Can we Pleaassee have 1 v1 random map only queue?

Why is a poll more relevant than the actual results of trying it out and looking at the player numbers?

Yea I don’t see any harm in trying it out for a month or something and then having some real data to make a decision on rather than speculating how people might behave. Mapgen week isn’t really comparable because everyone was forced to play mapgen only and some folks didn’t like that.

A seperate 1v1 mapgen que will definitely split up the playerpool. As we have seen before, one que will take players from another que. 3v3 mapgen just about killed 4v4. I noticed people tend to not que in empty ques, even if it doesn't really require them to click them.

From a gamedesign perspective, it also doesn't really make any sense to make a separate que. And trying out mapgen only for a single month is not really a reliable measurement either, since the novelty effect can skew the results.

In general, I think the player base is quite evenly split, where about as much people prefer mapgen as pre-made maps. Considering all this, I think a healthy mix of mapgen and pre-made maps combined in one pool makes the most sense. That being said, I would be fine with throwing in a few 1v1 mapgen only months, just to appeal to those who prefer that.

most online games make stuff like this as "events", so people that want to play a certain mode etc. know a specific time, when everyone else will play that mode.
You could for example a once/twice per week time window where 1v1 ladder will be mapgen only
Or if it's easier make mapgen week once a month.

@blackyps said in Can we Pleaassee have 1 v1 random map only queue?:

Why is a poll more relevant than the actual results of trying it out and looking at the player numbers?

Earnest answer attempt:
Despite my liking mapgen maps enough to want a few more in the 1v1 que, adding mapgen maps, or an extra queue, wouldn't really make me play any more or less games per week. It would be more likely to effect how quickly I got tired of the game over the years, than how often I played per week. (tbh games per week depends far more on my jobs) while more data from a poll, if it's easy to make, sounds valuable.

(I know, players don't even always know what we really prefer! )

@stormlantern said in Can we Pleaassee have 1 v1 random map only queue?:

In general, I think the player base is quite evenly split, where about as much people prefer mapgen as pre-made maps. Considering all this, I think a healthy mix of mapgen and pre-made maps combined in one pool makes the most sense. That being said, I would be fine with throwing in a few 1v1 mapgen only months, just to appeal to those who prefer that.

I generally agree with this. The exception, I think, is 1v1 lower ratings. I never got ANY mapgen games when I was at low rating, despite really wanting to play them. (I just checked to make sure, it still seems to be the case with the latest pool. I assume it's a trend by design.)

I can totally understand the 'reason' a map comittee probably decides on no random maps for newcomers / low ratings - keep it simple, give them just a small number of maps to learn etc...
The trouble is, I reckon; a genuine newcomer, (and many low-rated players), isn't trying to learn maps at all! They're not even building on the hydro! They're trying to get good at 1 or maybe 2 build orders, and get tanks on the battlefield.
Any map-specific 'tricks' feel like a hurdle thrown in when you're still learning to crawl; from reclaiming forests on ambush pass, to building up cliffs or pgens for transports on Auburn canyon, and don't even get me STARTED on reclaiming the entire middle city on Loki!!!

I just have a feeling that the idea of low-rating 1v1 being only a small number of maps might not actually feel 'fun' for newcomers, nor be particularly suited to players that should just be learning 'standard' play. And honestly, I felt completely cheated the first time I realised Loki buildings didn't give reclaim mass... Or that engineers could build up certain cliffs in certain spots etc. At least when it happens on a mapgen map, you feel like you got outplayed, rather than feeling like you're losing to knowledge-checks!

I could be totally wrong, of course!
(I understand the flip-side is getting out of low rating without that knowledge and then having to learn it in harder games!)

Those map tricks aren’t map tricks. They’re essential components for reading maps, they’re just “map tricks” because it’s the first exposure to them and the map often centers on that element of the game.

Loki is a map trick, though. Only one that was similar to me was the EotS stripper pole that was hard to notice and gave huge mass but these days ctrl+shift exposes that. Should always be looking at reclaim at game start so it fell out of that category for me. Ah, there is the og forbidden pass where you could walk over a mountain but that was made more obvious in the faf version iirc.

