Radix

I think the whole system of adding/removing (community) maps should be changed and most importantly be more transparent.

1. In my opinion more maps should get a chance of getting played and payed. The original Operations were a good thing.

2. Originally Valve wanted to (or at least they said it) keep popular community maps in the game and change the map pool according to popularity. One of the first maps treated like this was my Austria map. So it stayed in for a while and had pretty good queue times (most of the time shorter than Cobble, Nuke, Train and often even Overpass). But then they switched to "2 in, 2 out" mode. Some maps like Abbey seemed to be pretty popular but just got removed after 4 months or so. If players know that a map will be removed after such a short time, they will not want to learn it and keep playing Mirage or whatever.

3. Valve should give more feedback to the mappers what's going on. How many players do actually play a map? Which game modes are played? Balancing? At least to me they didn't give any information except some balancing stats. Maybe they gave more information to the others guys. But we won't get to know this because most people here don't share any knowledge anymore and just post their shiny screenshots to get some likes.

4. There should be a more transparent and consistent system of adding/removing maps. Maybe something like this based on popularity:

x <- -> Casual only (or Operation) <- ->(Scrimmage only for a short time <- ->) Scrimmage and Comp <- -> Active duty

5. Why are maps in Scrimmage mode OR Comp? Why not both? Scrimmage is good to learn a map without destroying your rating. But most people won't play more than 1-5 matches like this. I think pretty much all maps should be in Scrimmage game mode. Or if this increases queue time too much, there could be maybe 2 regular maps + another one that changes every month or so (atm Mirage)

6. What about Wingman game mode? I actually like this a lot. But most maps are just terrible and/or imbalanced. Would be nice to see some community maps there.

4 hours ago, Passerby97 said:

I respect that you took time to write this, but this is not really a response to anything I said.

Oh, then I probably just got you wrong. Sorry.

Quote

But maps that Mapcore doesn't acknowledge aren't accepted to CS.

Which maps do you think of particularly that were good but not added because of not being "acknowledged" by Mapcore?

Yes, there are a couple of maps that looked interesting but never got added (Grind, Import, Sanchez, Waterfall, Iris, ...). (Just talking about the visuals, not layout, because that's obviously Valve's first exclusion criterion)

But maybe there are different reasons. As I already said, they might just have failed when tested by Valve...

Or maybe there were just other maps that were better at the same time.

After all there are only a few mappers that reach the high requirements for the visuals. And these guys try to finish their maps by the end of a contest. Because of the fame/extra money I guess. So it's natural that most "good" maps are contest winners in the end.

Quote

Anubis has 50-100 fps lag spikes on medium rigs

I agree. Anubis looks awesome, but it's not well optimized yet. (Of course not! Because that's one of the last things you do when you make a map)

At it's current state it should not have been added to the game yet imo. It will surely run a lot better in a few weeks...

Quote

Ruby and Breach got completely remade midway through their in-game rotations. Not every map can be perfect when first accepted into the game, and that clearly does not stop them from being accepted.

Yes. But mostly gameplay changes. These maps looked very polished and played well already when they were added. The authors just wanted to improve gameplay based on player feedback imo.

What I said about Junction was that it's currently way to unpolished to be added. (Flickering props, nodraw surfaces, temporary assets, ...). Don't get me wrong. I like the map. But 80% finished is not enough.

After all you remind me of a guy called "Tynnyri". Just more polite.

2 hours ago, Passerby97 said:

Junction was awesome map and there's no good reason why it wasn't added to the game. But please also realize that Valve doesn't decide the maps accepted to the game anymore like they used to. Now they seem to purely accept Mapcore contest winners or maps from famous mappers.

I think they didn't add it to the game because de_junction was never finished. Can't talk about gameplay, because I never really played it.

But yes, the visuals are awesome at some spots, but some areas need a lot of additional work imo.

I don't know why, but it seems like the creators abandoned this map (?)

The fact that Valve only picks Mapcore maps is because pretty much every good CS:GO mapper is at Mapcore.

Valve also tests every candidate internally, that's why some awesome looking maps sometimes don't get added (because they are not fun, completely imbalanced, exploitable, not optimized (low fps) or whatever).

