tc_hydro study

Discussion in 'Team Fortress 2 Talk' started by fubarFX, Dec 28, 2010.

  1. fubarFX

    aa fubarFX The "raw" in "nodraw"

    Messages:
    1,597
    Positive Ratings:
    1,677
    so I've been studying hydro's layout lattely
    more precisely, the number of paths given in every given situations

    I noticed some strong patterns in the number of paths each combination offers

    first lets start with the 4 main combinations

    [​IMG]

    the first noticable pattern is that most combinations offer 3 entry points to the capture area
    which is pretty much the general rule in tf2. note that the only exeption to this rule is in the dam/dish combo.

    now lets take a look at the 2 crossed combinations

    [​IMG] [​IMG]

    the second noticable pattern here is that cross map combos feature only 2 entry points to each capture areas
    while both combos feature only 2 entries to the capture area, both paths are leading to the same battlement where flanking is out of the question

    hydro and stalemates
    from what I can recall of my matches on hydro, I see a correlation between problematic combos and combos featuring only 2 path to capture area.
    my unclear memories of previous matches and recent observations lead me to believe that the lack of flanking paths is directly connected to the number of stalemate happening in a specific combination.
    (that actualy makes a lot of sense) :O

    conclusion:
    2 paths to capture areas = stalemate friendly = not fun
    3+ paths to capture areas are required for a combination to be good

    now lets take a look at both final capture areas

    [​IMG]

    3rd noticable pattern, attacking/winning team always have many paths at their disposal.
    Even if the defending/losing team could attempt a capture at the locked point, the layout would totally destroy any
    chances of a successful offencive due to the low number of paths they can push trough
    everything in the layout here is against the defending/losing team (even height variations)
    this support the idea that valve implemented these final round as "punishment rounds" for the losing team
    while it's not a bad idea, these rounds usualy end up in steamrolls and are not that much fun for any of the teams

    final summation
    let's assume that:
    -the only good combinations are the ones featuring 3+ path
    -every 2 path combo are bad because they end up in stalemates
    -final stages suck because they're glorified steamrolls

    [​IMG]

    hydro ends up being a tc that has only 3 good combinations out of 8 possible combinations
    hence why it sucks?

    on a side note, the dish/dam combo being stalematish, I've wondered if balance was affected.
    stalemates on this combo would theorically help red's defence of either point should any of them be captured by blu.
    not sure if it's directly affecting balance but the stats support the theory.
    [​IMG]
     
    • Thanks Thanks x 14
    Last edited: Dec 28, 2010
  2. littleedge

    aa littleedge L1111: Clipping Guru

    Messages:
    983
    Positive Ratings:
    573
    This barely helps me in thinking how a tc map should work beyond giving me a place to look at how many paths inbetween each set. Thanks for that.

    Also, the stats thing has been crazy since it was created. Take it with a grain of salt and with a doubtful mind.
     
  3. fubarFX

    aa fubarFX The "raw" in "nodraw"

    Messages:
    1,597
    Positive Ratings:
    1,677
    how a map should work is entierly up to you. there's a lot of room for improvement in the tc formula.
    I don't really want to mislead people into thinking there's a good way to do tc beyond the number of path they should give (wich I really can't stress enough)
     
    Last edited: Dec 28, 2010
  4. Peachrocks

    Peachrocks L1: Registered

    Messages:
    10
    Positive Ratings:
    0
    I've stated this numerous times elsewhere but the biggest reason Hydro/TC stalemates is that the map breaks the very rules that were said in the commentary about how to prevent stalemates. Still, I think there's a lot to learn from Hydro in how NOT to make a map.

    Check one of the Well nodes, I think its fairly early on. They give the attackers faster respawn times in a 5 CP game mode to prevent stalemates and allow them to go on and make the final push.

    In Hydro regardless of whether you are turtling as an Engineer, defending as a sniper or trying to counteract them as a Scout or Spy you ALL have the same respawn times but you do not have the same travel times to start achiveving what you need to do.

    It's sort of like why instaspawn on Attack/defend maps favours the defenders so strongly, except worse because the 'attackers' will always be outnumbered, both by players actually defending and players who have just respawned and they almost certainly have no assistance in the form of teleporters from Engineers.

    The fact there are so few ways into to certain areas, and the fact that the map does in general have a lot of 'tough to break' Engineer spots does not help matters, but Hydro's problems stem from many many areas.

    Simply put. Problems with Hydro stem from the game type and Hydro itself.
    Hydro's problems as a map are...
    - Very easily defendable
    - On the most part the corridors are probably too tight
    - Last area is way too easy to attack

    Territorial control problem is...
    - Teams recieve the same respawn times regardless of what its players are actually doing. This favours defensive/slow paced play, which on the most part is undesireable as it leads to stalemates.

