Hi Thad, im Tom. I used to do your job, at Raven, iD, and Valve.
Thad Jantzi, on 07 April 2015 - 08:22 AM, said:
“would it be that PGI's priority in map design is to funnel players to the action quickly?”
Speaking only for myself, I can say that my approach to map design is to A.) try to provide a diverse and compelling playspace where all different types of mech can play a role, and B.) try to make each map feel distinct from every other. I'm not trying to "funnel" players to the action quickly, it's my hope that the players' actions define where the fighting takes place and I consider it my job to provide a playspace where that can happen anywhere.
This is unfortunately, and respectfully, backwards. You dont want to provide a playspace where action can happen anywhere, you want to provide three specific things:
Durability. A map should withstand thousands of game sessions without letting players feel bored. It must provide continuous tactical challenge.
Accessibility. Navigation in a map should be clear. Remember that complex map design is one of the main difficulties a new player is confronted with.
Entertainment. This need is obvious, but its rules are difficult to define.
Action can happen at the same places, as long as it happens in a different and interesting way. If you look at the most successful multiplayer maps in gaming history, action still takes place at a few map designer influenced places, but are dynamic in each of these places. Theres multiple ways into and out of these action hot spots and engaging and disengaging from the fight is how it remains dynamic. Tourmaline does a good job of this with sparse crystal stacks being truly impassable terrain, while rolling hills provide 3-6 paths into and out of a 'bowl' or 'glenn' where action is likely to take place. The fight can always be there, thats not a problem, its that it provides enough paths, to be dynamic, dependent on player choice. You dont need to shuffle players to areas "quickly" but you do need to funnel them to stage. Without the stage, the actors have no place to act.
The number one easiest way, since you could put rooms above rooms (post Doom II era engines, when I worked at Raven, we didnt have to worry much about verticality) to achieve durability, is, like Obi Payppy Wan said, "Use the Vertical, Luke". From a 2002 World Craft article
"M
y first recommendation, and probably the most important, is to put the third dimension to good use. Use and exploit the vertical dimension in your maps and give the players reasons to use the volume of the map and not only its two-dimensional layout."
Which leads directly to the second thing. Which is interesting and tactically challenging connectivity. Interesting ways to get from one part of the map, to another. This is what alleviates choke points, which is what makes games today terrible compared to the yesteryear of complex map design. You never have just two ways into an area. And it needs to be obvious. The way you climb Death Mountain™ on one side, is completely counter intuitive until you see someone else do it, then struggle turning your mech back and forth to overcome the not obvious hill climbing ability thats distinct to non obvious or listed movement archetypes. Death Mountain™ pretty much breaks every mapping rule...ever. The written ones, literally, not just some unspoken rulebook.
You need to hold the players hand. You arent providing a sand box. Youre providing the stage. You control the lights, you control the sound, the backdrops, where the actors should stand, where the effect should take place, where the solo should be delivered, what color the stage amp should be. You are the director. Youre not just a cog in a machine providing a play area for people and their toys. You need to grab that bull by the horns.
As to distinctiveness. Art wise, the maps are distinct, but id say that the very reason the OP here exists, is because most of the maps, are not distinct, and play very similar to others.
By comparison, and maybe this isnt fair given that they are different games entirely, but the technological and experience gap between the two, may even this comparison out....Quake 1 Death Match maps, were incredibly distinct, and were, for all intents and purposes, perfect. Even though the entire color palette was muted browns and greens and earth tones, every map was radically different and played different from one another. One of the ways this was achieved was to never, ever, make a choke point. If all maps flowed, and each map was designed differently, it was a sure bet each map would play differently.
Now taking into account they could literally do whatever they wanted answering only to Carmack and Todd, who were rather liberal 'bosses', as well as art styles not being dictated by any sort of even sci fi reality, Sandy and team had an easier job of it. Especially without Romero whining all the time.
Now, this is just a design philosophy disagreement, so take it with a grain of salt, and the industry has changed, and with change, will come different needs, I dont pretend to know yours, but thats my take on it.
Our role is to provide a linear experience, that appears to be dynamic. What I feel, and I promise you I mean no insult, what I feel has happened with MWOs maps, is that they are dynamic maps that appear to be linear. Thats why you have the complaints that you have, about the lanes, the singular huge dominating features, etc. While the maps may be designed from a dynamic standpoint, they act, in a linear manner. You want to do the opposite. Provide a linear experience that appears to be dynamic. Like any good play.
Quote
“And that it's your opinion that wide-open maps requiring longer/varied travel times and actual scouting would be frustrating to too many players?”
