CP Snowbase b20

b1_p3 - Beta Pre-release

Releasing a pre-beta of sorts, basically just a release of where the map is currently. Still working on a few things - some custom props are missing collision, one set of stairs is currently just a nodraw ramp, no soundscapes, etc.

Feedback on detailing is appreciated.

(This also marks the point where the map has been combined into a single multi-stage map.)

Read the rest of this update entry...

b2

-First playable beta. Some rough edges (lots of lighting seams) but otherwise a fully playable and fully detailed map.

Read the rest of this update entry...

Gravidea said:

The second spawn for the attacking team, the one I assume is supposed to be accessible upon capturing the first point, is automatically accessible upon the round starting.

Click to expand...

Thanks for the heads up! I thought I had tested for this, but evidently not enough. Combining multi-stage spawns and forward spawns is weird. I'll be uploading a fix later today.

b2a

-fixed forward spawns (probably)

Read the rest of this update entry...

b3
-added a detail room in stage 2 blue spawn
-fixed bug where capturing final point did not update the CP prop skin
-forward spawn signs start off and turn on rather than suddenly appearing
-probably other things but ive forgotten

Read the rest of this update entry...

b4

-ammo adjustments at 1-1 to hopefully gimp engineer stacks
-removed funny railing behind stage 2 fwd spawn (for chin)
-removed funny pipe
-fixed a bug where you could smash yourself into a duct and never escape
-fixed lighting bug behind 1-2 (still need to fix bug under 2-1, but it has proved elusive)

Read the rest of this update entry...

So, after playing the multi-stage version of this map a few times, something interesting has started to crop up - the second stage seems less fun and less balanced than it was in previous versions. Why?

After some discussion and some light number crunching, I'm discovering that this map may be falling victim to "survivorship bias": basically, data can be skewed when your data is only counting the "survivors" and doesn't count scenarios where no data is returned. In this case, we aren't getting the data on how teams perform on Stage 2 when Red wins on Stage 1.

Why is this a problem? Each stage was balanced individually around a 50% win rate, give or take. When Stage 2 was tested in isolation, this was including the scenarios where teams that would lose Stage 1 were still playing Stage 2. The actual win rate on Stage 2 is going to be much higher than 50%, since the only Blue teams who will play on Stage 2 are the same teams who already beat the 50% odds of winning Stage 1.

Part of this issue is also being a bit exaggerated because Stage 1 seems to have become harder over time as players learn how to better play defense (or perhaps are forgetting how to attack...), and the win rate for the past two betas is more like 33% (it was sometimes around 66% in alpha). This means that the teams who get to Stage 2 are even better and it steamrolls that much harder for a ~60% win rate. Not great when the Stage 2 win rate is about double that of Stage 1. Even if Stage 2 is harder in isolated testing, it doesn't feel that way for anyone actually playing the game.

So, I'm going to be doing a few things:

1. Test Stage 2 on its own to make sure it ever played well and I'm not just insane.
2. Test Stage 1 on its own with layout changes to help make it easier for Blue to win. I'm going to aim for a 66% - 75% win rate. Both points need tweaks.
3. Once Stage 1 is where I want it, test both stages together and make Stage 2 balance changes as necessary.

Hopefully this doesn't amount to a lot of actual labor on my end, just a lot of watching, thinking, and making small updates. Most of the work here has already done and I have a solid core for a map, but I'm not happy with it yet.

b5 s2

STAGE 2 ONLY VERSION

-shrunk structural "ribs" on balconies at B and added clipping to smooth the gameplay area out

Read the rest of this update entry...

b5 s1

STAGE 1 ONLY VERSION

A:
-adjusted rocks at point
-reduced medium health near blue forward spawn to small
-reduced medium health in garage to small, moved to ramp under point
-moved pickup pair near beach drain to side of large docks

B:
-reworked area behind point to encourage more dynamic and interesting play, making the area more valuable for both teams:
--replaced prop jump with gradual ramp
--added upper balcony route
--added spiral staircase room to red side of back area, connecting sewer and new balcony route
--moved large ammo into staircase room

Read the rest of this update entry...

A Ninja said:

Sorry about that, my bad. What you said makes a lot of sense and I'm looking forward to the final product!

Click to expand...

If you want to download the latest full version, you can grab it from the Version History page. I'm also holding any single stage test versions from the Steam Workshop, so it will always be a full version of the map.

Playtest thoughts:

- A is definitely easier to take as a result of the pickup changes, which makes me happy. It was never fully held, but red was usually able to put up a good fight. This is more or less where I want it to sit balance-wise, with red only holding it in extremely imbalanced matches.
- Sniper, unchecked, is pretty good on B. I'm not sure if this is a problem or not, as there are quite a few maps where unchecked snipers are an issue. If red is running a good sniper and blue doesn't counter with their own snipers or spies, is that an issue? The problem may also be that snipers and spies on blue are having a hard time doing their job because of unrelated issues (unable to get into position due to a lack of options or because the enemy team is able to hold too far forward).
- The new back balcony area at B didn't really see a lot of action. It's meant to be a fun side arena for both teams, but blue never actually went there. Sometimes this was because red was applying pressure that prevented them from doing so, but blue had plenty of opportunities that they did not take advantage of, probably because the route does not read as important from their side. After all, the only entrance is a doorway smaller than 128x128, which comes into a room with too many corners to reasonably check. I think blue needs an easier time rotating up here and the door needs to be bigger so players realize its importance. I also need to make the route easier to push if red has players actively defending the area.
- Sewer dropdown at B has always been a high risk route on purpose, but I'm starting to realize the high risk is almost never worth it when playing against a red team that knows to check the area. Is the high risk required to make this an interesting / clever / subversive route to use? I'm going to convert this to stairs and see what happens.

b6 s1

STAGE 1 ONLY VERSION

A:
-no changes

B:
-added rock to provide more cover for blue's main push & help manage sightlines
-added platform to back route to make it more useful for blue to clear out the low ground
-changed small health on flank route to medium
-widened door on flank passage and removed upper platform in room to make blue pushes easier and hopefully more obviously valuable
-changed sewer dropdown to stairs

Read the rest of this update entry...