    To resolve TC's problems, you simply need to turn it in a 3 CP mode. Capturing the center unlocks the defenders final point and gives the attackers a faster respawn. In Hydro's case this would look hideous, but TC would work, if you removed the stalematishness and 'win after one fight only' about it.
     
    Last edited: Dec 31, 2010
  5. Icarus

    aa Icarus

    Messages:
    2,246
    Positive Ratings:
    1,181
    TC logic starts to fall apart when you include more than 2 CPs per round. Doubly so if you have to work around the hard 8 CP limit.
     
  6. Mr. P. Kiwi

    Mr. P. Kiwi L5: Dapper Member

    Messages:
    244
    Positive Ratings:
    95
    A thing that bothers me most about Hydro is the tight cliff fights. I know it's necessary to have cliffs because of how big the map is, but if you'll find a way to bypass it, the maps will be much much nicer. Hydro, is one of the biggest sentry nests traps in TF2. Although having multiple routes all of them are really small. One big one and one little one should do the trick.

    On the other hand, one of the reasons I like Hydro is because the routes between the points... I really like the cliff that it doesn't matter what point you attack or defend you always have a familiar place which you can sigh in relief: "I know this place, now let's go to my favorite hiding place here (which I know nobody else knows) and breath for a few seconds." Hydro has done that pretty well. I know there are two levels of the cliff, but you know it's the same place. I'd also like to point out the deathpit that shows you the other pass...

    These are the thing I thing you should take this in mind when making a TC:
    1. Make the points attackable, not just defendable
    2. Find ways to make big open routes that a sentry can't dominate
    3. Multiple routes to each point - not all going through the same place
    4. Make sure that one path between all the combinations will have a common area

    Well good luck to all you people who want to make a TC!
     
  7. Peachrocks

    Peachrocks L1: Registered

    Messages:
    10
    Positive Ratings:
    0
    It's either make more than 2CP per round or have nearly every game end with a Spy/Scout cap, ends after one battle after most of one team is dead and is unable to defend the single control point or a stalemate because defensive players are cautious of the previous two and then when one side does it, the other has to reply if they don't want to lose which results in a stalematey game.

    Even if you open it up a bit and make it less favourable to defending players it would help but it would still be not a lot of fun simply because there is no respawn difference.

    As I said to start with TC contradicts Valve's very own design rule said in the Well commentary node, fix this problem or any TC map will be mediocre at best.

    Another way around it is making it single cap, either KOTH style or a rather long capture time. Mind you I don't know if any of this is possible... I didn't even know about an 8 CP limit.
     
    Last edited: Jan 1, 2011
  8. honeymustard

    honeymustard L9: Fashionable Member

    Messages:
    698
    Positive Ratings:
    273
    You do realise that they did fix the spawn times a few months ago?
     
  9. tovilovan

    tovilovan L6: Sharp Member

    Messages:
    391
    Positive Ratings:
    94
    I've been considering making TC a kind of capture point race where the capture progress doesn't decay. I'm not sure this is possible, however.
     
  10. Peachrocks

    Peachrocks L1: Registered

    Messages:
    10
    Positive Ratings:
    0
    You are missing the point. For as long as both teams have the same spawn time, a problem will exist. A player on defense will always be treated more kindly than a player on attack simply because they have less walking distance to their objective.

    Ugh, I'm trying to find a TL;DR version of this but there isn't one.

    Take this example.

    8vs8. Both teams have 2 Engineers, 2 soldiers, 1 medic, 1 scout, 1 heavy and 1 Spy.

    The Engineers on both teams are commited to the defence. The rest are commited to an attack. Now say that 2 of these attackers are dead on both teams and are respawning, lets say the Scout and the Soldier.

    Team A attacks with a medic, soldier, heavy and a spy.

    Team B is defending with a medic, soldier, heavy, a spy AND two Engineers AND the two respawning players are going to get to the combat sooner on the defence than the same will happen for the attack.

    Take the same situation in an attack/defend game.

    Team A will have a medic, soldier, heavy, spy AND engineers providing reinforcements through teleporters and perhaps giving it a go fighting themselves as they do not have to defend too excessively.

    Team B will have exactly the same defence as earlier but longer respawn times.

    See the problem with TC now? 5 CP doesn't have this problem because it's not all over if one point is captured and the teams that have more points in their control have faster respawn times. The losing team should know this, and play defensively to compensate for the weakness until the attacking team attacks too rashly. With TC, none of this can happen, and that's why it drags and stalemates typically.

