agree, we shouldn't get rid of vgui for now
will it use the html motd renderer or does the html motd renderer SUCK?
SUCK
DAMN
Yeh I figured. I just saw that so I was like “oop better hop in”
*Yeh I figured. I
just saw that so I was like
“oop better hop in”*
We gotta stop that thing
I thought you were playing this game in geforce now cuz of the menu lol
Too bad I'm gonna
truly amazing how everytime i work with vgui i find out its worse than I could've ever imagined
I was hoping to find a way to make global conditionals (i.e. is_mvm, is_comp, but not limited to specific files), but as far as I can tell without changing how VGUI works a lot of the functions needed to do that are too low down, set to private, etc.
could do it manually i guess
no it's simple
how so
LoadControlSettings requires a res file input
as far as I can tell there's no way to get the res file programatically from the panel itself
everytime LoadControlSettings is used its with a hardcoded res file path
thats why things are limited to specific res files in the first place
if you open build mode you can see the path IIRC
im not sure why that would matter anyways...
because it requires a .res input?
yes so whenever a res file loads for vgui it uses this function which you can add more conditionals to im not sure what the problem is
its not whenever a res file loads, its that specific hud classes call LoadControlSettings from their ApplySchemeSettings overload with the hardcoded .res path of that specific hud file
im not sure what you're getting at
in the case of a panel that can be used in any res file, it can't have a hardcoded path
and I can't find anyway to get the path
you asked for global conditionals you should just be able to just edit loadcontrolsettings to add global conditionals
it's not actually compiled by tf2's project, so the only real way you can figure out if its tf2 is this, which is stupid
Ideally I think it would be good if, say, the gamemodes themselves (or anything for that matter) could send a signal down to activate a condition, but I'm not really quite sure how to do that yet
is -game tf even default anymore
the mac version of tf2 is the only true version of tf2
thats true, it has debug symbols!
I wonder if it would still work if you fix the sendtables or if there's other issues
don't really care enough to spend time on it
hm actually I might be able to figure out how to do this globally
Yeah -game tf gets automatically appended to the commandline internally
even if you launch without it
you can check by running a timedemo and then checking sourcebench.csv, the output includes the full commandline that the game was launched with
do you think valve would mind if I edited a VPC file to add a public header file :)
specifically, string_t.h
?
whats confusing
.
yeah but what about that is confusing
*VGUI Controls doesn't
import string_t, and I'd like
to compare a string*
can't you just #include it?
doesn't seem like it
#include "string_t.h"
I may be stupid, but that also didn't fix it
just to clarify, when i say it doesn't import it I mean that its not part of the VPC includes
if that makes sense
like this whole list
at least for this project, its used in several other ones including client and server
i mean if it works
which files were you trying to include it from? i included it from vgui2/vgui_controls/BuildGroup.cpp and it worked without modifying the vpc file
I was importing it from the BuildGroup header file instead of the cpp file
idk i'll try it again
setting up the conditions themselves is more complicated than I thought
I forgot they had a hudtype variable that overwrites whatever hud is usually used for a gamemode
give spy a ping
could you pretty please make ctapping consistent on even slightly slanted surfaces
Was that ctapping? Normally it works fine on flat and down slopes, its only an issue on up slopes iirc
I was walking slightly up slope
One thing I'd want to see fix would be the class selection screen, it has a lot of issues like the animations not playing correctly, the facial animations not resetting when surfing the class models, broken lightning and missing sound effects. It's been 'bugging me' for as long as Valve broke it.
Maybe it's skill issue? 
patch ctap and make it a feature instead of a bug
ctap is a bug?
yeah
i propose crounching becomes an uninterruptible animation that onky works on the ground. Two bird with one stone, no more crouchjump exploiting either
Ficool already made a fix for that screen, it just hasn't been implemented and added in the game yet.
is it possible to do game state hand off in the current engine?
*is it possible
to do game state hand off in
the current engine?*
like, solely with game code and some commands
for a steam networking listen server, could the host upon disconnect, hand off the server state to a client to rehost and redirect to?
if so, then we could easily make a scalable matchmaking system
couldn't you use the redirect command for that?
ofc, there's still the issue of unclean exits but it's an improvement
yes thats the easy part
what i mean is setting up the same game state
all entities and game rules preserve state
ohhhh
given to an ideal client as a keyvalues or some more efficient blob
doesnt gmod do that?
how does the singleplayer saverestore system work?
even with engine access it seems like it would be a hell of a lot of work
since you'd have to check and annotate like every entity at the very least
plus theres extra game state in globals, physics state, etc
when will valve release the engine source code 
regarding global conditions in VGUI, I have a few questions regarding what people want feature-wise
For a bit of context, TF2 decides which gamemode hud to use through one of 2 ways, the gamemode of the map (usually determined by presence of a logic entity, although if its tc or sd they just use the prefix in the map filename lmao), and by a "hudtype" variable, which maps can use to override the default hud for that gamemode.
