I am Peyton "monitor", and I make maps for the ESV Mapmaking team. I play Protoss primarily, but I fool around with Zerg and Terran often too. Right now I practice in custom games at a low-mid masters level. My map Korhal Compound is in the Blizzard Ladder. I am also the organizer of Map of the Month here on TeamLiquid.
A lot of people on Team Liquid make maps. Many of the maps are quite good, too. Unfortunately, most of them make the same error- incorrect proportions. You guys are getting really good!!
There will be two topics that I will cover as concisely as I can: Map Size and Proportions. Each of these need careful attention when creating a competitive melee map, because they can make or break a map. Once you get down the proportions, which comes with experience, you can make a well balanced map.
Map Size
When I say 'Map Size', I mean Playable Map Bounds. There are three measurements that the editor gives you- yellow, blue, and grey. The yellow dotten lines indicate that the players vision cuts off there (except player camera sees a bit farther). The blue line indicates where buildings cut off- you can't build anything past this. The grey layer, only visible by clicking Map --> Map Bounds, lets you adjust the Full map size. In the bottom right corner of this screen, it tells you the dimensions of Playable and Full map sizes. I refer to Playable map sizes when I talk in this thread.
When you start a map, you don't need to know exactly how it will fit- by the end though, it needs to fit within certain boundaries, or guidelines. As in Brood War, maps that are excessively large or too small have inherent imbalances (anything above or below 128x96 in a two player map was almost always considered flawed). Starcraft II maps aren't fully matured yet, but they do follow similar size rules.
When creating a map, keep in mind that a standard two-player map needs to be smaller than a standard four-player map, because they contain different spawn and symmetry properties. Part of the magic of a two player map is that it is smaller than a four player map. It doesn't need to be as big as a four player map because it will only have two mains, and less of the map is symmetrical (two sections instead of four sections).
A two player map is ideally around a 136x114 map size. This will allow space for 10-12 expansions and comfortable space to maneuver in. In most cases, a map larger than this can lead to having too long of a distance between expansions, and it will discourage aggression and/or harassment. And conversely, too small of a map will make aggression too powerful.
A four player map should be sized about 144x144. This will give enough room for 16 expansions. In my opinion, a 12 base four player map that uses rotational symmetry can almost never work. The problem is that Starcraft II is balanced off of a close third expansion, but rush distances need to be quite long. If you only have 12 bases, either the third base ends up being too far away or the rush distance is too short. I could be proved wrong though.
The rules are occasionally bent in competitive play, but without a plethora of knowledge of the game, larger or smaller sizes will not work. Every map will have some problem if it is too big or too small, whether it be distances, void space, open space, or proportions.
Proportions
Getting down your proportions takes knowledge and time. After you've had a lot of mapping experience, it will come more naturally- you'll be able to see how many gateways it takes to wall a choke, how many creep tumors it takes to connect the 3rd to 4th, how much vision the Xel'Naga Tower has, how many barracks fit in the main, etc.. You'll also get a feel for each of the race's preferred terrain as you play the game more.
Zerg players like to have wide open space. This is because their units have shorter range overall compared to Protoss and Terran, thus rely on getting surrounds on the opponent's army. Having too many chokes will make it impossible for Zerg to comfortable engage. Make sure you've got enough open space in the appropriate areas for the swarm to succeed.
Protoss and Terran both like tight spaces vs Zerg. In particular, Protoss likes corridors and Terran likes chokes. Both of their armies are primarily composed of long range units (stalkers, marauders, marines, tanks, colossi, etc.) which are better when balled up and only attacking at one angle. Protoss also has forcefields to decimate the battlefield in small corridors by cutting off the opponent. Make sure that your map has chokes dispersed appropriately so that Protoss and Terran can exploit them with good micro- but don't overdo it!
To have good proportions, ultimately you must balance each of these elements. Do your best to offer a balance of chokes, corridors, and open space for surrounds. Mastering space distribution can be very difficult, but will come in time.
Mains When you create the main base (a base with a spawn in it) the most important thing is getting the size and shape right. The optimum size of a main can differ with the size of the natural-- so if you have a lot of space in the natural, you want less space in the main and vice versa. A typical main is nicely sized with space for 30-35 CC's. Make sure enough of this space is located between CC and ramp, otherwise it will be very awkward for Terran and Protoss to rally units out of their base.
