About Neroxis map generator...

Just some dev updates. We (as in mostly @Sheikah) have worked out the basics of how we need to change the database schema and have some code in the works.

Remember that TMM was created primarily in the spirit of facilitating casual play and enticing new players to stick around, so it is my hope that once we have the code changes in place, we can throw up a casual/unrated/global rating queue that incorporates some random maps. Regardless of where the discussion on competitive play goes, I think this is a valuable addition to the matchmaker.

i think it is good,gen map should be appended the ladder.It can avoid the player cause the gap to map familiarity,we need to make myself bo.

I think it will be more test a player and the basis of the random strain capacity

@Jip said in About Neroxis map generator...:

It shows quite a lack of respect to start such a discussion.

I am completely willing to have a cordial debate with anyone and start off giving them respect. But that does not mean they will always deserve to retain that respect, completely irrespective of their repeated conduct and demeanor.

For an example, try reading through the posts ftx made on the "increase t3 mex cost" balance post. (Even just the ones that haven't been deleted.)

After a certain number of attempts at a civil discussion, you realize it is pointless to keep trying.

@Askaholic said in About Neroxis map generator...:

we can throw up a casual/unrated/global rating queue that incorporates some random maps

Why an entirely new ladder and pool? Genuine question, don't wanna sound unappreciative. It's really great to hear this is being worked on and not just shutting down a good idea just because it's work for you btw, thank you and thanks @Sheikah so much for doing this - this idea about creating a whole new queue seems unnecessary on first glance to me.

I guess to try to ease tensions? I also think it being directly in the current ladder pool is fine.

Tbh I don't think I've seen the 1800+ crowd so united behind a ladder idea before, if the 1800+ pool was partly or wholly mapgen for a month I think it would go down very well.

Guess it's not a big deal either way, just might see all the tryhards in the "casual" queue which will be a laugh but people get what they want, that's what matters.

edit - guess the risk is splitting 1v1 player base, especially if the "casual" queue isn't pure mapgen. If it's pure mapgen then the hosting of custom mapgen 1v1s will just move to an automatch feature, which would be an improvement, but if it's not then it might get a bit weird in a three-way split between the two queues and custom games.

I agree with Sir-Prize that it might not be best to separate the ladder, if there are so few player to begin with.

My idea for a solution would be to make it opt-in. E.g. in the matchmaking tab there is Box A you can check, where you would like a 10% probability (or whatever) to play on a mapgen map. Box B would be if you want to always have a mapgen map and none of the ladder pool maps. So two players that select B can always play mapgen on ladder if they are paired, and any player that never wants it will never get it.
Just a suggestion, no idea how difficult this would be to implement.

@Sir-Prize said in About Neroxis map generator...:

Tbh I don't think I've seen the 1800+ crowd so united behind a ladder idea before, if the 1800+ pool was partly or wholly mapgen for a month I think it would go down very well.

take into consideration that usually only the people say something that want change. To get the opinion of the others we would need to introduce it and get feedback then.

people are just used to the same pool which grants most quite some skill boost during games.
not to mention that the bowhoring is one of the main things in big tourneys like LoTs,i wouldn't enforce that change but it's just bad when you look at high ranked players doing the same stuff for about 15 minutes when they have to play seriously and we call that "skilled" gameplay

queuing with a newbie to show him the beauty of tmm and meeting tagada be like:
https://www.youtube.com/watch?v=yLcRpdZ0Xb0&ab_channel=Tomoko

I think one of the problems that you aren't just willing to look at is the presentation to new players.
This is something that Biass and FtX have to watch out for.
Sure the map gen is great addition for us seasoned players who are all sick of BO whoring and want less competitive enviroment where burden of map knowledge is kept to minimum. We can also live with smaller and bigger mistakes like terrain not being obvious or something, as we are getting the kick out of finding new random map to play on. We are fine with ugly decals as long as we don't get to play the same game of burden of knowledge for the first 10 minutes of the game as to how to expand on certain maps or how to not lose the game at minute 11 by having worse Air BO.