So for example, a map that has CTF logic will automatically use the CTF hud, but then the map creator can override that to force, say, the CP hud instead.
Valve uses this for training mode (or maybe planned it for the more indepth scrapped training maps) to force the training HUD instead of any other gamemode HUD.
So my question is: should the hudtype variable also overwrite the global conditions, or should all the conditions for the current scenario be active (i.e. CTF, CP, and training conditions all enabled at the same time)
This won't change how specific hud elements are enabled, just the conditions for other, already enabled elements
my thought process is that the way valve currently has their hud stuff set up is that there is only one active at a time, but it could maybe useful to know if CTF and Payload or smth are enabled at the same time
BUT it could also break things because maybe a hud uses specific placement that only really works when one is enabled
@chilly nova what about demos 
question since I actually was just thinking this, would adding a netvar break demos
i think there was one commit that added a netvar
i think it does break demos but i'll check
hmm i mean it doesnt have to be perfect tbh
not a server side demo?
even stv demos
okay
why not?
im would be fine with taking a reinit path to redo physics and such
and even be okay with a 30 second interruption to roll through the demo incrementally
i'm more thinking the amount of server-side entity state that never gets networked
yeah
i think the saverestore system has a separate description for class members that you might be able to use
but yeah then you'd have to annotate everything
personally i think #worthit
i mean the crazier option is hosting dedis all over the world
*i mean the crazier
option is hosting dedis
all over the world*
and hoping theres enough funding per month
i suppose
what about not having handoff
and heavily penalizing the host stopping the match
but then its not casual enough
ah yeah
and still fucks over players
what state is there anyways
gamemode progress
kd, stats
player and building positions
timer
respawn times
etc
what if
you compiled the server to wasm
and then to do handoff you send the entire vm's memory

actually no that probably wouldn't work cause of external interfaces
unless you wrapped everything
also i don't think you would need to do that since the server can just serialize the entire entity state
Is it feasible to add support for Precision Time Protocol into the dedicated server binary?
we're killing you
wait this hasnt been fixed yet :D
no it's never getting fixed 
Hey! I'm sure you have a million bug reports and these are probably known in TF2, just throwing these out there. I don't know if this is a problem in normal servers, but community servers with 0 respawn time still have to wait for the kill cam, but quickly swapping classes twice let's you spawn faster. It's annoying because it's an unnecessary and obscure tech that can't be used by engies without losing their buildings.
Also if you get ignited by the degreaser, and then the pyro respawns and changes their flamethrower, you start taking more afterburn damage.
(sorry to bother if these are already fixed in TC2)
There's already cvars servers can set to disable the kill cam wait time
Thanks for the report, very interesting
Just remembered a bug, unsure if fixed yet but tested some time ago. If you're playing on a breadspace map and then a different map with the normal announcer you'll still hear the breadspace announcer in the next map
I tried compiling TF2 SDK and loading into a stock map, but it crashes as soon as I'm supposed to enter the map. Anyone get any similar problem?
nevermind, it fixed itself after I ran it a second time. weird that it crashed the first time!
Elon if he mad tf2 configs
concerning
is there a way to query workshop subscriptions for a different app
load will not go well :(
"generic fail", even better
ugh looks like the api doesn't like that i'm not app 440
I wonder if objector decals work
please dont work please dont work please dont work
they do
get owned
but I think they're using a similar system to the workshop maps aren't they?
I guess they're different
Did you try just lying to the steam API and telling it you are 440
to see if that's really the problem
I tried that and it still doesn't work :(
hm
has anyone figured out why debug builds crash yet?
as far as I can tell its unfortunately in code which is not part of the SDK, since it looks like the inventory request works fine
it just goes to some mystical gcclient place that we only have a header for and then breaks
isn't it because its not expected for mixed release modes to be used?
that would make sense i guess, but it does seem rather silly that valve would just not want people to compile as debug since thats kind of... helpful
is there any documentation with all balance changes listed?
