What is needed in a good map (A developer's perspective)

[Guest]

Welcome, Guest. Please login or register.
Did you miss your activation email?

1 Hour 1 Day 1 Week 1 Month Forever

Login with username, password and session length

[Ucross]

A few things maps should be IMO:

1) Reasonably small.  Better to side on making a map smaller than bigger as bigger maps mean more capping and less fighting.  Hill 331 is too large because the front is across a square turned like a diamond.  This makes the front too large.  However, if it was a 4v4 it would be much more appropriate.   It is often better if you are making a 2v2 to make it a 3v3, or if you are making a 3v3 to make it a 4v4.  This is because you can still play a 2v2 on a 3v3 map, and most mappers make maps too large.  Infact, it might be reasonable to make all maps 4v4s and let the public choose to play a 2v2 on it.  So point is make the maps smaller, but keep the number of players who can play on it to large.  A 2v2 should be medium, a 3v3 should be large, and a 4v4 should be 'just' extra large IMO.

2) Shot blocking walls.  Open maps make combat a lot less interesting and add a lot less strategy.

3) Multiple pathways to attack/defend that cannot be accessed easily between each other.  Rectangular maps offer limited pathways, smaller fronts, and easy access to all pathways for the defenders.  This makes gameplay more predictable, is good for not having to cap too much, and defense very strong.  Square maps offer more pathways, huge fronts, and long distances between pathways making it hard for the defense to cover them all in response to attacks.  This makes gameplay more unpredictable, players become unable to defend all aspect of the front at once, and defense and attack more even.  I believe the best is somewhere in the middle with a wide rectangle.

4) Sector sizes.  Every point in a sector should be reasonably close to every other point so that units cannot hide in a sector or cause the opposition to 'search' for periods of time for those units.  Sectors should not be too small though either.  An ideal number of sectors for a map is 4-8 across depending upon shape and size.  A very important point is that if players are testing your map and find that there is too much capping and not enough fighting you need to INCREASE your sector size.

5) Sector layout.  Sectors should not sneak across shot blocking and pathway blocking areas such as hedges so that it is possible to capture those sectors from the other side of a hedge.  If they do go over a hedge or something they should include the entire next sector.  Sectors should also be 50/50 in the middle of the map so that one side does not have more than them.  Sectors should also be ideal shapes so that when you capture a sector you don't have to walk across the entire map to get to the edge of the sector to clear out the defense.

[Tkaudi]

Sectors that do not cutt off half the map.

There is one of those St. maps with the city, where the road comes out and goes into the city, that road is a sector by itself.  It has essentially allowed the defender to sit inside the center of the city and cutt off that half of the map from being captured unless the city itself falls.  Why have the flank if it can't be captured?


Balancing the amount of sectors per side of the map.  again the same map, st laurant or something, had many many sectors clustered in the city, then those before or after and to its sides.  Meaning that the defender, abusing the road sector, could easily hold 3 or 4 closely packed sectors and stall and bleed their enemy out.

Maps like leprovinence and lambert and even abbyville don't have these issues because they offer enough avenues without any 1 sector stopping a flanking move while being controlable from a realistically unreachable position.

[Ucross]

Totally agree with Tkaudi, excellent post man.

[Tkaudi]

I personally think that the best size of a sector is one that an infantry can cover with LOS by moving less then half of the sector either way.

ie standing in the middle he can not see the full thing, but moving a tad in any direction he can see it all (laprovidence is the best example)... but again lambert has great flanking sectors that are bigger on either side of the city and more condensed sectors IN the city itself but because the city flows through the whole center of the map, its not like st laurant where its clustered on one end rather the actual north/south/east/west center... so it works

[Ucross]

While I wouldn't say that all sectors should be like that it is a great ideal IMO.

I would also say that sectors should not go across large walls like hedges, or if they do, they should go quite far across.  Some maps have the sectors just across a hedge so it is possible to hold the sector on the very edge of it on the other side of the hedge.  This is bad.

[salan]

While I wouldn't say that all sectors should be like that it is a great ideal IMO.

well its funny in one word i say it would be ideal (of course there is leeway in interpitation) but i did go out of my way to prove it also wrong by citing examples of lambert .. lol

[Nevyen]

OK so can we look to idetify maps that need to be changed to reflect these requirements and set out what needs to be done?