I'm all for huge maps. It is, however, certainly true that maps requiring long travel times are frustrating to many players. You need only glance at the forums to see that. There is a consensus here that the sweet spot for map size is somewhere in the range of Caustic->Tourmaline and while I'd be perfectly happy to make maps bigger than that, or even bigger than Alpine, unless I'm instructed otherwise I think you'll see most maps henceforth be in the Caustic->Tourmaline range.
Id go out on a limb and say that Alpine is really the only problem and its because theres no chance that the fight will ever take place anywhere but the same two squares, which can be rather far from the slowest mechs spawn area alot of the time. Remove those complaints, then there are no longer frustrations for those "many players". The consensus is much more akin to "Death Mountain™" is terrible, not that the size of Alpine is terrible. I think many of us would definitely appreciate it if someone did instruct you to create something larger than Caustic and Tourmaline. Putting all the mechs into a single spawn area and not separated would increase the value of the "scouting" people want to do. As it is, even on the largest maps, the time to contact is still ridiculously low, and as soon as a full dorito comes up, you know what lance is where and what to do. Its not so much a tactical choice as a tactical certainty, which hurts map durability.
Quote
“Part of the problem is that a lot of the maps were designed for 8v8 and don't scale up well to 12v12.”
It's certainly true that a lot of the maps were designed before 12v12. What's even more important, though, for a number of maps is that they were designed before we even had a game. Forest Colony, Frozen City and River City were necessarily deep into development long before we could even play matches on them. At the time I was working on River City I thought it was enormous - turns out it wasn't. It's way too small - everyone agrees on that and the size of the maps that have come out since certainly reflects our awareness of that. We will ultimately be revisiting some of these maps with a view to improving the gameplay experience as much as we can.
Fantastic.
Quote
“Thad might not be the problem it might be the Cry engine design itself limiting what a designer can do on a scale that works for MWO.”
CryEngine is not a limitation for us in any way. Every map size is the result of us having decided to make the map at that particular size. This isn't to say that there aren't tradeoffs - the larger a map gets the more art will have to be made for it, the more resources it will demand from your computer, and so on. You could have an enormous empty map that performs well and takes a short time to make. You could have an enormous map full of art that performs like crap and takes a very long time to make. You could have an enormous map full of art that performs well and takes an exceptionally long time to make. Like everything in life we are bound by the strictures of reality, and that will always involve making tradeoffs.
A larger map, where youre concerned about total vertices, faces, whatever, showing, should mean less art. You could streamline that process alot by having large single feature rich maps, like a forest, or Tourmalines crystals, where you can reuse the same map models, with different textures, and modes (with leaves, without, alot easier than making new models) then plop them everywhere to break up lines of sight. Since these models wont be actually cutting line of sight for processing purposes, the map has to be extremely low detail, which should speed up the process. A large single feature rich map should take less time than a small detailed map. Something is going wrong here.
And back to the Quake example. No one cares if you reuse the same 4 textures and colors in all 12 of your DM maps, if each and every map plays entirely differently. Reuse reuse reuse. All those lovely doodads Viona makes, should really be in every map.
Quote
“But even in that case, there's still the "three lanes" rule and "one big central feature" rule that we see repeated in almost every non-CW map and that don't really allow scouting or emergent gameplay into the game. “
The thing to realize is that there are different people involved in this process. It's not like I'm Rembrandt at the easel making whatever I want. I work in a company and that means that numerous people have input into the process and that's how it should be. If I made the maps 100% myself without anyone else's oversight I can assure they would be much more divisive than they currently are. Some of you would probably love them. The majority of you would probably hate my guts.
You know as well as I do, if you were doing what moves you, designing the maps, doing the art, you want to do, that you feel in your soul, with no oversight whatsoever, people would love it, because you love it. And the haters be damned.
You need more freedom. I hope this isnt a gun shy/they took meh artistic freedom bro, thing over Alpine and youre holding on on Death Mountain over some sense of egalitarian righteousness. You do good work. They should let you expand your horizons a little bit. Experiment. Do a map you really want to do for a change.
Quote
I have certain preferences, I like the mechs to go wherever they want. My favourite map is Caustic Valley, runners up are Tourmaline and Alpine. Other people prefer more discrete paths and arenas, and we work with that in mind as well. Everything you see is the aggregate of the work of many people. In any case I don't agree with your assertion that there is a "three lanes" rule and "central feature" in every non CW map, and I think there's plenty of room for scouting or emergent gameplay in all the maps with the possible exception of the ones I've already agreed are far too small.
There pretty much is. Were not lying to you. I wouldnt lie to you. Death Mountain™ and The Citadel™ are things. Lanes are a thing. Wether its two, or three, or four lanes (Terra im looking at you) your objective of emergent gameplay isnt being met. People arent utilizing most of the map, and for obvious reasons, the edges of the map arent where the enemy are. Any browsing of a Call of Duty kill location heat map will tell you that. Unless you want to put teleporters at the edge of the maps, no ones going there, unless verticality comes into play (where say you could go to the edge to get to a rise or cat walk that took you back to the center). Again, you are the director (well if its a collection of chiefs, as stage designer, that might be a part of your problem. No one should be handing the lead map designer a sketch or a flow chart, of how a map should go, thats...bad sauce) you need to tell us, the actors, where to stand, so we can play our parts and get our rocks off.