I was wondering the same thing.
was looking through sdk forks and this caught my eye (in part because of demoknight tf2)
i've been a tf2 classic player for the past couple years, and only started playing live tf2 again recently - classic and similar leaked code projects were largely siloed off for a long time, both for better and for worse, but between those mods coming into the mainstream and having known for a while that none of these games can really be what i want from team fortress, i want to see what other tf2 communities are working on
planning to look at contributing since i have a programming/it background, but i don't want to overpromise since source itself is new to me and i have other obligations
also looked through mcoms' recent posts about design/balance out of curiosity and i agree with a lot of what's been said about the philosophy behind changes
Could TC2 make koth’s first-cap spawns more intuitive, or are those controlled by the map?
Basically koth’s spawn times are longer for defending team and shorter for attacking team, but on the first cap of the game (neutral to teamed) spawn times don’t be changed for people actively respawning.
BUT if there is more than one spawn wave, the first cap will lengthen affect any spawn wave after that first one.
I’m not sure if it’s an intentional decision or not, but there are a lot of high level players that think first cap is just flat out bugged and don’t know the mechanism. So making it more intuitive would be a big plus.
opinions about returning to v_ and w_ models?
also for forced first person taunts, i think more viewmodel animations should be created for the sandvich
that's a common item used in comp
Lol C_models are way better, just change the textures and VMT parameters of em, hell just changing the VMT parameters would be enough too since VALVe never messed with em to make em look good.
r_drawviewmodel 0 problem solved
like here's the scattergun, left is vanilla, right is with VMT parameter changes.
i dont really see the point
nvm
yeah it's subjective, something that could be easily resolved by modding without upsetting people who like the models as they are.
i cant really tell the difference other than the reflection which seems entirely subjective to me
i mean i was saying because of this
but decompiling is shit
can you add a taunt for the jarate if you try taunting with it right now nothing happens
oh yeah, moving taunts
forget about the forced taunts lol i got confused
but still, i think an animation for the sandvich (and for scout's sodas) would be good
cus he heavy
is there a reason why valve didn't include audio stuff in the sdk?
its probably mostly lumped in with the engine code
only the code needed to build the client and server dlls is included (well, mostly. there is some extra stuff. but no source code for other dlls)
@umbral siren has leveled up! (28 ➜ 29)
yeah, nothing to do with audio it seems, sorry
well I managed to get the game to load and apply the global conditions relatively easily but as it turns out, its surprisingly hard to get the game to reload the files when necessary
at least without just doing a full hud_reloadscheme
I mean in theory I could do it in the client code but that seems like an absolute nightmare to maintain and I'd like to be able to keep it out of game code as much as possible
Hey idk if this is the right place for this, but one thing I would love to see fixed is how getting airblasted affects air control. JI update notes say airblast would "cause the target player to have reduced air control for a short period", but in reality the reduced control lasts until you are grounded again. The current behavior is clearly not intended, and also very unintuitive and disruptive. A pyro airblasting you when you rocket jump away from them causes you to practically get locked into your trajectory, which makes no sense.
the current behavior is actually intended but changes can be tested out if wanted
personally i dont see a reason to change it at the moment unless airblast was reworked again
and im not really interested in doing a third airblast rework
*and im not really
interested in doing a
third airblast rework*
the current airblast is very much designed around killing momentum
actually very much designed as a rocket jumping counter
if we wanted to remove that, it would warrant a whole rework to the mechanic
give pyro airblast infinite range
pyro should be able to charge their airblast
I don't understand how the mechanic is intended, unless "a short period" means until the player lands
the reduction in air control is definitely needed, as airblasting airborne players would be nearly useless otherwise
but I do not understand why it shouldn't last for a certain amount of time instead of until being grounded
It doesn't affect anything anyways I genuinely didn't even know that effect existed until solarlight pointed it out and i have 5000 hours on this game
Well yeah I didn't know about it either until I started playing trolldier lol
It's mainly noticeable when jumping away from pyros
But the fact that it is quite inconsequential but annoying when it happens is why I would like to see it fixed
@gilded bobcat has leveled up! (1 ➜ 2)
Haven't looked into the game code but I would guess changing the air control reduction to last for x amount of time would be quite easy, and almost no one would notice anything, and I doubt anyone would dislike it
