Hacker News Reader

Comment by ozmbie | Hacker News Reader

I’ve worked as a level designer on FPS games. Dynamic generation might suit certain types of games (or something like an event mode), but a huge amount of design work and iteration goes into the layout of multiplayer game levels to make them enjoyable.

It’s also important that players/teams can memorise the level layouts in order to form strategies.

Still a cool idea though!

View on HN

I haven't designed levels, but playing maps across a few games, I'm constantly amazed at how well small details are thought out. A seemingly random crate actually blocks off a line of fire that would give one team a major positional advantage; a tree breaks up the line of sight between two objectives; a decorative fire provides visual cover in an otherwise overly open lane; a curved passage has exactly the right proportions to give cover on one side while being open on the other, letting players choose when to engage.

In a well-designed map, every part has to play well with everything else, at least within line of sight. Even small changes to the layout result in unbalanced maps that lead to frustrating play. Randomly generated maps might be fun out of novelty, but they won't play well in the long run, at least for games like Counterstrike.

Compared to Fortnite, where every match is a unique experience with the addition of player-crafted buildings, and the static game map is under constant transformation each season.

"Guided by human actions" isn't exactly random; chaotic and unpredictable, sure!

On big budget games like Call of Duty, some of the multiplayer maps would be very finely tuned over thousands of hours of testing. A friend and I play regularly (just us against bots) and not a session would go past without either of us remarking that they'd nailed the maps. There's nowhere you can truly dominate a map. Everything has been tweaked to balance it all.

Oh yeah. I remember modding Quake and Unreal and such back in the day--after a lot of practice and experimentation, one learns to read a map like a book. Gameplay is built right into the structure of the world. When the map is made by an expert, players can still enjoy it without directly understanding it or even noticing it. It's an incredible example of intuitive design.

Not to say there isn't a way to make some form of satisfying gameplay out of a randomized sandbox, but it certainly won't be the same as an FPS map designed with experienced intent. That aside, it would probably produce a spot with excellent gameplay every so often out of pure serendipity, which would be fun to hunt for! I'd try it.

> It’s also important that players/teams can memorise the level layouts in order to form strategies.

I'd like to see someone challenge that notion. I mean, that's certainly a known and reliable model for multiplayer play, but we see single-player games like Spelunky where familiarity means being able to recognize how common elements interact and construct your strategy on the fly, as opposed to a Super Mario where you're just memorizing the layout. Has anyone seriously attempted that in a multiplayer game? Obviously there's a lot of ways it could fail, but I feel like it could be spectacular if done well.

Look for blind races of Super Mario Maker at Games Done Quick.

You kow what would be more fun - is a game with a squad that needs to advance along a path to a goal co-op style taking out hoardes coming in the opposite direction - and the prize goes to longest life of the team.

I can't tell if you're trolling by commenting with a basic summary of Left4Dead versus mode, a 10-year-old game.

Left4Dead?

If the level was ever changing and expansive it wouldn't matter if one side has a major advantage. Balance is important for small amounts of limited maps, balance is not so important when you have infinite maps and someone might never encounter that same setup as before after hundreds of games. Plus it will be way harder to find advantageous points to utilize when you have to find that spot through observational skills, rather than seeing somebody else use it or kill you from it and copying them.

It is definitely a different play style this way, but im sure many people would have a preference for it.

> but im sure many people would have a preference for it.

hardly, for shooters it will become very frustrating very fast, because a lot of the time you're going to get outrandomed rather then outskilled.

I mean I don't get why everyone is focusing on shooter style games, this would be really great instead for assassination games or hide&seek games, including path findings, runners, treasure hunt etc.

I think Arma is proof against it not working. Sure the maps are static in Arma and not random, but they are also large enough that I might utilize a same spot twice over hundreds of hours of play, while in something like CS or CoD I use the same spots hundreds if not thousands of times. Plus with the larger maps without artificially small arena limits there is very rarely a 'safe' area to camp anyways. In CS or CoD you can block off two lanes and be immune from a surprise flank, but when somebody can literally go 1km circle around you and your team without you ever knowing about it there really isn't a safe area.

If you want a more direct comparison on how map limits effects playstyles, compare Insurgency with Arma, they both have similar weapon deadliness and player types where running and gunning isn't really possible, but the overall game strategy players use and rely on is very different. You regularly see players in Insurgency using the map limits to their advantage and can just assume they won't get shot from certain directions because of it, in Arma you might try and utilize a similar area but there is no guarantee someone might not take the time to completely flank you or pop out of the sea or just take pot shots from 2km out on a hill top.

You could design to have repetitive features, ex. you know that any tunnel will always have stairs to a higher level next to a lamp.

Reply on news.ycombinator.com