    Instaspawn near the end didn't help things sure, but fact is the problem exists elsewhere.
     
    Last edited: Jan 1, 2011
  11. TMP

    aa TMP Abuser of Site Rules

    Messages:
    948
    Positive Ratings:
    551
    No, what he means is, if you have made progress on the cap of the enemy team in Hydro, your spawn time lowers, effectively helping to counteract what you're saying.
     
  12. Peachrocks

    Peachrocks L1: Registered

    Messages:
    10
    Positive Ratings:
    0
    Ah I see. However... it doesn't do anything if there is no progress made and the points in some cases are tricky to access especially if there is a sentry and/or defensive Sniper on them.
     
  13. Jimmy

    Jimmy L7: Fancy Member

    Messages:
    421
    Positive Ratings:
    222
    I love how my whole map study thing has taken off :D
     
  14. YM

    aa YM LVL100 YM

    Messages:
    7,099
    Positive Ratings:
    5,739
    Just so you know, you're totally misrepresenting the diagonal rounds which, while technically have two routes, actually only have one.

    In the middle of both of the diagonal rounds, no matter which of the two routes you pick you have to pass through a common <1024 unit wide area, there isn't even any height variation in these spaces.

    This effectively makes them single-route paths because ALL the traffic has to squeeze though one single chokepoint on the way to the other side.

    You're analysing it as though 2 paths isn't a good situation when really 2 paths is fine (as is 3/4) but one single path causes problems.

    The Dam-Dish section is fine apart from those two points being poorly laid out for attacks from each other, (Dish has that hugely OP ledge above both exits and they're close) and the dam overlooks both exits from a raised point (highground).

    The reason for hydro's diagonal rounds being an utter disaster is due to there simply not being enough space to have two routes each. An alternative was suggested in chat just the other day that solves the spatial issue whilst only removing one possible round by offsetting the areas by half.


    Main point is:
    3 = Ideal. - main routes
    1 = Horrific torture, what the hell are you doing you evil bastard. - diagonals
    And:
    2 = OK but just make sure you can't ultra easily cover both exits at once. - dish/dam catastrophe
     
    • Thanks Thanks x 2
  15. DaBeatzProject

    aa DaBeatzProject

    Messages:
    1,276
    Positive Ratings:
    1,091
    I think the optimal way is to just make both diagonal routes go through the same "Area" and not on different height levels or next to each other or whatever (=blending them together). Maybe valve did use two different levels though to have more variation (even more??), but it doesn't really work out well. So what I'm going to do is make both diagonal routes go through the same area, but with several walls/doors changed. That means I (should) have more space for fighting, sideroutes and so on.
     
  16. LeSwordfish

    aa LeSwordfish semi-trained quasi-professional

    Messages:
    4,102
    Positive Ratings:
    5,991
    One thing i'd like to see with TC would be a central Arena-style area in the center, for both the diagonal routes- in my minds eye, this looks/plays like nucleus. This could also be utilised for extra routes around the other four points.
    By sharing space, the two rounds can be larger and less of a clusterfuck.
    Plus, this creates the ability for a single central landmark- landmarks are what i feel TC maps miss out on. Both Hydro and Meridian go for making each point a landmark, which has the effect of devaluing the awesome dam and robot. If the robot was in the center, being fought around in 2/8 rounds, and visible as a side-route in 4/8, it would be way more interesting and focal. You can then also have landmarks for the final points for each team.
     
  17. Jimmy

    Jimmy L7: Fancy Member

    Messages:
    421
    Positive Ratings:
    222
    Not a half bad Idea, if I ever get round to finishing bedrock I'll probably try something like that, even just as a wee experiment... also is that Garrus in your avatar? I love the mass effect games! /offtopic
     
  18. LeSwordfish

    aa LeSwordfish semi-trained quasi-professional

    Messages:
    4,102
    Positive Ratings:
    5,991
    yes, it is, and i'm glad you like the idea.
     
    • Thanks Thanks x 1
  19. tyler

    aa tyler snail prince, master of a ruined tower

    Messages:
    5,031
    Positive Ratings:
    3,975
    The idea of a TC map that utilizes even more shared space is really awesome. Really really awesome.

    I wish Valve would refine the idea.
     
  20. LeSwordfish

    aa LeSwordfish semi-trained quasi-professional

    Messages:
    4,102
    Positive Ratings:
    5,991
    An idea that i've rejected is one where the central area is ALL the routes, with the points being off from that, but i doubt any area could be interesting enough to make 90 degree angle changess feel like different rounds.