With 30-35 CC's of space, each race is comfortable. Protoss and Terran will have good room for production and tech buildings. Zerg won't have too much space to cover with creep and scout for proxies/drops. It will also lead to the ideal distance between the main and natural of about 1-2 creep tumors.
Naturals Most players don't realize, but naturals don't need to be big. If your main is correctly sized, then your natural can be quite small. In Brood War, building placement was extremely careful because the areas were very small. Starcraft 2 hasn't reach the same level because building positioning doesn't matter quite as much, but making the natural too big can result in bad space allocation.
Starcraft II naturals need three primary things. Firstly, they need space to wall off. This means however many barracks are required to complete the wall, and some room behind them to put bunkers and a medium size army (same applies to Protoss and Zerg). Secondly you need some space to build a few production buildings. The most comfortable spot for this space is in between the natural's CC and the wall. Thirdly, you need a little space behind the mineral line for static defense, and possible space for proxies.
Thirds Thirds can vary a lot once you get into them. At the most basic level, you need space to build enough structures to produce off of 3base constantly and enough room for an almost 200/200 army. If there isn't room in the third, there needs to be a logical place to build and put your army somewhere else. Players often like to park their armies between their natural and third to defend from harassment at all bases. If there isn't a good spot to do this, make sure there is enough room at the third for a small army.
Take a look at the following maps for reference on this. They all do a extraordinary job with proportions.
When making maps, I ask that you please pay attention to your proportions and space allocation. Use the above maps for reference, you can find them on most servers. Check them out in-game, and good luck mapping!
I'm slightly confused about this post, although do find it extremely helpful. My question is should the 128x98 be the PLAYABLE size or the MAP size? Thanks for the post!
good guide! I always thought there should be some kind of general understanding among the mappers that certain sizes are the most balanced. This is definitely something a lot of mappers need to consider but it's a shame that so many maps wont see the light of day because of how the popularity system works. Even some of these GSL maps are horrible when it comes to proportions but are in the lime light simply because they have a GSL tag, so I hope even the GSL mappers and the GSL admins come to an understanding about proportions and truly figure out the best maps for competition.
On February 09 2011 15:24 Final3 wrote: I'm slightly confused about this post, although do find it extremely helpful. My question is should the 128x98 be the PLAYABLE size or the MAP size? Thanks for the post!
On February 09 2011 16:03 butter wrote: How do you feel about Tal'darim Altar?
Tal'Darim Altar has a very well thought out concept-- one of the best yet. The only problem is its overall map size, which is too big. The following numbers explain this by using comparisons to Shakuras Plateau, a commonly accepted "good map" by Blizzard:
Tal'Darim Alter would be one of the best maps, but its size makes Zerg too powerful in ZvT and ZvP, and possibly Protoss too powerful in PvT. The extra seconds of rush distance gives Zerg too much time to rebuild their army between attacks, and allows them to have 5+ hatcheries with stacked larvae very early and easily.
I am currently in the process of contacting the GSL mappers about map size. I have already talked to winpark, creator of Auir Gardens, and he agrees it needs to be smaller, so he is remaking it.
TL;DR: Tal'Darim Alter would be near perfect if it was smaller.
Yes those sizes are very good. If you cut out the space for air units. If Blizzard made these exact maps they'd be at least 10x10 bigger. Generally I think it's dumb to say "a map should be this size." As example, you didn't consider the LT/Meta style maps that are diagonal, which requires a lot more space.
Also pretty much all the Blizzard maps that seem right in size (KR, XNC, BS, SP) are slightly larger than your numbers so that makes it sound a bit weird too.
On February 10 2011 03:34 FlopTurnReaver wrote: Yes those sizes are very good. If you cut out the space for air units. If Blizzard made these exact maps they'd be at least 10x10 bigger. Generally I think it's dumb to say "a map should be this size." As example, you didn't consider the LT/Meta style maps that are diagonal, which requires a lot more space.
Also pretty much all the Blizzard maps that seem right in size (KR, XNC, BS, SP) are slightly larger than your numbers so that makes it sound a bit weird too.
If Blizzard made their maps end at the edges of land, they would all be a lot smaller. Currently they have a lot of space around the outside of the map.
That is a good point, the sizes I give do not include space for air units around the outside.
LT and Meta are only slightly bigger than I listed, but they too have had balance issues regarding space-- LT is hard for Z to maneuver late game, Meta has easy crescent expansion layout for Z in cross positions. Square maps and Diagonal maps differ, but not by that much-- especially with a well thought out concept and use of space.
