Map layout problems

Discussion in 'Mapping Questions & Discussion' started by Simulacron, Apr 4, 2016.

  1. Simulacron

    Simulacron L6: Sharp Member

    Messages:
    305
    Positive Ratings:
    173
    I have a question:
    In which part of the mapping progress your layout becomes really "good"?

    I submit two of my maps to the gamedays, but the first one was to open for snipers (pl_timberbridge [Deleted]), so I gave up on it. My second one was a small koth map (koth_canyon), in this map I tried to reduce the sniper lanes to a minimum, but by doing this the map became to tight, cramed and clunky. My question is if problems like these happen to all maps in development or if I'm just too self-critical and give up to fast.
    These two examples are just the two I submited, but I have a about ten maps that just collect dust, because I wasn't happy how the map layout was!
     
  2. sooshey

    aa sooshey :3c

    Messages:
    522
    Positive Ratings:
    418
    No matter how much you study layouts, they're always a trial-and-error process. As you use hammer more and keep making maps, it should come to you. Having only made 3 maps (two of which have been tested), I'm still a long way from making "good" layouts, so I might not be the most qualified person to say that :p

    You should try to aim for a mix of sightlines and tighter side passages that players can use to flank the snipers. Take upward for example, a red sniper can stand on that building near A and shoot straight into blu spawn, but there are ways to stay out of the sniper's field of vision and get to them.
     
    • Thanks Thanks x 1
  3. Simulacron

    Simulacron L6: Sharp Member

    Messages:
    305
    Positive Ratings:
    173
    Thank you for the sniper advise, that will change how I look on (my) maps.
     
  4. UKCS-Alias

    aa UKCS-Alias Mann vs Machine... or... Mapper vs Meta?

    Messages:
    1,264
    Positive Ratings:
    748
    dont fully focus on such sniper paths aswel, since there is still a situation where other classes can become OP. The sniper is simply the most complained about.

    Instead its usefull to analyze the idea before even building. drawing on paper helps for some, some just keep it in their head for a few days while the idea forms. Rarely should you instantly build, even if it seems the most direct approach to fix a problem. The only thing you do is shift the problem to something else and make the map more complex (as the easiest fix is often just 'add prop' or 'add door/hallway'). Keeping your map simple remains more important than fixing every single sniper sightline. Look at badwater at the amount of sniper sightlines for example.

    A map often features a direct path and a direct flank, on upward the direct flank is the right side. And the reason it works is because that flank is very hard to be covered at the same time as the main. By this logic you can essentialy make a map using just 2 paths. But as experienced before, both the main and alt flanks can often be easily covered, and for these you will check which areas are the key holding points. You design counters to those and will also check on ways they cannot be abused by defenders that easily. But if we look at viaduct notice there are 2 main paths people take (the sides because of the higher ground advantage they give), the middle is only used when they dont specificly need to go to the side and want to flank instead. These low sections often provide cover which can be used to take out the ones on the higher ground by for example sniping.

    To force the requirement of flanking you often make a sniper/sentry spot that can lock down the main path quite well (a well performed uber must be able to get through and shouldnt be forced to walk around the path again - an uber should mean a dead sentry). People will not take this path knowing they would die, they will take the direct flank, or attack from a distance. This flank however on its own often is weaker to defend by providing a better cover for blue, but when red would fully focus on that path, its locked down just as wel as the main.

    For this you have the smaller flanks, and they are ment to counter specific holding points. On upward the left flank allows you to shoot quite safely into the engy spot, even though keeping a little spot blind. This makes a huge sentry nest quickly destroyed, but a lone sentry can hold out well. But this capability makes it an effective flank.

    Also, in A/D maps the closer you reach the hatch the more safe zones for red start to appear. And many times change the major flank to a minor one to allow red to focus the defense a bit more, which also reduces the need for a larger sentry gun nest - although in some cases that nest still helps (dustbowl stage 3, goldrush stage 3).
     