Balance this, balance that. I want more killbind types
let me turn into an ice statue at will to confuse friend and foe
*let me turn into
an ice statue at will to
confuse friend and foe*
I didnt do that deliberately but somehow felt like it would be a haiku
i can see how infinite time would be weird
but its hard to determine a good time for it to end, ill try to think of things
the lazy thing to do would be to treat it as same as the old stun, and set it to 0.5s
but its a new mechanic, obviously
so it requires new time bounds
i think there should be some differences in how duration is computed based on some different factors
yeah that is true, I hadn't actually thought about what the duration should be
killbind that spawns a cp_well train to run you over
you should default sv_pure to -1 since it doesn't work
pyro has always been in this weird spot where two of the main reasons you'd use pyro over other classes (airblast and afterburn) are both just miserable to play against, but they're both baked into the class's mechanical identity, power relative to other classes, and player expectations of what pyro is to the point that there's no easy way to address them
you could realistically argue that most classes are trapped in an established identity that makes the game less fun for at least one person playing
but for pyro you can say that about basically every useful characteristic they have
you forgot about comboing which is actually fun and airblast isnt really that bad to play against unless it's this #1072696252162117752 message
someone made like 100 prs in sdk repo in like 40 minutes lol
Dont worry Niterux will review them
the only part of combo pyro that has notable interplay between the weapons is using airblast to remove a form of counterplay (it also does this by removing the strategic commitment of changing weapons rather than adding new tools), and when its main utility is making stuff pyro can already do better it neither adds anything that would make pyro more fun to interact with nor adds anything new to make pyro worth picking
also that video is what optimal pyro play has been for years, even at the peak of puff n sting where pyro could do 200 damage in half a second it wasn't as useful as babysitting or uber stalling
literally me
hopefully they're all merged
there's stuff i like about pyro (for most of my time with tf2 i was a pyro main) but the moment you start optimizing all the things i like about the class collapse
They're all for hl2mp too
When the airblast hitbox was just a giant cube with no directional cone that clipped slightly behind the pyro, so if a spy went to stab them and they airblast at the right time, it would airblast the spy away from behind.
For a better understanding, click this link: https://www.youtube.com/watch?v=sZ0axNXvNik&ab_channel=Jane
All roads lead to screw over Spy.
"Life is pain! I hate m-"
127 branches
i don't remember which source 1 game had an extra slot for keys in the keyboard menu
oh wait it's hl1
0
Bizarre.
wasnt the shape fixed in the 64 bit update
ohh
some weird shit
can you add the jalorpy (half life 2 episode 2)
I've actually seen someone in the tf2maps discord who was trying to get it to work in tf2 with the sdk
on an unrelated note, does anyone know if there are any known issues with VGUI and paths using backslashes? they use CUtlDict to store cached res filepaths, and even though the dict class has an option to do compares where forward slashes and backslashes are handled the same, its not used anywhere in the code
just a weird thing I noticed and I'm curious if anyone has ever seen an issue due to it
I don't think it is an issue there since like 99% of the LoadControlSettings calls use hardcoded strings that use forward slashes, but it looks like vgui screens use a keyvalue for the res filepath it uses as an input
so maybe there it would break
maybe its actually intentional to force programmers to use forward slash so that they don't break compat with linux 
where are you rn at the game?
jalorpy
what's that
how feasible would it be to try and replace all the core engine dlls
theres already vaudio-minimp3 and vphysics-jolt
but what about stuff like matsys
or filesystem, or inputsystem
gonna take some time over the next week or so to look through the issue tracker
haven't really done public repo prs before but both it and the engine are things i'd like to learn
@chilly nova unreal engine layer 
you're not a good person
What the fuck that's so mean
@shell temple You're a lovely awesome person 
"That's what it says, and we're not even testing for that."
Was this issue with the BASE jumper parachute appearing in the middle of the air ever fixed?
Valve: fix it yourself
would be possible to force high res model and texture only for viewmodel or is something more engine sided?
Wtf why is there code in the game for translucent viewmodels
I thought that was a hud hack
jill was working on a solution without a hack
Why did they stop 😞
Valve not finishing what they started is a common occurrence
Hopefully the community can find a fix 🙂
Other tasks were more important I'd assume
Hunt Down The Freeman is a Fan-Made creation of Royal Rudius Entertainment. Not created by Valve.YOU ARE THE VILLAIN...It's time to tell the story of the villain. Hunt Down The Freeman takes you into a journey like no one has before. Witness the pain of the villain firsthand with over 2 hour gameplay, over 40 levels, immersive gameplay, cinemati...
