Snowbase

CP Snowbase b20

  • This map is featured! Our best maps, all together in one place for your viewing pleasure.

Idolon

they/them
aa
Feb 7, 2008
2,105
6,106
b7 s1

STAGE 1 ONLY VERSION

OVERALL:
-raised max round time from 6 mins to 7 mins

A:
-optimization pass
-adjusted rock to block sightline
-moved a bush

B:
-adjusted rocks under bridge to block sightline
-changed red respawn wave time from 9s to 10s

Read the rest of this update entry...
 

Idolon

they/them
aa
Feb 7, 2008
2,105
6,106
Whenever I'm developing a map and confidently identify a major issue, I tend to halt development until I can properly solve that issue. Since I have a lot of projects going on simultaneously, I tend to switch projects whenever this happens, letting the problem simmer in the back of my mind until a solution starts to form. I've been in this spot with this project for the last month-ish.

Snowbase Stage 1 has been living in the shadow of this particular specter:

upload_2021-1-30_7-43-50.png


Spawncamping to win is more or less a valid strategy on any map, so this isn't the most directly helpful feedback. In addition, spawncamping isn't really required to win on this map, as the red spawn is quite some distance from the point. However, playing the map and watching demos reveals the actual issue - Blue players who are playing to win can use the sewer route at B not for high-risk flank plays towards the point, but instead to just spawncamp.

upload_2021-1-30_7-49-9.png


I was aware of this possibility when I had originally designed the point, but didn't consider it a big issue, as blue players going this way are quite vulnerable to attacks from behind. However, I failed to consider a few things:

1) Long respawn times for red mean that blue players managing to get even just a few kills this way will have a significant impact on the round.
2) Even if this is somehow balanced, it isn't fun for red.
3) If we assume that red's possible counterplay (attacking the spawncampers from behind) makes this balanced, red's counterplay requires good team coordination to pull off, with players getting spawncamped alerting their teammates on the point of the threat, and alerting them quickly, and the teammates on the point coordinating a proper response, splitting enough resources off to deal with the threat without completely abandoning the point. You can't expect that from a pub team. On the other hand, all the blue team needs to do this is a power class with a medic who both want to spawncamp. Worst case scenario, this strategy becomes the meta for blue and they don't even need to coordinate at all.

So, what to do about it? My first thought was based on the history of the map - this sewer route has always been here, but spawncamping only started being an issue once I had changed the sewer route from a dropdown to a stair, so surely these things must be related. What if we made the route a bit harder to use as a retreat? This would make it easier for red players to catch blue players out while they are retreating.

upload_2021-1-30_8-1-27.png


On reflection, this probably doesn't solve the issue. Which is more likely: Making this a two-way route enables spawncamping, or making this a two-way route lead blue players to discover spawncamping was a viable strategy? Maps are not black boxes where you put players in one end and always get the same results out the other end. The strategies that players use will develop over time, and map changes may help unearth strategies that were always there, just undiscovered. We need bigger changes.

Ok, so what if we redesign spawn to make it less campable? Let's give the spawn some high ground, maybe adjust some cover.

upload_2021-1-30_8-6-59.png


This also doesn't work. It will be easier for red players to flush blue players out of here, but the worst-case spawncamping scenario is a demo lobbing stickies to the spawn door and just waiting for red to respawn (which they can track by just watching the scoreboard). This change doesn't really do anything to fix that. We could also consider moving red spawn entirely to a different side of this area, but the same demoman can still do the same thing. (I didn't even bother mocking this up.)

The conclusion that I've come to is that, because of the wide variety of classes and weapons in TF2, there is simply no way to give blue a shortcut route directly to red spawn and make it not abusable by at least one class. That sounds really obvious in plain text, but sometimes you need to do a lot of thinking to make sure your simple answer is correct.

So, the solution is to just block off this route. Let blue use the sewer route to get under the point and flank behind a red forward hold, but don't let them walk directly to spawn. A transparent barrier like metal bars or grating will help keep some element of risk that I want for this route. Blue is in a lot less immediate danger with this route, and it also removes options for red... on the other hand, it will make a red hold in this back area significantly safer. This change will probably require some balance changes elsewhere to make it all work, but removing a match-ruining strategy should be worth it.

upload_2021-1-30_8-13-29.png
 
Last edited:

Idolon

they/them
aa
Feb 7, 2008
2,105
6,106
b8 s1

STAGE 1 ONLY VERSION

OVERALL:
-minor detailing adds

A:
-no changes

B:
-added grate to lower sewer route to help prevent blue team from getting to red spawn too easily / quickly
-to compensate for red losing a route to the area behind B, added new route from spawn to behind B (opened previously out of bounds area)

Read the rest of this update entry...
 