To me 128x96 is very small - one of my next map is 130x114 (157main2main , 120nat2nat < both slightly bigger than Xel'Naga Caverns) but i agree on your overall guidelines
On February 10 2011 04:14 dezi wrote: To me 128x96 is very small - one of my next map is 130x114 (157main2main , 120nat2nat < both slightly bigger than Xel'Naga Caverns) but i agree on your overall guidelines
I noted below that often 1v1 maps can go to 132x104, sorry its not in the chart
Making maps this big is rather risky, but can be balanced fairly easily. Does your big map have air space around the map? My guidelines were (mislabeled mind you) for to map space, not including void space around the map.
There is space for air and if can shrink this but i don't feel comfortable with it and also you stated 136x104 is fine so i'm not way out of the bounds.
On February 10 2011 04:56 dezi wrote: There is space for air and if can shrink this but i don't feel comfortable with it and also you stated 136x104 is fine so i'm not way out of the bounds.
Having air space is fine, that part is excluded from the sizes. I actually meant 132x106, but you're fine anyway- you know your maps :D
dezis maps feel a bit bigger than icup maps. I found the scaling in my last map with 12expansion in 128x128 quite good, 10expansions works too and give you more space for terrain. trying to make something special or many high levels forces to make biger maps, e.g. 10expansions on 128x128 (all 1on1)
On February 10 2011 05:01 Samro225am wrote: dezis maps feel a bit bigger than icup maps. I found the scaling in my last map with 12expansion in 128x128 quite good, 10expansions works too and give you more space for terrain. trying to make something special or many high levels forces to make biger maps, e.g. 10expansions on 128x128 (all 1on1)
Yah the idea with my post was directed at relating proportions to map size, which is rather impossible to explain lol. Take a look at some of the listed maps; they've been revised.
You should consider linking map analyzer images for the maps you recommend as having good proportions, because I know I find proportions specifically much easier to judge that way.
On February 10 2011 05:23 funcmode wrote: You should consider linking map analyzer images for the maps you recommend as having good proportions, because I know I find proportions specifically much easier to judge that way.
I've become very interested n melee map production recently and I find it easy to make them too big. I have now developed my own set of reasonable scales for the various parts of the map and they work very well. My maps tend to be large in size and space due to that fact that I worry more about imbalances hindering Zerg than favouring them.
When I read the proportions in the OP I was just about ready to un-publish all of my maps and remove my map thread but after having a look in the editor I am not so sure. Proportion is something that I believe has a level of intricacy that is difficult to sum up with simple rules. Ten expansions for 1v1 sounds good to me, maybe even twelve (it's always nice to have options as well as expansions.) But I created a map with a playable size of 132x104, made a mineral patch and pasted it until I had ten. This is what I found: + Show Spoiler +
Taking into account cliffs, way the edge of the map is laid out, dynamic terrain structures plus main bases (and not to mention the fact that the editor seems designed to make maps at a 45 degree angle within a perpendicular square [more on that here]) this seems a little tight to me.
Also when talking about imbalance in large maps we need to remember that in reference to Zerg having too much time to produce and the warp in ability giving too much of an advantage it is rush distance that is the issue. Rush distance is only tied to map size as long as the main bases are in the corner which is practical, instinctive and standard but not essential. Having a main nearer the center of a large map to decrease rush distance is effective and useful as long as the appropriate steps are taken to counter the inherent problems with such a map.
I am all for the idea of this post and was thinking of making a similar one before I saw it. I would like to make it clear that I am not arguing or trolling, just discussing a subject that greatly interests me. It's important that all the mappers out there (including myself) are getting it right and that we have lots of good custom maps to play. It would be great to also see information on pathway proportions, wall-in sizes and possible imbalances with high ground structure.
These guidelines are good for a starting point. They reflect the state of the art presently, and a bit of the iccup squad's doctrine. That is okay, since monitor shared his guidelines.
As you say, deviating from these dimensions can be balanced depending on the layout. Imagine if Xel'Naga Caverns was enlarged 15-20% and had one or two more expos per side. (The routes and whatnot would have to be reproportioned to accomodate this, it's just a general hypothetical.) This would be totally okay for balance. The nat to nat rush distance would resemble Shakuras. Because there are so many pathways through the center or near-center, all expos are reasonably accessible from any other location. This style lets you have bigger maps. Once you start making zigzags without intersections, big maps get unwieldy.