23 hours ago, Squad said:

how newer (official and custom) maps do it nowadays

It's pretty inconsistent even in officials maps. In the older official maps the first step is mostly skipped like you did. But in some maps like nuke or cobble both variations are used (for no reason), while newer maps like dust2 and inferno are completely smoothed (all steps clipped).

I think they initially skipped the first step for the following reason and then adapted it for nearly all stairs without thinking about it (even if it didn't make sense.):

If you clip all stairs, in some situations the player "stumbles" over the ramp when he just wants to pass in front of the stairs. That feels pretty irritating and it's the reason why they skipped the first step (in my opinion). But this can only happen if there are no walls at the sides of the stairs.

Edit: You already said it here. I should read more carefully before wasting time

23 hours ago, Squad said:

because sometimes the bottom clip extends beyond a corner

I prefer to clip all stairs and add an extra bevel at the side if needed like here in dust2: (Yes, the wall clip is too large there... Valve please fix :P)

53 minutes ago, Tynnyri said:

Something like Nuke? That would very interesting to see!

I think Nuke is functionally still a 3 lane thing but with bombsites on top of each other, which is a terrible idea nowadays. I think something like train (but with less annoying bombsites) would be more reasonable.

But I suggest to move this discussion somewhere else. This thread belongs to de_swamp.

I think it's a good idea to build something new (slow movement in water) on something proven ("3 lane"/"4 square" whatever layout).

So players feel immediately "at home" in the map and are willing to learn the new things. If you add do too many new things at once players might be frustrated.

I think you did a good job here.

58 minutes ago, Lefty said:

They got till the end of Feb 1st lol

Yeah, but looks like a photo finish...

30 minutes ago, jubjunior said:

I've been having problems with loading screen in workshop maps, it doesn't load overview/background, i tested in 2 PCs. Official maps and operation maps works fine. Is that a bug? If so, the judges will be alerted?

Yes, it's a bug. The loading screen/overview thing is broken for years. Before it was just broken for former Operation maps but now it's all Workshop maps, so equality for all I guess.

23 minutes ago, blackdog said:

Use WallWorm

20 minutes ago, That50'sGuy said:

I've been using that tool since the very beginning. I can't imagine what modeling would be like without it hahah!

I've always been a little astonished because apparently everyone uses this software (3dsmax) that cost 2k bucks every year.

Well, there is this students version, but you may not make money with your stuff then (CS:GO Operations etc.).

Am I the only idiot that uses Blender, which is for free and far good enough to make shitty low poly props for CS:GO?

What's up with that?

1 hour ago, Vaya said:

@Radixthank you for catching it!

Yeah, just posted it here because nobody else did, so that people don't miss it.

Do you know who made that image? Love the Win98 (?) vibe!

https://twitter.com/MapcoreHub/status/1208070633097244674

1 hour ago, blackdog said:

Who said games are not educational?

Btw... The other day I saw the picture below while surfing the web and immediately recognized Hallstatt. For a second I thought that I would have to read something like "terrorist practiced bomb attack on Austria in popular video game" bullshit, but it was just a fire. Luckily the fire fighters could prevent greater damage. Still sad to see this. Hope they can reconstruct the historic buildings...

57 minutes ago, Vaya said:

Can you not just play outside?

Just go like this

2 hours ago, csWaldo said:

8 months is enough for anyone to make a contest winning map.

Apparently it's not. That's why we have a discussion and why >90% of the map entries will not be finished in time ever. 

1 hour ago, birjolaxew said:

Is this the right place to have a discussion about rule 4? I think we need to talk about it as a community, now that de_zenith has exemplified how problematic it is.

I understand your concerns. But I don't think that there is an actual alternative for "rule 4".

The contest is NOT about who can create the best map in ~8 months. Mapcore is holding this contest to encourage people to create/finish maps and so that some maps get a little more attention.

If you really expect that everybody starts from zero when the contest starts, you can just forget it. Because nobody (except some hackers and intelligence agencies ) can control what's going on the hard disks of people. People can easily keep finished maps secret until a contest starts. The only way to avoid this would be to have a very limited theme (something like "snowy castle" and NOT something open like "exotic places"). But this would have the consequence that very few people would participate in the contest. Because nowadays most people need something like 2 years or more to create a new map.

So just accept it, or we won't have any contests ever again.

So... No answers yet.

I guess most people have the same problem that I have: I don't see anything.