I have a few opinions but before i lauch into a crtique id like to hear what other think. 

[DerangedFerret]

Maps should have a good level of detail in order to make them visually stimulating.

Bland maps suck to play on. It's a fact. What you need to do, is to look at your map, and try to imagine how it would look if it were real. Then try to get it as close to that image in your head as possible. Or, you could imagine your map as a scene for a war movie, a la Saving Private Ryan. Then, try and work it out so that a movie could realistically be depicted there.

[Ucross]

Sometimes though detailed maps lag like shit.  Especially when they are large or extra large.

[Kolath]

But having smaller maps should help with the lag of detailed maps.  And if you look at relic maps, they are quite detailed and most do not lag significantly.  It is all about optimizing the number of objects vs splats to create the ambiance.

This is a very good thread.  What I think would be useful is for people to start posting hypothetical sector layouts to illustrate their points.  Do some for rectangular and for square maps.  For reference Abbeville is 544 x 384 meters playable area so to look at a generic map of that size you'd want a rectangle with a 24:17 aspect ratio.  A map like St. Hilliare is 512x256 meters for a 2:1 aspect ratio.

Using the right scale sizes on the sample maps will help get the sectors the right approximate size.

[Lolto]

Personally I prefer big but narrow maps, which is what I am aiming to create mainly as a mapper.  The smaller width of the maps allows for more intense combat and relies less on MCP but more towards what the players are capable of, this is why I prefer such maps as St. Hillaire or Rails, etc.

Optimization of detail is another strong point in smaller maps like stated before, I don't have the most ultimate computer of all but I know that many look at all the details so I strive as I hope many other mappers do, to make a map look the most natural flowing detailed maps they can become.

[salan]

sometimes I think they had it right with rails and metal, having that second spawn closer to the middle then the back end.  THe time to conflict is greatly reduced and can actually impact the game a hell of a lot more then expected.  Most of our maps have spawns on the back ends, we should test out spawn locations a bit more on certain maps which have difficult layouts?

[scrapking2]

Map shape is quite important to the relative effectiveness of arty.  Square maps favor on map arty, less distance from end to end.  For that reason alone, I think rectangles are better, although the exact dimensions are open for discussion.

Not that I am a mapper, but I think one of the most important things about a map is "implied symmetry".  That is, you want both sides (from which players spawn) to be the same, without looking the same (or the map will look gamey).  So, have about the same amount and distanced,  good cover, shot blocking terrain, access points, sectors, etc.  All of our current good maps have it.  If there is a no brainer side from which you'd prefer to start from, there is something wrong.

Salan hit a big one on the head - road sectors.  Nifty, but don't make the sector longer than adjacent sectors.  Feel free to chop up the road sector into several smaller ones.  This is the no "penetrating tentacle" rule.  No sector should be cap-able from a position of complete safety and once capped, allow sectors deep into your opponents control to be capped.  The road sector on Rails and Metal is a prime example of this.

I also like Salan's idea of what a good size estimate is.  "Infantry cover with LOS by moving less than half the sector" is a good rule of thumb.  What we don't want in games, and what turns alot of players off to MCP, is games of hide and seek.

Ucross brought up a good point about sectors flowing through obstructions.  Shot blockers, LOS blockers, barriers, etc, either need to be ENTIRELY contained within the sector - with free access around it within the sector, or define the sector boundaries.  What you don't want is a squad on one side of a barrier, safe and sound, capable of stopping a cap for that whole sector.   

Number of sectors is especially important.  Distribution equally so.  You do want half of the sectors on each half of the map.  You do want the same ratio of sectors on each spawn area, but not too heavy either.  IMO, the middle third of the map should be weighted with around half to two-thirds of all sectors.  Not exactly, but thats the general idea.  Implied symmetry applies here also.  You don't want one side having easier access to, or more sectors.

I'm going to disagree with Salan on the idea of moving spawns around too much.  Put them on the sides, or too far spread apart, and it not only becomes more likely to isolate and contain 1 player, but it will fuck with spawn arty quite a bit, making too much of the mid-map undeployable at start.  It can work, but be conscious of these points.

Lastly, I'd really like to see FL-'s idea of banded supply line sectors, possibly with my idea of objective sectors contained within the middle ones, implented.  I really think it can work well.