I don't think the current game knowledge is sufficient to say why big maps don't work. Have you seen the IM vs zenex game on Crevasse? Exactly the opposite of your ZvP imbalance rationale occurred. I'm not saying this invalidates it, but it shows that there are unexplored situations, necessarily: we haven't had extensive top level play on huge maps. One of the problems with Crevasse is the in base natural, which allowed Protoss to do that. I think Tal'Darim is a good candidate for pushing just beyond the size limit. We should watch that one closely.
Protoss and Zerg spawned cross positions, both FE'd obviously. Protoss can wall with one pylon, forge, gateway because of the rocks. He tech'd hard behind a single cannon while scouting and harassing with one void and one phoenix. Zerg got 4 or 5 bases and had tons of roach hydra and some air. Protoss pushed out with the money comp: void rays, colo, some sentries and various ground guys, taking his third. Zerg came in and was obliterated. Remaxed with more corrupters, got obliterated on the center platform. Rebuilt some guys and got obliterated as Protoss is walking into his outer territory, gg. Zerg was severely underprepared for a deathball that early, which was possible because of the laughably easy two bases.
Perhaps I did go a little over the top with my argument. I just think that something like this would be really a great tool for bettering the mapping community if it was more comprehensive. It needs to challenged, and expanded. (also I may have certain insecurities about the size of my maps :p)
I think we wrote our posts at the same time and my pronouns were ambiguous, saltydog. ;D I was speaking in the direction of OP, for general discussion, sorry for confusion.
This is potentially an enormous discussion / "guide".
On February 10 2011 08:05 EatThePath wrote: I think we wrote our posts at the same time and my pronouns were ambiguous, saltydog. ;D I was speaking in the direction of OP, for general discussion, sorry for confusion.
This is potentially an enormous discussion / "guide".
Yes, it is a gigantic topic. I'll add more and more when I have time....
For now, I've updated with OP with a slightly more detailed guide for proportions. Give it a read and let me know what you think!
On February 09 2011 16:03 butter wrote: How do you feel about Tal'darim Altar?
Tal'Darim Altar has a very well thought out concept-- one of the best yet. The only problem is its overall map size, which is too big. The following numbers explain this by using comparisons to Shakuras Plateau, a commonly accepted "good map" by Blizzard:
Tal'Darim Alter would be one of the best maps, but its size makes Zerg too powerful in ZvT and ZvP, and possibly Protoss too powerful in PvT. The extra seconds of rush distance gives Zerg too much time to rebuild their army between attacks, and allows them to have 5+ hatcheries with stacked larvae very early and easily.
I am currently in the process of contacting the GSL mappers about map size. I have already talked to winpark, creator of Auir Gardens, and he agrees it needs to be smaller, so he is remaking it.
TL;DR: Tal'Darim Alter would be near perfect if it was smaller.
Have your overall thoughts on Z being too powerful on large maps changed after watching the GSL and GSTL? It seems to me that when any of those games made it past the early mid-game (that is to say, into the mid-game, but early on in the mid-game) that Z was at a huge disadvantage in both T and P matchups.
I completely agree with your analysis of PvT based on the evidence.
Dont forget , theres a map size, and the mapsize you actually fight on.
For example, my new 3p map for MotM #4 is about 220 x 220, but the actual fighting size is a more of a 140x140 ish thing. Dont get confused between the two
On February 09 2011 16:03 butter wrote: How do you feel about Tal'darim Altar?
Tal'Darim Altar has a very well thought out concept-- one of the best yet. The only problem is its overall map size, which is too big. The following numbers explain this by using comparisons to Shakuras Plateau, a commonly accepted "good map" by Blizzard:
Tal'Darim Alter would be one of the best maps, but its size makes Zerg too powerful in ZvT and ZvP, and possibly Protoss too powerful in PvT. The extra seconds of rush distance gives Zerg too much time to rebuild their army between attacks, and allows them to have 5+ hatcheries with stacked larvae very early and easily.
I am currently in the process of contacting the GSL mappers about map size. I have already talked to winpark, creator of Auir Gardens, and he agrees it needs to be smaller, so he is remaking it.
TL;DR: Tal'Darim Alter would be near perfect if it was smaller.
Have your overall thoughts on Z being too powerful on large maps changed after watching the GSL and GSTL? It seems to me that when any of those games made it past the early mid-game (that is to say, into the mid-game, but early on in the mid-game) that Z was at a huge disadvantage in both T and P matchups.