What's the problem? How does it look like in Hammer?

Water in Hammer/Source:

https://developer.valvesoftware.com/wiki/Adding_Water

All brushes are black because the vrad.exe crashes for some reason.

22 hours ago, Donskioni said:

zero area child patch

I guess this causes the problems. So there is probably some messed up brushwork.

• Check for invalid brushes (Press "Alt+P" or go to "Map" -> "Check for problems", then search for invalid solids).
• Check for misaligned brushes. Brushes which are not "on the grid" with small gaps between them.
  • If you can't find them, try to compile parts of your map via cordon tool. So you can isolate the problematic brushes.

- There are also some errors with nodraw texture on displacements (select these faces with "Toggle texture application" (Shift+A) then go to "Displacements" -> "Destroy")

- Also some wrong usage of models (prop_static using a model, which must be used on a dynamic entity (i.e. prop_physics) )

The more you do, the more you do wrong.

In expert compile mode, add -StaticPropLighting to your compile parameters.

https://developer.valvesoftware.com/wiki/Static_Props_show_up_black

On 10/6/2019 at 1:01 AM, JorisCeoen said:

I am curious to know about the benefits of autocombining props. I know Dust2 uses literally almost exclusively this feature for its props, but I just don't understand what it does. The only possible use I know for it is if you have lots of grass props and stuff like that, you can easily combine them to make a single prop and save on having each seperate prop_static entry being calculated seperatly... but in Dust2 I've discovered that sometimes 3 or even 2 props are being combined and there would literally be no reason to do it, as far as I can tell.

It's about "draw calls". Many small props mean many performance-eating draw calls. The static prop combine feature merges multiple props into a single big prop, which is less performance-hungry.

Combining very few props may not make much sense (rendered geometry vs draw calls?), but that's because it's done automatically. All in all it's still an improvement. Valve has stated that Nuke runs 40% faster after they implemented static prop combine.

@untor showed me how to use the autocombine feature and we will post a short guide soon...

Edit: Here it is...

https://www.mapcore.org/articles/tutorials/static-prop-combine-in-csgo-r111/

19 hours ago, untor said:

If you want, i can told with you in Discord and try to help you with Autocombine, and then you can help me write a tutorial on english)) 

Sure, I can try to help you if none of the "tutorial guys" wants to.

Further discussion via pm I suggest, I don't want to hijack your thread.

Looks really cool so far! Take care of player visibility!

On 8/10/2019 at 12:18 PM, untor said:

Hi Mapcore! Today, I want to show you, great stuff!
I finally figured out the issue of auto-combination of props_static.

Seems like you are still the only one who managed to make the autocombine feature work...

I tried it for a while, but it still doesn't do anything. No errors, nothing at all...

- So the *.qc (and the other source files) of your model has to be in "SteamApps/common/content/csgo/", right?

- Anything special about the qc? Is there a chance that you could share an example *.qc file?

- What about the other source files: Which file format? *.smd or *.dmx?

- What compiler do you use to compile the model? (Is "Crowbar" ok? https://steamcommunity.com/groups/CrowbarTool )

On 8/12/2019 at 7:55 AM, untor said:

and now, need compile a map with Combine settings in VBSP
-StaticPropCombine
-StaticPropCombine_AutoCombine
-StaticPropCombine_SuggestRules
-KeepSources (for checking in hammer)
-StaticPropCombine_ColorInstances (for checking after compile in game, it add random color for combined props) 

- So these are your compile commands for vbsp: -StaticPropCombine -StaticPropCombine_AutoCombine -StaticPropCombine_SuggestRules -KeepSources -StaticPropCombine_ColorInstances -game $gamedir $path\$file

( -StaticPropCombine_AutoCombine makes it ignore your spcombinerules.txt  afaik)

- stub qc and spcombinerules.txt as you described above (according to this: https://developer.valvesoftware.com/wiki/Static_Prop_Combine )

- anything else?

What happened to the community maps (Ruby and Breach) was pretty much what I expected (looking at the queue times), but the fact that they were not replaced with other community maps but with Cache and Mirage is interesting.

Cache because there will be a reworked version in a few days, so it can be tested easily.

But Mirage? Will there be a reworked version soon as well? It's currently one of the maps of the comp pool that need a visual rework the most, so I'm pretty sure this is what is going to happen.