$9.99
3372
ev&tho berkan denizyaran likely never played Half-Life before and made HDTF to grift off of the starved Half-Life community
what backstab changes are made in this version?
https://discord.com/blog/announcing-discords-social-sdk-helping-power-your-games-social-experiences
Integration coming in two hours
i'm holding you to that
does l4d2 have a different vgui version?
so...
never specified which two hours
que
what
so
Btw i was saying becuase i was looking if we could implement l4d2s radial menu
for emoticons?
But for now I don't want to see more tf2 code, I'm tired
To replace the voice command menu
I know he did a long time ago, but it's been years and the fix hasn't been given a community fix tag on the workshop, the chances of seeing it get in the actual are about as low as getting a new weapon.
Can someone tell me how to install the Linux version manually
*Can someone tell me
how to install the Linux
version manually*
Or windows version with proton?
I put it in the sourcemods folder and don’t show on steam
just click tc2 64 exe
run the tc2.sh script, that should work no matter what
For me, it was as simple as moving the downloaded content of the zip, in your sourcemod folder. As sourcemod/tc2/
Launching from itch.io, doesn't work at all of course
Il try that
oh?
Can you give me the directory path to check the right path
it'll be in the files you downloaded
Sourcemods folder path?
wherever you extracted the tc2 game files to
The Windows version will launch through wine, and the Linux one might still be sandboxed. Even without enabling it, it didn't managed to launch it that way.
Steam/steamapps/sourcemods
there is a file named "tc2.sh" in the files that you downloaded, run that
~/.steam/Steam/Steamapps/sourcemods
that's odd, it worked when i tested it. is it possible to get logs from when you launched it
Or, at .local/share/Steam. .steam is just a symbolic link
Tomorrow if I remember
It has a error
what is the error?
./tc2.sh ─╯
[S_API] SteamAPI_Init(): Loaded '/home/medrinnn/.local/share/Steam/linux64/steamclient.so' OK.
Setting breakpad minidump AppID = 243750
SteamInternal_SetMinidumpSteamID: Caching Steam ID: 76561199769827413 [API loaded no]
[Source Mod Launcher] Launching default game: tc2
SDL video target is 'x11'
SDL video target is 'x11'
Using shader api: shaderapivk
info: Game: hl2_linux64
info: DXVK: v2.3
info: Vulkan: Found vkGetInstanceProcAddr in libvulkan.so @ 0x72481e39ff00
info: Built-in extension providers:
info: SDL2 WSI
info: Enabled instance extensions:
info: VK_EXT_surface_maintenance1
info: VK_KHR_get_surface_capabilities2
info: VK_KHR_surface
info: VK_KHR_xlib_surface
MESA-INTEL: warning: Haswell Vulkan support is incomplete
warn: Skipping Vulkan 1.2 adapter: Intel(R) HD Graphics 4600 (HSW GT2)
warn: DXVK: No adapters found. Please check your device filter settings and Vulkan setup. A Vulkan 1.3 capable driver is required.
Convar sdl_displayindex has multiple different change callbacks
Using Breakpad minidump system. Version: 120022 AppID: 243750
Setting breakpad minidump AppID = 243750
Using breakpad crash handler
[S_API] SteamAPI_Init(): Loaded '/home/medrinnn/.local/share/Steam/linux64/steamclient.so' OK.
Forcing breakpad minidump interfaces to load
Looking up breakpad interfaces from steamclient
Calling BreakpadMiniDumpSystemInit
03/19 20:24:33 minidumps folder is set to /tmp/dumps
03/19 20:24:33 Init: Installing breakpad exception handler for appid(243750)/version(120022)/tid(6869)
Looking up breakpad interfaces from steamclient
Calling BreakpadMiniDumpSystemInit
SteamInternal_SetMinidumpSteamID: Caching Steam ID: 76561199769827413 [API loaded yes]
SteamInternal_SetMinidumpSteamID: Setting Steam ID: 76561199769827413
Did not detect any valid joysticks.