I'm curious as well. I don't really watch GSL at the moment (I'm more into the European scene) but apparently Protoss and Terran are pretty strong on these maps with their lategame armies against Zerg, right?
On February 09 2011 16:03 butter wrote: How do you feel about Tal'darim Altar?
Tal'Darim Altar has a very well thought out concept-- one of the best yet. The only problem is its overall map size, which is too big. The following numbers explain this by using comparisons to Shakuras Plateau, a commonly accepted "good map" by Blizzard:
Tal'Darim Alter would be one of the best maps, but its size makes Zerg too powerful in ZvT and ZvP, and possibly Protoss too powerful in PvT. The extra seconds of rush distance gives Zerg too much time to rebuild their army between attacks, and allows them to have 5+ hatcheries with stacked larvae very early and easily.
I am currently in the process of contacting the GSL mappers about map size. I have already talked to winpark, creator of Auir Gardens, and he agrees it needs to be smaller, so he is remaking it.
TL;DR: Tal'Darim Alter would be near perfect if it was smaller.
Have your overall thoughts on Z being too powerful on large maps changed after watching the GSL and GSTL? It seems to me that when any of those games made it past the early mid-game (that is to say, into the mid-game, but early on in the mid-game) that Z was at a huge disadvantage in both T and P matchups.
I'm curious as well. I don't really watch GSL at the moment (I'm more into the European scene) but apparently Protoss and Terran are pretty strong on these maps with their lategame armies against Zerg, right?
Yeah after the GSL games, it seems to me that Protoss is really powerful in PvZ. I was completely wrong, I underestimated the power of the deathball and Protoss macro. It could still change in the future as metagame evolves, but I do agree Protoss is at an advantage on large maps in PvZ at the moment.
I'm not really sure ZvT, it seems like its relatively balanced to me on larger maps. From the high level games I've seen, whenever the Zerg loses, he fails to expand when he needs to, and also gets the wrong army composition. There was a game with StrifeCo(Z) vs. GoOdy(T) where StrifeCo lost, but lost because he overdid his aggression and failed to expand when he needed to (and saw his opponent expanding). This is just one example, and it very well might be that Zerg is at a disadvantage in ZvT too.
I updated the OP to make reading a bit easier... my writing is not very good, sorry.
I also removed Adagio from the 'good proportions', because I felt like Amazon is much better. Adagio was decent in proportions, but they layout, balance, and originality is not particularly good compared to Amazon.
On February 10 2011 06:08 EatThePath wrote: These guidelines are good for a starting point. They reflect the state of the art presently, and a bit of the iccup squad's doctrine. That is okay, since monitor shared his guidelines.
As you say, deviating from these dimensions can be balanced depending on the layout. Imagine if Xel'Naga Caverns was enlarged 15-20% and had one or two more expos per side. (The routes and whatnot would have to be reproportioned to accomodate this, it's just a general hypothetical.) This would be totally okay for balance. The nat to nat rush distance would resemble Shakuras. Because there are so many pathways through the center or near-center, all expos are reasonably accessible from any other location. This style lets you have bigger maps. Once you start making zigzags without intersections, big maps get unwieldy.
I don't think the current game knowledge is sufficient to say why big maps don't work. Have you seen the IM vs zenex game on Crevasse? Exactly the opposite of your ZvP imbalance rationale occurred. I'm not saying this invalidates it, but it shows that there are unexplored situations, necessarily: we haven't had extensive top level play on huge maps. One of the problems with Crevasse is the in base natural, which allowed Protoss to do that. I think Tal'Darim is a good candidate for pushing just beyond the size limit. We should watch that one closely.
Protoss and Zerg spawned cross positions, both FE'd obviously. Protoss can wall with one pylon, forge, gateway because of the rocks. He tech'd hard behind a single cannon while scouting and harassing with one void and one phoenix. Zerg got 4 or 5 bases and had tons of roach hydra and some air. Protoss pushed out with the money comp: void rays, colo, some sentries and various ground guys, taking his third. Zerg came in and was obliterated. Remaxed with more corrupters, got obliterated on the center platform. Rebuilt some guys and got obliterated as Protoss is walking into his outer territory, gg. Zerg was severely underprepared for a deathball that early, which was possible because of the laughably easy two bases.