Idolon

they/them
aa
Feb 7, 2008
2,105
6,106
b9 s1

STAGE 1 ONLY VERSION

OVERALL:
-clipping and optimization pass

B:
-changed big rock to big silo to simplify geometry / collisions and improve aesthetics
-added grate to main bridge to give red an additional "early warning" opportunity against blue flankers
-increased door height at far flank
-decreased red respawn time from 10s to 8s
-optimization pass includes adjustments to geometry for better visleaf structure

Read the rest of this update entry...
 

Idolon

they/them
aa
Feb 7, 2008
2,105
6,106
It's been a few versions since I originally set out to tweak stage 1. When I started, my goal was to hit a 66% - 75% win rate for blue. How close have I gotten?

upload_2021-3-10_12-31-54.png


As you can see, previous versions had pretty bad win rates for blue. Since beta 7, the blue win rate has rocketed up and has not gone back down since. I want to attribute this mostly to the change of the sewer dropdown to stairs, making a flank route for blue much more effective - however, this change was made in beta 6, not 7. I'm still guessing it is because of this change, but it took another version for players to notice.

There have been 31 matches total since beta 6. In the 31 times that A has been contested, red has defended it 6 times, for a blue win rate of about 81%. In the 25 times that B has been contested, red has defended it only 2 times, for a blue win rate of about 92%. Overall, blue has won 23 times, for a win rate of about 75%. This is exactly what we were aiming for!

However, given the above theory that the sewer route change is the major cause of change in win rates, and that it took a version for players to take advantage of the flank, we should take the average win rate since beta 7, not beta 6. This blue win rate is 21 / 26, or about 80%.

I think this is good enough, because red is able to defend some of the time, and despite whether or not red is able to defend, people seem to be having fun. I also have a hunch that red will slowly get better at the map, as it seems that the back balcony area is an effective hold position, but it is not an obvious one. As far as I'm concerned, I've achieved roughly what I set out to do with regards to win rates, but I am going to continue to tweak B on stage 1 to make sure it is fun to play as red.
 

Idolon

they/them
aa
Feb 7, 2008
2,105
6,106
b10 (includes both stages!)

(changes since b4)
-major changes to layout of stage 1 B
-detail and optimization pass
-redo of several "propper" models necessary to get full map under engine limits, improved lighting and collisions as a result

Read the rest of this update entry...
 

Gravidea

L5: Dapper Member
Apr 17, 2017
227
102
Hey there! B10's looking pretty good so far, and I'm liking the visual changes so far as well! Just one thing I've noticed is that these two doors on Stage 2's A (shown below) open, and then do not close afterward no matter how far away you move from them. Not sure if this was intentional, but I assumed it wasn't. Good luck with the map! I'll be keeping an eye on it.
upload_2021-3-17_22-26-41.png
 

Idolon

they/them
aa
Feb 7, 2008
2,105
6,106
Hey there! B10's looking pretty good so far, and I'm liking the visual changes so far as well! Just one thing I've noticed is that these two doors on Stage 2's A (shown below) open, and then do not close afterward no matter how far away you move from them. Not sure if this was intentional, but I assumed it wasn't. Good luck with the map! I'll be keeping an eye on it.
upload_2021-3-17_22-26-41.png
This is intentional, the idea being that it helps teach players what doors will open. Gorge also has doors like this.
 

Idolon

they/them
aa
Feb 7, 2008
2,105
6,106

Idolon

they/them
aa
Feb 7, 2008
2,105
6,106
I published b10a and forgot to actually read feedback for bugfixes, so b10b is a follow up patch for some other little things. I also spent ages trying to optimize the water and got nowhere.
-fix cringe peek at 2A helipad
-fix stuck spot in stage 1 blue spawn

Read the rest of this update entry...
 