[CFontManager] Forcing freetype2 for GorDIN
[CFontManager] Forcing freetype2 for GorDIN-Black
[CFontManager] Forcing freetype2 for GorDIN-Bold
[CFontManager] Forcing freetype2 for GorDIN-Light
[CFontManager] Forcing freetype2 for GorDIN-Regular
[CFontManager] Forcing freetype2 for GorDIN-Semibold
[CFontManager] Forcing freetype2 for GorDIN Black
[CFontManager] Forcing freetype2 for GorDIN Bold
[CFontManager] Forcing freetype2 for GorDIN Light
[CFontManager] Forcing freetype2 for GorDIN Regular
[CFontManager] Forcing freetype2 for GorDIN Semibold
Fontconfig error: Cannot load default config file: No such file: (null)
crash_20250319202437_2.dmp[6883]: Uploading dump (out-of-process)
/tmp/dumps/crash_20250319202437_2.dmp
crash_20250319202437_2.dmp[6883]: Finished uploading minidump (out-of-process): success = yes
crash_20250319202437_2.dmp[6883]: response: Discarded=1
crash_20250319202437_2.dmp[6883]: file ''/tmp/dumps/crash_20250319202437_2.dmp'', upload yes: ''Discarded=1''
/home/medrinnn/.local/share/Steam/steamapps/common/Source SDK Base 2013 Multiplayer/hl2.sh: line 113: 6869 Segmentation fault (core dumped) ${GAME_DEBUGGER} "${GAMEROOT}"/${GAMEEXE} "$@"
that the log
coredumpctl dump \E2\94\80\E2\95\AF
PID: 4973 (steamwebhelper)
UID: 1000 (medrinnn)
GID: 1000 (medrinnn)
Signal: 4 (ILL)
Timestamp: Wed 2025-03-19 20:27:17 +00 (55s ago)
Command Line: $'/home/medrinnn/.local/share/Steam/ubuntu12_64/steamwebhelper --type=renderer --user-agent-product=Valve Steam Client --user-data-dir=/home/medrinnn/.local/share/Steam/config/htmlcache --crashpad-handler-pid=4905 --buildid=1741737356 --steamid=0 --lang=en-US --num-raster-threads=2 --enable-main-frame-before-activation --renderer-client-id=5 --time-ticks-at-unix-epoch=-1742413594666924 --launch-time-ticks=816290402 --shared-files=v8_context_snapshot_data:100 --field-trial-handle=3,i,985147375092270763,11602722209531856295,262144 --enable-features=PlatformHEVCDecoderSupport --disable-features=BackForwardCache,DcheckIsFatal,DocumentPictureInPictureAPI,SpareRendererForSitePerProcess,ValveFFmpegAllowLowDelayHEVC --variations-seed-version --log-file=/home/medrinnn/.local/share/Steam/logs/cef_log.txt'
Executable: /home/medrinnn/.local/share/Steam/ubuntu12_64/steamwebhelper
Control Group: /user.slice/user-1000.slice/[email protected]/app.slice/app-cosmic-steam-native-2168.scope
Unit: [email protected]
User Unit: app-cosmic-steam-native-2168.scope
Slice: user-1000.slice
Owner UID: 1000 (medrinnn)
Boot ID: 506f856afabe425396b49b6b900d38a0
Machine ID: 74b390ecd88e455b9ce4955b8b6ff725
Hostname: cachyos
Storage: /var/lib/systemd/coredump/core.steamwebhelper.1000.506f856afabe425396b49b6b900d38a0.4973.1742416037000000.zst (present)
Size on Disk: 65.6M
Message: Process 4973 (steamwebhelper) of user 1000 dumped core.
Refusing to dump core to tty (use shell redirection or specify --output). @chilly nova
oh sorry do coredumpctl dump hl2_linux64 instead
Wait a sec I restart my pc
coredumpctl dump hl2_linux64 \E2\94\80\E2\95\AF
PID: 6869 (hl2_linux64)
UID: 1000 (medrinnn)
GID: 1000 (medrinnn)
Signal: 11 (SEGV)
Timestamp: Wed 2025-03-19 20:24:37 +00 (7min ago)
Command Line: $'/home/medrinnn/.local/share/Steam/steamapps/common/Source SDK Base 2013 Multiplayer/hl2_linux64' -game /home/medrinnn/.local/share/Steam/steamapps/sourcemods/tc2/tc2
Executable: /home/medrinnn/.local/share/Steam/steamapps/common/Source SDK Base 2013 Multiplayer/hl2_linux64
Control Group: /user.slice/user-1000.slice/[email protected]/app.slice/app-cosmic-com.system76.CosmicAppList-6724.scope
Unit: [email protected]
User Unit: app-cosmic-com.system76.CosmicAppList-6724.scope
Slice: user-1000.slice
Owner UID: 1000 (medrinnn)
Boot ID: 506f856afabe425396b49b6b900d38a0
Machine ID: 74b390ecd88e455b9ce4955b8b6ff725
Hostname: cachyos
Storage: /var/lib/systemd/coredump/core.hl2_linux64.1000.506f856afabe425396b49b6b900d38a0.6869.1742415877000000.zst (present)
Size on Disk: 6.6M
Message: Process 6869 (hl2_linux64) of user 1000 dumped core.