I disagree on a few points here. I believe IdrA explained it best on State of the Game, when asked for his thoughts on balance and the MLG map pool. He said "Zerg has a big problem on the GSL maps because it relies on creep spread to control space and connect expansions in the midgame" (he then elaborated to talk about the role of mutas and whatnot, but you get the idea.)
ICCup operates under constraints to map sizes and distances that are very similar to what has been described in the OP because that is what we have found over the course of custom map's existence in SC2 in testing to be the most effective. Large maps like Enigma and God's Garden were removed because exactly these issues.
You can have a big macro map, but it doesn't have to be 178x178 if you use space efficientely and plan your design correctly. To be honest, I personally find giant maps to be sloppy execution of a potentially great concept.
On February 09 2011 16:03 butter wrote: How do you feel about Tal'darim Altar?
Tal'Darim Altar has a very well thought out concept-- one of the best yet. The only problem is its overall map size, which is too big. The following numbers explain this by using comparisons to Shakuras Plateau, a commonly accepted "good map" by Blizzard:
Tal'Darim Alter would be one of the best maps, but its size makes Zerg too powerful in ZvT and ZvP, and possibly Protoss too powerful in PvT. The extra seconds of rush distance gives Zerg too much time to rebuild their army between attacks, and allows them to have 5+ hatcheries with stacked larvae very early and easily.
I am currently in the process of contacting the GSL mappers about map size. I have already talked to winpark, creator of Auir Gardens, and he agrees it needs to be smaller, so he is remaking it.
TL;DR: Tal'Darim Alter would be near perfect if it was smaller.
Have your overall thoughts on Z being too powerful on large maps changed after watching the GSL and GSTL? It seems to me that when any of those games made it past the early mid-game (that is to say, into the mid-game, but early on in the mid-game) that Z was at a huge disadvantage in both T and P matchups.
I'm curious as well. I don't really watch GSL at the moment (I'm more into the European scene) but apparently Protoss and Terran are pretty strong on these maps with their lategame armies against Zerg, right?
Yeah after the GSL games, it seems to me that Protoss is really powerful in PvZ. I was completely wrong, I underestimated the power of the deathball and Protoss macro. It could still change in the future as metagame evolves, but I do agree Protoss is at an advantage on large maps in PvZ at the moment.
I'm not really sure ZvT, it seems like its relatively balanced to me on larger maps. From the high level games I've seen, whenever the Zerg loses, he fails to expand when he needs to, and also gets the wrong army composition. There was a game with StrifeCo(Z) vs. GoOdy(T) where StrifeCo lost, but lost because he overdid his aggression and failed to expand when he needed to (and saw his opponent expanding). This is just one example, and it very well might be that Zerg is at a disadvantage in ZvT too.
I think it's also fair to take into consideration that not only are the maps large, but players are essentially gifted 3 bases to start the game as any race, which could leave Z, which is believed to require at least a 1-base advantage to compete in a macro game, at a disadvantage due to the ease at which T and P can keep up in the base-count to the point that an extra base or two isn't as beneficial. (wow, that's a run-on sentence if I ever typed one)
On February 09 2011 16:03 butter wrote: How do you feel about Tal'darim Altar?
Tal'Darim Altar has a very well thought out concept-- one of the best yet. The only problem is its overall map size, which is too big. The following numbers explain this by using comparisons to Shakuras Plateau, a commonly accepted "good map" by Blizzard:
Tal'Darim Alter would be one of the best maps, but its size makes Zerg too powerful in ZvT and ZvP, and possibly Protoss too powerful in PvT. The extra seconds of rush distance gives Zerg too much time to rebuild their army between attacks, and allows them to have 5+ hatcheries with stacked larvae very early and easily.
I am currently in the process of contacting the GSL mappers about map size. I have already talked to winpark, creator of Auir Gardens, and he agrees it needs to be smaller, so he is remaking it.
TL;DR: Tal'Darim Alter would be near perfect if it was smaller.
Have your overall thoughts on Z being too powerful on large maps changed after watching the GSL and GSTL? It seems to me that when any of those games made it past the early mid-game (that is to say, into the mid-game, but early on in the mid-game) that Z was at a huge disadvantage in both T and P matchups.
I'm curious as well. I don't really watch GSL at the moment (I'm more into the European scene) but apparently Protoss and Terran are pretty strong on these maps with their lategame armies against Zerg, right?
Yeah after the GSL games, it seems to me that Protoss is really powerful in PvZ. I was completely wrong, I underestimated the power of the deathball and Protoss macro. It could still change in the future as metagame evolves, but I do agree Protoss is at an advantage on large maps in PvZ at the moment.