Idolon

they/them
aa
Feb 7, 2008
2,105
6,106
upload_2021-5-1_12-13-35.png


The stairs in this lobby got revised several times in alpha, and they're due for another look. (Thanks to Bec and abp to bringing this room back to my attention.) They're currently in a spot that's a compromise - designed to be used by red for defending A, and also designed to be used by blue for attacking B. They work well for red, but blue tends not to use them because they're out of the way and run backwards. This is intentional, as I want the rotation to the high ground to be slower. I'm also fine with having routes that aren't immediately obvious or intuitive, because I know that players will learn a layout over repeated plays, and some layouts are simply not possible if everything must always be perfectly intuitive.

However, what I'm finding is that the high ground in this lobby is too important to leave this obscure. This isn't a flank for sneaky players, but instead one of only two routes that aid in pushing the tightest choke in the map, and arguably the more important of the two, since having high ground presence in the next lobby is essential for holding forward as either team. The way up needs to be more intuitive to help guarantee that more players use it, because otherwise the offense falls apart, and the blue team can't even tell why.

(Side note: This is a good example of how playing your own map can skew your testing data, as I often use this route for pushing B. By playing my own map the way I think it should be played, I don't get to see what happens when players are left to their intuition.)

So, what to do about this?

upload_2021-5-1_12-49-47.png


I can slide this stair side to side like this to make it more obvious to blue, but this doesn't solve the awkward flow, it gets in the way of the main path for blue on the low ground, it creates a very tight u-turn to the path into the next lobby (where fights will happen), and also hides it away from red when defending A. If we're keeping the stair in this rough area, its current position seems to be pretty much ideal.

upload_2021-5-1_12-52-6.png


I could try moving the high ground entirely to make it way more obvious to blue. However, this makes it way more effective for red holding forward here, causes sightline issues when flowing into the next lobby, and is awkward flow for red defending A. It also looks bad, and would interrupt the main open space of this lobby, making everything feel more cramped.

upload_2021-5-1_12-54-27.png


There's the entry room for this lobby, as well as the raised area in the back. Any stair here would be a fast rotation time and have an awkward fighting angle into the doorway to the next lobby, so I don't think anything here works.

We have one last area to use, and I think this makes the most sense:

upload_2021-5-1_13-2-44.png


I think part of why this change wasn't obvious to me up front is because I had framed the issue as, "How do I design a stair that works for both teams?" This ignores the possibility of simply having two stairs, one that works for each team. This change may upset balance for any number of reasons, but giving blue a more intuitive route upstairs is my priority.

---

An unrelated change I'm considering:

upload_2021-5-1_13-41-53.png


Not sure what will come of this. It would solve an awkward fighting angle, and it would also make this route a more valuable forward hold for either team, which seems like a good idea for making this point more intuitive and dynamic. It just also upsets the existing flow and balance in a way that I'm unsure how to predict, so it feels pretty risky. I'll probably try it regardless.
 

Idolon

they/them
aa
Feb 7, 2008
2,105
6,106
STAGE 2 ONLY VERSION

STAGE 1 (not compiled):
-changed forward spawn door to use no entry skin before A is capped

STAGE 2:
-added new route to high ground from lobby to pyramid room so blue can more recognize that the high ground exists
-removed boiler from B flank
-widened blue-side doors into B flank
-addded small ammo to cliff route near blue spawn
-added pallet in front of a door to better indicate it cannot open
-changed texture of underside of pyramid room bridge for better visibility

OTHER:
-misc. clipping fixes

Read the rest of this update entry...
 

JeanPaul

L6: Sharp Member
Aug 5, 2010
284
279
Final in its current state feels borderline impossible with a decent red team. In the latest playtest I noticed it seems the most important area to hold as red is this back alley:

20210620165550_1.jpg

The problem is red can get a very strong hold here and shut down all blue attempts to get into the area. I think you realized this a little bit as you gave blue an elevated entrance into the area:

20210620165555_1.jpg

But in order for any blue to reach this balcony from the lower area takes a 10-20 second walk depending on which route you take:

20210620165701_1.jpg

This route also puts you in front of three potential sniper sightlines from inside the final cap area.

So the main diagnosis of the problem is blue needs help in capturing the back alley.

My first idea would be to try would be adding a staircase for blue to access the top balcony immediately from the lower area:

20210620165555_1 - Copy.jpg

I noticed that means youd have to get rid of this hallway:

20210620165836_1.jpg

I actually think removing this hallway might help blue push into the general choke area as well :p

EDIT: Extending the elevated balcony to reach further into the alley also might help blue a bit too
 

Idolon

they/them
aa
Feb 7, 2008
2,105
6,106
Final in its current state feels borderline impossible with a decent red team. In the latest playtest I noticed it seems the most important area to hold as red is this back alley:

...