But new players won't be as amazed, they will be pissed for having to fight against game again and again where nothing makes sense to them. Them getting a bad map will be much more impactful compared to us. Them seeing ugly map compared to what they can see on current ladder will also look strange. Why are they playing on ugly maps with pathfinding problems when there are hundreds of great hand crafted maps available to them.
Not to mention the fact that with random maps they will never get to feel at home seeing the same map for 3-4 time and now kinda knowing what to play on that map. Instead they will be faced with the daunting experience of having to play against not only enemy but also against brand new map with each new game.

Not to mention the fact that the newbies will have even harder time getting accustomed to setting economy up and running when they have to learn how to improvise BO each and every time they enter the match. What you all want to do is all nice and dandy if you are seasoned player who already have played the game for a while and is accustomed to it's problems.
But a new player? You are literally inviting living hell upon them and making sure that the ladder experience will become even more daunting to many of them as they get scared of yet another layer of unknown.

@RandomWheelchair
I agree with your points but dont recall anyone proposing to put mapgen into the <300 or 300-800 ladder pool. If blodir or archsimkat talk about how they would love like 10% mapgen in ladder, they clearly talk about their corresponding ladder map pool and not about forcing mapgen onto the low rated ladder brackets. So I dont see a conflict in this regard.

Yah, we already have map pool which changes based on your ladder ranking. The higher you go in ladder, the larger/harder the maps get, and map pool size increases. I would assume you could have map gen be introduced into pool the same way. Have it unlock at 'X' rating, to be determined.

I like the idea of map gen in ladder because it does not allow people who have PhDs in build orders to reign supreme.

I like having actual battles in my games rather thinking about how to survive first bomber and the game going to shit in less then 5 minutes.

@Sir-Prize said in About Neroxis map generator...:

Why an entirely new ladder and pool?

TMM is still lacking the casual element which was a big part of the design. The 2v2 queue we added is still pretty laddery because it has its own rating. I want to see at least one queue that uses global rating so new players can use it to "get out of the grey." Something that has the casual feel of “all welcome” games. It seems that there wouldn’t be as much resistance to adding mapgen to that sort of a queue, so it just feels like a nice way to test out the idea.

Fair enough, a global queue does make some sense when you put it like that. I still slightly worry about splitting player bases and making it harder to fill games overall but in fairness the moment you're not talking about 1v1s I probably should butt out of the conversation due to a total lack of knowledge.

On a (possibly) related note, what's the latest with the ladder divisions/hiding matchmaker ratings?

@harzer99 said in About Neroxis map generator...:

take into consideration that usually only the people say something that want change. To get the opinion of the others we would need to introduce it and get feedback then.

This is exactly what I'm suggesting. I will say that when map pool sizes have changed in the past there's always been some debate among higher level players, although they overwhelmingly preferred larger pools (keyser basically stopped laddering when the pool got big again). The only argument I'm seeing against mapgen is from people who are worried for newbs (rightly, imo) - but no one is suggesting this for newb map pools.

Small map pool = everyone has a build order for all maps => bo no longer has impact on the result of the game

Large map pool = some players have build orders on some maps => better prepped player wins

Mapgen = nobody has a buildorder on any maps => better bo improv wins

@Blodir said in About Neroxis map generator...:

Small map pool = everyone has a build order for all maps => bo no longer has impact on the result of the game
Large map pool = some players have build orders on some maps => better prepped player wins

This was/is why I prefer a small pool, but I'm in a minority and it's obvious small pool kills your level ladder play 💔

let's not forget that 1v1 is already considered a competitive gameplay

queuing with a newbie to show him the beauty of tmm and meeting tagada be like:
https://www.youtube.com/watch?v=yLcRpdZ0Xb0&ab_channel=Tomoko

I leave the decision for map gen for ladder up to arch. Of course it has development requirements etc but if he wants it for ladder I’ll trust his opinion.

Ultimately I’d rather run a dual queue with map gen and curated ladder maps both operating with the same rating system.

With regards to tmm, I see no need for implementing map gen in those formats. Most reasonings here hold no real sway because:

  1. Tons of new maps that saw near zero competitive play exist for these formats
  2. The format has hardly exists long enough for map fatigue to set in
  3. Map gen itself is a lot less tested for these formats