@ftxcommando I tried ctrl+shift in loki! Kindof annoyed me that the buildings didn't show their mass value!
Regarding the other things though - my point is simply that at the lowest level, players are focusing on a basic build order, and while an enemy figuring out a clever cliff-build on a mapgen map feels like 'wow, I got outplayed', having them do it on a pre-generated map feels more like 'I lost because I don't know maps yet'... And then after learning that lesson you have to figure out how this affects the build order that you're already struggling to get right, next time you get that map!
While low-rating players playing mapgen maps would see it as an opportunity to practice that build order without any 'funny stuff' messing it up!

I'm certainly not against pre-generated maps in low-rating 1v1 though! They're great, and teach good lessons... I just think mapgen might not be excluded in principle; if anything it feels more 'fair' to newbies, and can still present lessons while feeling that way.

I agree with Sylph. If you want people to learn to read maps, instead of just memorizing certain maps, mapgen is the way to go. In general I feel mapgen is better and more interesting for new players because they aren't starting off even further behind by lacking the knowledge of the map. Everybody's seeing the map for the first time together, and from there it comes down to how well they read the map.

Wow so much good conversation.

I think making 1v1 only map gen was a bad move as a lot of people love the maps in rotation, and enjoy perfecting a map. Just look at sentons. A certain type of player loves memorizing and refining a build. A certain type of player hates memorizing and wants a purely level playing field and reading a map on the fly.

When 1v1 was set to mapgen only we were excluding a type of players preference so it is understandable that the game count when down. This also supports my theory that game count will go up if there is a mapgen only queue that complements the standard 1v1 queue.

I think 3v3 ate into 4v4 because it is fresh and new and if people queue both 3v3 will match first. Less chance of a single lagger messing up the game also.

Could we just try a 1v1 mapgen queue for a month and see what happens?

Facilitating players to more accurately express the game that they would like to play through queuing cant be a bad thing?

Yeah previously we didnt include mapgen in the lower brackets for two reasons. First because we thought the newer player would want some predictability in gameplay by playing maps they are familiar with. And second we noticed that a lot of small size (5x5 and some 7.5x7.5) are often pretty low quality mpagens with weird gameplay elements. But given the feedback in this thread we can start by trying out some 7.5x7.5 mapgen in the 200-700 bracket (if the team agrees), starting 1 november. Lets see what kind of maps are generated and how they will be received.

I think having a few random map gens in the lower bracket is a good idea.

But maybe I didn't make sense in my previous post, is any of my logic wrong or questionable? Otherwise could we please trial a 1v1 mapgen queue for a month that is pure mapgen regardless of bracket?

Lets see what people think and if the total amount of games goes up or not?

Because weirdly enough its been done and there is feedback:

https://forum.faforever.com/topic/3332/every-ranked-1v1-map-is-generated

https://forum.faforever.com/topic/2328/introducing-mapgen-week-on-ladder

https://forum.faforever.com/topic/2526/map-gen-week-feedback-thread

Not only does it disincentivize mapmakers from creating its not wanted by everyone. The goal needs to be a balance between curated maps and mapgen. Splitting the queue is unlikely to work as has already been pointed out.

The better solution to the different individual preferences is to implement map vetoes. Then everybody can tune the amount of mapgen to their liking (within limits).
The matchmaker team made a detailed concept, we just need someone to implement it. I can help with showing the ropes, so you know where to start. Overall it's not too complicated to implement. If anyone is interested in making it become reality, let me know.

@blackyps Because I do not think a week on/off is long enough for either to show a clear effect.

It doesn't have to be one or the other - could just have mapgen be more likely to occur in the existing 1v1 pool. Despite liking mapgen I wouldn't be keen on only mapgen 1v1 all the time.

I find mapgen only 1v1 very dull. The majority of maps generated are pretty trash B(

Mapgen is fun and diverse until you realise it generates only like 10 different types of maps with 1% variance in looks and gameplay within each type, and their quality peaks at like a 7/10

human-made maps dont have the 1% variance in looks and gameplay but atleast they exceed being 7/10 maps, and we have the ability to filter out the <5/10 trash, unlike with mapgen

I stopped stressing out because of random maps, and now we have something to do while the 1st factory is being built: studying the map.

The biggest issue with random maps is that sometimes you don't know if a unit can fit into this or that pass