I noticed that means youd have to get rid of this hallway:

20210620165836_1.jpg

I actually think removing this hallway might help blue push into the general choke area as well :p

EDIT: Extending the elevated balcony to reach further into the alley also might help blue a bit too

I agree this back alley is currently one of the major issues with final, though I'm not quite sure on what the issue is exactly. I've been thinking about removing that hallway in question so that I can widen the space and make it less cramped, but I'm not confident that this will actually fix things. This flank was intended to be the way for flank classes to get in and do their thing, but it provides access to two of the better ways to push last. Blue's top balcony is a weird place to push from because it doesn't have good sightlines onto where red is actually holding most of the time.

In other words, the flank's problem might just be that it is serving the wrong purpose and the overall layout doesn't make sense.

That ties in with another complaint I've seen, which is that red just doesn't have a lot to do on B. Main is good and I think a fun area to play, flank is cramped and dangerous, and the back balcony area is smart to hold but ultimately requires sitting and waiting for blue to come to you, which isn't the most fun. What I think I'd really like to do is give red other good things for power classes to do so that they have more fun and leave the flank alone for the classes that really need it. I think the way to do this is to make blue's upper balcony more valuable as a way to push last, rather than just valuable for pushing the flank. (And it's not even great at that! The walk time is intentionally long and fine imo, but the area is easy to miss as a blue player and easy to splash as a red player.)

I'm going to just start throwing some ideas out and see if I get anywhere:

upload_2021-6-20_17-43-53.png


Turn the high ground into one contiguous area so that blue can use it more aggressively for attacking the point (rather than just for the awkward angles it gets them onto red's defense), and in turn allow red up there as well. This would open up a new area for teams to duke it out, hopefully making it more fun for everyone and also easing pressure on the flank.

This idea worries me because older versions of Snowbase tried similar things, and it showed that giving red access to the high ground over the lobby so easily was too powerful. Here's an image of a1, where red can get directly to blue's balcony by stairs on the far side of the point:

upload_2021-6-20_17-49-42.png


I'm also not so sure about giving red such quick access to the back balcony from spawn. Part of the reason why the flank is the only access to the back flank is because it is so valuable, and a quick rotation (with a quick retreat) makes it just too good. Again, an image from a1 where this was the case:

upload_2021-6-20_17-55-0.png


I don't see a good way to make this back balcony easily accessible to red without also making it overpowered. So, my next thought is to make it less powerful. What if the whole thing was lower?

upload_2021-6-20_18-2-29.png


After drawing this image out, I'm feeling pretty good about this idea. Red gets a new holding area that isn't super powerful, but is still useful. In addition, if red is holding behind the point here, then the blue balcony actually gets a more direct angle onto red's defense. I'm not sure how this ties into anything I've just said about the flank, but I think I'll figure something out eventually.

This change would also have a funny consequence at blue's spawn rollout, but this is probably fine? Might be an opportunity for signage that helps draw attention to blue spawn??

upload_2021-6-20_18-12-3.png




This post is barely even a response to yours, but it helped me work through the problems with this point. Hopefully I'll have a new B out sometime within the next month that's more fun to play.
 

JeanPaul

L6: Sharp Member
Aug 5, 2010
284
279
All good points really.

I think the bridge is a great idea and so is lowering reds back holding area. Letting red get onto the elevated back area (first paintover with the arrow question mark) works only if you lower it as both the height and direct access would probably suck like you mentioned.

I like the idea of being blue and successfully getting to the back of the alley flank and being rewarded with getting the drop on a few red players.

EDIT: another idea could be making the bridge but only half the width of the room so blue can attack/clear out the alley flank without having to do the weird 180 degree drop down turn thing
 
Last edited:

Idolon

they/them
aa
Feb 7, 2008
2,105
6,106
b12 s2

STAGE 2 ONLY VERSION

B:
-Added new route from blue balcony that extends across last to the back balcony
-Lowered back balcony by 128 units and opened it up directly to red spawn exit
-Lowered red spawn by 64 units
-Removed high ground option from flank, made exit to point more covered (thanks to new route mentioned previously), expanded room size by removing underused route from garage
-Removed door on main choke into B to allow for safer pushes, create sentry closet
-Lowered point into ground by 32 units to make consoles less annoying to shoot over

OTHER:
-clip pass
-optimization pass

Read the rest of this update entry...
 