I'm not really sure ZvT, it seems like its relatively balanced to me on larger maps. From the high level games I've seen, whenever the Zerg loses, he fails to expand when he needs to, and also gets the wrong army composition. There was a game with StrifeCo(Z) vs. GoOdy(T) where StrifeCo lost, but lost because he overdid his aggression and failed to expand when he needed to (and saw his opponent expanding). This is just one example, and it very well might be that Zerg is at a disadvantage in ZvT too.
I think it's also fair to take into consideration that not only are the maps large, but players are essentially gifted 3 bases to start the game as any race, which could leave Z, which is believed to require at least a 1-base advantage to compete in a macro game, at a disadvantage due to the ease at which T and P can keep up in the base-count to the point that an extra base or two isn't as beneficial. (wow, that's a run-on sentence if I ever typed one)
In my opinion, this is the most problematic feature of the GSL maps presently. It seems to me, a "macro map" is supposed to provide expansions that can be defended against rushes or very early timings, given proper scouting and good defense. It's not supposed to give you 2.5 bases essentially for free, subject only to harass. On top of this, taking a 4th is vastly more difficult, encouraging 3 base turtle.
Anyway, if I were tasked with creating a set of maps for GSL or what-have-you, I would err on the side of conservativism and stick to sizes +/-10% of what has been the size limit so far, I would say 144x144 ish. You can make a map feel larger or smaller depending on the terrain and expansions quite easily. The GSL maps are just one style of large map. You simply can't say it's definitely imbalanced, or why, even though we all agree there are probably particular problems. You could only say so after some serious warfare with a significant population of top players over months to allow for metagame innovations. On the flip side, you certainly can't argue it is balanced.
In any case, I mostly support the current ethos, which is to improve the design of maps on a similar scale, not experiment with the scale. Tweak one knob at a time.
Regarding the camera bounds, mappers need to make a decision regarding the ability of 'fly bys' for air units.
If you take a missile turret and count 7 placement-grid squares away, that is the range it will hit units, so making a decision of how far or how close you want this is important to any matchup, as it determines how easily a muta flock or dropships can sneak in.
I like to use a sensor tower as the 'marker' for the range. you can then adjust the camera bounds to this on the screen (its easily visible).
On August 11 2011 11:19 a176 wrote: Regarding the camera bounds, mappers need to make a decision regarding the ability of 'fly bys' for air units.
If you take a missile turret and count 7 placement-grid squares away, that is the range it will hit units, so making a decision of how far or how close you want this is important to any matchup, as it determines how easily a muta flock or dropships can sneak in.
I like to use a sensor tower as the 'marker' for the range. you can then adjust the camera bounds to this on the screen (its easily visible).
Typically I find that wasted space is really unnecessary. I don't think you should be able to sit air anywhere for free, I'd rather require micro if the opponent is perusing you. You'll find that almost no BW map had wasted space around the edges (and almost none in general)- instead, they encouraged drops to require more micro, and defending to be more effective if you are a good player (because you could chase the opponent with good micro).
[edit] No wasted space also encourages the better player to win, because micro pays off.
A lot happened since this thread was first created and we have a lot better understanding of the proportions. A lot has been standardized by now so I am thinking that it is time for an update, maybe an even more detailed guide? Not to be harsh on the new guys(we have all been there) but with the large influx of new maps since the announcement of the TL map making contest I am sure we can help them and raise the standard a bit.
If you dont have time or dont feel like making it you could ask Barrin to do so. I am sure he is up to it
On October 12 2011 22:36 Barrin wrote: 30-45 in-game seconds is a good place to be at for natural choke <-> natural choke distance + Show Spoiler +
not 100% on rush distances tbh
Your numbers seem wrong.
I used your mesuring methods and I got these resoults: Bardiche with rocks still on 27 seconds Emerald jungle close spot 27 seconds Emerald jungle cross spot 30 second
And then the map everyone and their pet on this subforum complains is too big tal darim altar close position 33seconds tal darim altar cross position 43seconds
I'd be glad to expand this to include more relevant information for mapmakers. I know a lot of it isn't very useful when actually creating a map. More numbers would help.
I really hope the judges of the TL contest don't rely on the openness value provided by SC2 mapanalyzer. It doesn't take into consideration unpathable doodads but considers the pathing layer (the red map that tells if you can walk/drop/climb). The way mappers tweak the pathing can completely mess with the openness values sc2 mapanalyzer calculates.