    • Thanks Thanks x 2
    • Agree Agree x 1
  5. RaVaGe

    aa RaVaGe

    Messages:
    728
    Positive Ratings:
    1,054
    Don't listen too much to playtesters on tf2maps, especially when it comes to sniper sightlines. Just take a look at how valve's maps are made, having a long sightline is fine, as long it is tight.

    Anyway, creating the perfect level is never easy, and you will need to iterate your work, the main problem is to understand why it is not working, once you get that, you can make the changes that are needed.
     
    • Agree Agree x 2
    • Thanks Thanks x 1
  6. Benoist3012

    Benoist3012 L3: Member

    Messages:
    147
    Positive Ratings:
    201
    As said RaVaGe look at valve's maps. And compare them to your map.
     
  7. Moonrat

    aa Moonrat The end of an era

    Messages:
    896
    Positive Ratings:
    530
    666 MESSAGES YOU ARE THE DEVIL
     
    • Agree Agree x 2
  8. Illius

    Illius L2: Junior Member

    Messages:
    65
    Positive Ratings:
    8
    I never stopped and considered that. That's something I'll keep in mind, actually.

    On a note for the OP, don't get frustrated on scaling. I've seen big name mappers (Such as UEAKCrash, where he made the final point waaaaay too small) screw up scaling from time to time. It's about fixing the problem after you find it, not getting frustrated at it. Experience will help mitigate it but so many people screw up scaling it's hardly something to fret over.
     
  9. IrishTaxIDriver

    IrishTaxIDriver L6: Sharp Member

    Messages:
    271
    Positive Ratings:
    144
    Never. You'll tweak it until you can't anymore. No one is 100% happy with a layout.

    What you're asking is how do you become comfortable with a layout where you're okay with other people seeing it and providing feedback on it.*

    For me its when I feel the layout represents "The Idea" to a point where other people can understand it without prodding. Let some friends (or the fine folks here) run through it at that point and give some feedback, make iterations based on those (or acknowledge concerns), and repeat a few times.

    Once that stuffs done, you need to take the leap of faith and get it tested. The worst thing that happens is the map doesn't work, you don't want to save it, and you shelve it. Use what you learned to make the next map better. Its only a failure if you don't learn anything.

    Keep the layout in a state where you feel comfortable rapidly iterating on it. This means different things for different people. I've seen people be completely comfortable iterating on final detail maps, I've seen people only be comfortable iterating with dev textures. Everyones process is different. You need to be able to get a decent turnaround time for feedback. A good rule of thumb is a week of iteration time between tests, even if you get a small amount of feedback. If you finish your feedback list early, take a break and do something else for a little while and come back. You'll see things you missed.

    Feedback will be vitrolic. Feedback will be shitty. You need to not take it personally. Look for the nugget of truth, and try to address that.

    Sorry for a long dramatic rambling answer, but it all contributes to a layout being "good". Just practice, be open to change, and eventually you'll start cranking out home runs.

    *For the TLDR shortcut, when I catch myself doing detail. If I'm doing detail then I know I'm okay with the layout on some level and have accepted the consequences if it needs to change.

    EDIT: A note about Valve maps, don't take them as gospel. Use them as guides for what to do or what not to do. The best maps bend or even break hard level design rules. Also, look at how many maps or gamemodes they've stopped working on or haven't touched in a long time. Even the pros struggle with this.
     
    • Thanks Thanks x 1
    • Useful Useful x 1
    Last edited: Apr 5, 2016
  10. Hyperion

    aa Hyperion L16: Grid Member

    Messages:
    810
    Positive Ratings:
    617
    Try and try again and make improvements based on the feedback. Personally I work with first version very long, I almost get stuck to testing all sightlines and sentry spots etz... I don't suggest overdoing but playing even with bots can help you notice problems with the map. Usually bot snipers find long sightlines and sometimes engineers find good sentryspots but real people are by far the best.

    When you start to improvement your map, remember that props aren't very efficient way to reduce sightlines in a long run. Also people may get frustrated if you make only small updates to your map