Idolon

they/them
aa
Feb 7, 2008
2,105
6,106
With images, here's what changed in b12:

cp_snowbase_b12_s20005.jpg


The back balcony has been lowered by 128 units, making it crouch jump height from the point. Red can now access it directly, though it is a one-way for most classes. (There is medium health/ammo behind the consoles.) This back balcony has a stair directly to a new high ground path across the back of the point area, which connects directly to blue's balcony. The idea here is that red players have a long commit to this route, which should balance it against the height advantage they get.

Spawn has been lowered 64 units so that the height is less extreme. Now that red has another way out of spawn that doesn't funnel into a single choke, they don't need such a height advantage over that choke. This should help spies maybe??

The point has been lowered 32 units so that the consoles are easier to shoot over. I toyed with removing the consoles entirely, but the point felt barren, both in terms of gameplay and aesthetics. Lowering the consoles should allow them to serve a gameplay space without also being too cumbersome. I also just really like how this looks.

cp_snowbase_b12_s20006.jpg


This new high ground turns the flank area into a room, which looks nice. This also helps with navigation, as you can no longer dead-end yourself when leaving spawn to go to A, as the closed doors are obvious and visible. The arrow sign at round start blocks out the entire window.

cp_snowbase_b12_s20007.jpg cp_snowbase_b12_s20008.jpg

The flank area itself has been enlarged to have a second, more covered route. This should help flank classes prevail in combat more easily. In addition, there is no longer an important high ground route here, and red has a newer, more important route to hold besides this one, so I hope that this loosens things up a bit.

cp_snowbase_b12_s20009.jpg


This area got changed to make room for that new flank. I am a bit worried that this might make a red forward hold too effective since blue is more limited on options, but the route lost here wasn't used very much anyways. It's probably fine.

Overall, I'm very happy with this update. I'm fully expecting there to be problems that need fixing, but I've known for probably a year that B needed changing up, and I've finally found something I'm happy with.
 

Idolon

they/them
aa
Feb 7, 2008
2,105
6,106
Some takeaways from tonight's test. We had three rounds:
- One where red held A
- One where red held B, and blue barely ever got access to the point room
- One where red held B, where blue made multiple pushes into the point room and made good progress but was unable to win
I want to run this version again to see if blue ever wins, and to get more data on how people are playing.


I managed to accomplish some of the goals I was after! Defending B is more fun than it used to be. The B room itself is more fun to fight in than it used to be, I think. The flank area sees less traffic than it used to and is more useful for blue flanking classes.

As predicted, I also introduced some new problems. The new balcony is way easier to hold for red than I anticipated. It is a good place to hold the point itself (as you can easily see blue players coming in from main) and it is also a great place to push into lobby. Blue's counter-strategy is to also play the balcony, and so the gameplay is a bit stale.

Letting red up onto the balcony from their side of the map might be a mistake. For comparison, Gorge B has roughly the same set of routes (main, flank, high ground) and doesn't let red up to the high ground. Why does Gorge work? The answer, as best I can tell, is that Gorge B's main accommodates way more players, and all you really need is two (and a half) routes to give everyone something to do. Snowbase B main is just a 256 wide corridor of spam and isn't very fun for red.

I bring up Gorge B because the lobby for Snowbase is modelled off of Gorge B's lobby, and that lobby was not designed for red players to be on the high ground frequently once blue has pushed that far forward. I don't think allowing red onto the high ground this easily works at all. Maybe this idea could work if I redid lobby to make the high ground somehow a more natural hold, or less useful for red, but I'm not sure how to do that very easily.


One other issue that came up is that holding forward as blue is difficult because red always has a good angle. I think this is partially because of the balcony issue that I've created, but also because of the location of the flank. When red uses the flank, they end up coming into the side or even behind where blue team is holding, which is very disruptive. The only fix I can really picture is adding a one-way, which I'd rather not do. There's not really any other place in the layout for a flank to go.


The best idea I have right now to address these issues to mirror the entirety of B, putting the flank in a more natural position for both teams (and also freeing me up to do more interesting things with the geometry of the flank). I'm not super sure of this idea, especially because it would even further complicate the navigation from red spawn to A, but it is currently the best idea I have for fixing some of the fundamental issues I'm seeing with B.