For example on my map I have an open center with a few small unpathable doodads as obstacles. If I let sc2 mapanalyzer calculate the openness it gives 4.32 which is up the limits you gave for reference. But now if if use the layer brush to tell sc2map analyzer theses doodads are unpathable it calculates an openness of 3.66 which is barely in the lower limit (you said the limits were 3.58-4.22 and 3.8-4.0 is great). So I guess my map's openness should be somewhere in between.
Another example that messes with this figure is how mappers unpath cliffs. Mappers should use the pathing brush to make sure reapers can't climb up cliffs or get stuck up there, or that colossus can't get up either. But most mappers just put unpathable doodads. Yesterday Hero vs White-Ra in NASL s2 w5d1 on the map Crevasse, White-ra got a colossus stuck inside doodads in the middle of the battle and lost (that bug revealed the pathing layer wasn't painted on Crevasse). Sometimes Blizzard doesn't fix cliffs with the pathing layer either. If you don't paint the doodads out, sc2mapanalyzer thinks your unwalkable cliff is an open space. Some mappers paint the pathing layer extensively, some mappers don't. In the end sc2mapanalyzer will give a wide range of openness values.
I'm a bit angry now I discovered that and saw your openness reference values, because I sent my mapanalyzer picture for the TL contest with an openness of 4.32 which is way out of your references. I hope I don't get dismissed because sc2mapanalyzer doesn't properly calculates the openness.
On October 14 2011 01:37 chuky500 wrote: I really hope the judges of the TL contest don't rely on the openness value provided by SC2 mapanalyzer. It doesn't take into consideration unpathable doodads but considers the pathing layer (the red map that tells if you can walk/drop/climb). The way mappers tweak the pathing can completely mess with the openness values sc2 mapanalyzer calculates.
For example on my map I have an open center with a few small unpathable doodads as obstacles. If I let sc2 mapanalyzer calculate the openness it gives 4.32 which is up the limits you gave for reference. But now if if use the layer brush to tell sc2map analyzer theses doodads are unpathable it calculates an openness of 3.66 which is barely in the lower limit (you said the limits were 3.58-4.22 and 3.8-4.0 is great). So I guess my map's openness should be somewhere in between.
Another example that messes with this figure is how mappers unpath cliffs. Mappers should use the pathing brush to make sure reapers can't climb up cliffs or get stuck up there, or that colossus can't get up either. But most mappers just put unpathable doodads. Yesterday Hero vs White-Ra in NASL s2 w5d1 on the map Crevasse, White-ra got a colossus stuck inside doodads in the middle of the battle and lost (that bug revealed the pathing layer wasn't painted on Crevasse). Sometimes Blizzard doesn't fix cliffs with the pathing layer either. If you don't paint the doodads out, sc2mapanalyzer thinks your unwalkable cliff is an open space. Some mappers paint the pathing layer extensively, some mappers don't. In the end sc2mapanalyzer will give a wide range of openness values.
I'm a bit angry now I discovered that and saw your openness reference values, because I sent my mapanalyzer picture for the TL contest with an openness of 4.32 which is way out of your references. I hope I don't get dismissed because sc2mapanalyzer doesn't properly calculates the openness.
Not to worry, none of the judging is based on the openness of the analyzer.
I updated the OP a bit. Not that anybody needs to read it though, because almost all of the maps are following the guidelines.
Zerg players like to have wide open space. This is because their units are slower and have shorter range
You might want to correct this
For Chuky, if you still want to read this after all this time, just use the unpathable brush of the editor : the analyser will recognize it and you will avoid small chipsets of pathable space that stray blinkstalkers could get stuck in, or dropships could abuse.
I want to disagree about the main size. I think 25-30 is too small. For example from my own experience and from what I heard from several pros Entombed Valley has too small of a main to build stuff in, at it's 27 CCs. Custom tournament maps have 30+, for example Cloud Kingdom has 36, Tal'Darim has 39 (source). I think 30+ is way more reasonable.
On February 22 2012 22:41 Barrin wrote: btw I have indeed strayed to 30+
but I still think 25-30 is fine for small maps. You mentioned 3 of the heaviest macro maps (I personally consider entombed the heaviest macro map right now).
EPT![[image loading]](http://i.imgur.com/JBE6C.jpg)
![[image loading]](http://imgur.com/Td01Y.jpg)