Module [dso] without build-id.
Stack trace of thread 6869:
#0 0x000072481f77cf0d n/a (n/a + 0x0)
#1 0x00007248193af80b n/a (n/a + 0x0)
#2 0x000072481a99d822 n/a (n/a + 0x0)
#3 0x000072481a99ddbb n/a (n/a + 0x0)
#4 0x000072481f508cc2 n/a (n/a + 0x0)
#5 0x000072481f508d54 n/a (n/a + 0x0)
#6 0x000072481f508e58 n/a (n/a + 0x0)
#7 0x000072481f508e67 n/a (n/a + 0x0)
#8 0x000072481f4e544a n/a (n/a + 0x0)
#9 0x000072481f6278ce n/a (n/a + 0x0)
ELF object binary architecture: AMD x86-64
Refusing to dump core to tty (use shell redirection or specify --output).
how?
@crystal lynx would you happen to know specifically, i've never done it myself
.
ohhh
Need to set the Source SDK 2013 Multiplayer via Proton
if valve managed to make proton be able to run autocad reliably i would instantly change to linux
valve dont care about software support unless there making a linux computer
You mean, CodeWeavers, with Wine ?
Proton is for Games.
well afaik
https://www.codeweavers.com/compatibility?browse=&app_desc=&company=&rating=&platform=&date_start=&date_end=&name=Autocad&search=app#results You may pay CodeWeavers for better support, especially if Professional
mmm i'll check that later
Run Windows apps such as Microsoft Office/Adobe in Linux (Ubuntu/Fedora) and GNOME/KDE as if they were a part of the native OS, including Nautilus integration. Hard fork of https://github.com/Fmst...
./tc2.sh \E2\94\80\E2\95\AF
[S_API] SteamAPI_Init(): Loaded '/home/medrinnn/.local/share/Steam/linux64/steamclient.so' OK.
IPC function call IClientUtils::SetAppIDForCurrentPipe took too long: 97 msec
Setting breakpad minidump AppID = 243750
SteamInternal_SetMinidumpSteamID: Caching Steam ID: 76561199769827413 [API loaded no]
[Source Mod Launcher] Launching default game: tc2 now i have this log
bro, dont tell me you still use AutoCAD in 2025
what is one supposed to use?
I like fusion 360
anything with a parametric feature based approach
Minecraft block bench
using AutoCAD is like using Microsoft Paint to do graphical design
good!
But i still have to use autocad since it's the only thing my father uses
aw
He's not into the bim sector yet
i always wanted to see this fixed
Are we finally able to fix that shadows through walls bug
no
ah so it's tied to the engine itself
1 FPs
I love that bug, love getting called a cheater for it lol
portal prelude implemented qt inside source for hud (i guess)
qt for comtress when
ops
portal revolution**
or to more specifc, strata source
ah
makes more sense
panorama for comtress when
not ironically wouldn't be a bad idea to use godot ui feature on tf2
the issue with integrating any ui system is that we can't modify the engine
afaict the only way to really do custom ui is to wrap it in a vgui panel
or completely bypass the input & rendering systems like how cheats do it
the source sdk isn't the csgo version?
no
valve couldnt do that?
yes valve could do it
but you can't do change the engine with just the sdk
(if i get one person saying to use hooks i will)
actually
we could do a lot of nice stuff with hooks
but idk if coms would want to do that
why was gameui code removed too?
Don't need spawn wallhacks, the very game engine itself hates Spy.
It's one thing I'd really love Comtress to think about, tweaking the engine so it mildy dislikes Spy, rather than outright hating him.
I also hate spy, so I think engine should hate it too
Come on, the engine should be somewhat fair at least.
maps should be tweaked to have thicker walls
that could fix the problem without messing with engine
that would be harder than fixing the engine
Honestly it’s surprising how thick of walls will show shadows through. Walls need to be kinda counter intuitively thick to not show through.