Quote
“These core map designs are part of the DEVS SDK tools so Thad takes a already designed map scrapes off what he does not want and proceeds to build a new 3 lane map design.It is much faster than starting with a clean map and building everything from scratch.”
I have no idea what you're talking about.
I think he's saying youre copy pasting your WAD files bro (sorry, DoomED joke). I think he's saying you have a library of art assets, including a basic 'format' that you can paste into the editor, starting a new map, to get you started. Which is ... of course way more annoying than a clean start. You spend tons of time cleaning up things instead of getting work done.
Quote
“Doesn't explain why the constant emphasis on a central feature, though. Just an orientation device for similarly-minded "quick action" players? “
Look, I don't know what to say to this. There's a feature in the center of the map. What's your solution to this? All the features must now be at the sides of the map? Some maps have a big thing in the middle, some maps don't. Whatever.
Man you really are sticking to that Death Mountain™ gun aint cha.
Ok, off the top of your head, name one other famous multiplayer map from our constituents, and peers, that has a single dominating terrain feature you absolutely have to hold to win the map. And before you answer, think about how the single most powerful weapon or item, was never anywhere near that piece of terrain, without some ridiculously dangerous way to get there. Lava. Rocket jump. Zipline. Interesting tactical choices. Do you go for the dominate terrain, or do you go get the red armor first. Did someone already get the red armor and the rocket launcher? Man. Better avoid the central room.
Since we dont have pickups to draw you to different less used but more interesting parts of the map, there needs to be not one, not two, not three, not five, but multiple, great places to be. Some great places, the greatest places, should be difficult to get to, and countered by even more difficult places to get to.
Think HPG, but, 4 HPGs, instead of walls, and a larger HPG, over those 4 HPGs. The high ground is always good because of hardpoints. The longer it takes you to get to the highest ground, the better. Its a tactical choice. Do I take the pretty good, easier to get to high ground, or do I go the long way, and get to the highest ground. Or do I neutralize those spots by going underneath them all where I cant be hit. Youre weighing options.
Thats what leads to good map durability.
Quote
“I don't see why you can't provide varying terrain without limiting gameplay to one distinct feature (a la the hill in Alpine, or D*mbass Dome in Terra Therma). “
There is no map in this game that limits gameplay to one distinct feature.
It depends on how you define limits. Youre not LIMITED from going to the corner of the map, shutting down, and walking away from your computer. But thats a terrible idea if you want to shoot mechs. If your goal is to shoot mechs. you have one obvious choice, thats so obvious, its like a giant blinking neon sign that says FREE MONEY HERE, only, the money is worthless, and you have to do this, at least a few times a play session, day after day, ad infintium.
Quote
“The thing about the OP'S post is even if we get bigger or better map designs without new game modes and actual objectives other than destruction it wont make much difference in game play. “
Well I'm not the guy who makes up the game modes.
But youll probably be the guy who gets to work with map entities and set all their values and get yelled at when things dont work right because someone else vaguely notated their additions and is out sick the day you do it.
Quote
“Let me clue you in on something.. the lack of collision in this game in conjunction with whatever you call it, geometry ,clipping, whatnot, makes a map like bog an abomination.”
And believe me, this frustrates me more than it frustrates you.
Completely understand. Imagine what it was like for me, going from 2D sprites for 'map fluff' to actual 3D models. Then going from using them as fluff, to integral parts of the map geometry. And 2D sprites were no fun either. Its not like you just shot the little pots in Hexen and the spirte detected a hit. No, I had to individually set X,Y size so it would detect your hit. Far simpler than today, but time consuming and annoying, and of course since sprites rotated with you, you had to make sure the geometry wasnt cut into by the sprite at any conceivable angle. Placing 'fluff' took more time than actual geometry. Not sure if you do all of that your self. Ideally you should be.
Quote
“I would love to know WHY... they just dont level the mountain in alpine, and take advantage of the rest of the map? “
I don't level the mountain because I don't want to.
Ill go pick a fight with Gordie Howe or something. Please? Pretty please? Ill kneecheck Ovetchkin, ill make Crosby cry...just please god. Get rid of Death Mountain™
Quote
Okay BYE.
Talk to you again if the Canucks win a playoff round.
-T
Nice talking to you.
Stop picking on the Avalanche.
Edited by KraftySOT, 07 April 2015 - 09:06 PM.
3