You could maybe texture them as cinder block to justify it, but brick is so common place by now that I think it’d be less awkward to just add invisible walls to keep people away from the wall. Or just accept shadow wall hacks
Just turn shadows off and then you can feel morally superior to the shadow wall hackers when you beat them
It's a feature 
i decided to turn shadows very low to get 8fps back 😦
blob shadows that you find in 3D platformers to help you see where you land better.
what if you just make a cinder block wall inside all the walls
In the Attic of Luigi's Mansion on the Nintendo Gamecube, the attentive player may see, amidst a flash of lightning, during Luigi's phone call, an unnerving shadow is cast on the wall. The disturbing thing about this is that shadow is seen hanging above the floor making it appear as though Luigi's shadow has hung itself. This is, of course, just...
we knew that
not much 
kinda bored
until dedicated server hosting is available from valve i dont wanna do anything tbh
shrimply run a client with textmode
Well there must be something different about them right
Keyvalues got messed up when using the wrong one
you can't use community servers?
I think she means like valve releasing a working version of the standalone server utility
for the source sdk 2013
that actually works without having to do extra manual stuff to get it working
i mean there is a dedicated server app
Cant wait for the dedicated server to come out right alongside the heavy update
gonna make a hack fix where players can't do dmg in setup time
Would it be possible to create a bind_per_class command?
I think keybinds rely on stuff in the engine, it may be doable though
Even if it is, is it worth the effort of implementing it instead of just using class configs?
Yeah that seems bad
Plus what would it write
would it write bind key thing
or like bind_class class key thing
if the former then you'd still have to bind the key to something else on other classes
it just seems bad all around
Doesn't "bind" do that already?
What do you mean?
Bind writes to config.cfg right?
yeah
And just do nothing with regards to the key on other classes?
If you don't also bind it in all your other class configs, it will be bound on engineer, but when you switch class it will stay
Also you would you deal with the mastercomfig issue
mastercomfig class configs are no longer in cfg
how would the game know that
(it can't write to the ones in the mastercomfig vpk)
how would you deal with configs beig inside read protected dirctry
i guess in ther other class configs it would write unbind
Yeah I mean certainly you could do it
Like just add a console command that writes whatever you like into any file
You'd have to decide how you want to handle the issues of writable class configs potentially not existing or being in another location
I wonder if it would be possible to achieve the same goal with aliases
I guess it wouldn't probably
Don't know whether it had already been asked or not and don't know where to ask this if not here: I'm new to this channel and discord as a whole (yes, really), sorry.
But is TC2 coming to Steam in the future? I know there's a lot of work to be done, so I'm not expecting it anytime soon, but is there's a possibility?
Have a good day, everyone.
I don't get what this means.
@tight nexus Thank you good sir
The dedicated server in the source sdk 2013 is kinda broken right now
You have to do special steps to build and run it
Valve probably plans to fix that since misyltoad commented on the issues about it
So mastercoms is waiting for valve to act in that regard
ahhh
It doesn't matter to me even if I will have to wait even 10 years till TC2's Steam release.
It will outshine vanilla TF2.
why
o-o
can we have every single mannpower change from the past 5 years reverted
Ain't a rebalance mod.
Not since. Saw the pinned post.
Even with notifications on the topic, I feel i don't get updated enough. Oh well, that's on me.
anyway, every single change to mannpower thats been made recently has made the gamemode worse
every class that was fun to play isnt anymore, including changes that just make some of them unplayable
heavy can barely move around the map anymore because they removed his ability to jump out of grapple that everyone else had
so hes just sluggish and doesnt fit the pace of the gamemode at all
not to mention some maps are designed for jumps because you wont always have a ceiling above where you want to grapple
so he just gets totally owned by simply existing
demoknight is just laughably useless now
if a demoknight charges with a powerup he has to wait an entire minute before his charge is refilled
its fucked
while using an item like the splendid screen which reduces your cooldown its still like 45 seconds
the persian persuader which refills a portion of your charge meter on hit takes 14 hits to get you back to full charge
at least the tide turner still fills up your meter if you get a kill, but it wont matter because your damage is nerfed compared to any other melee weapons which get a damage bonus for the gamemode
im not even touching on the "dominant state" they added which nerfs players doing well by giving them a lazy marked for death debuff they cant do anything to remove
so theyre just punished for getting good at the game???
idk, i think its one of the worst balance updates tf2 has gotten
and its for a gamemode that already barely gets players! i used to play the gamemode until those changes!
oh and not to mention that the game ALREADY HAD A COUNTERMEASURE FOR TOPSCORING PLAYERS
you get crits when your team is falling behind!!!
All those rebalances and I still dont wanna play mannpower LMAO.
After playing it countless times.
(The matchmaker would constantly put me in mannpower despite most maps being ticked on til I ticked mannpower off)
Moving to the new channel for better visibility: https://discord.com/channels/389089828249010188/1362247891913281786