-
We'll have to see how effective Jip's optimization efforts will be. Sounds like they've been making a pretty big difference on the fafdevelop branch
-
No amount of optimizations can combat a dual core laptop
. -
@fruitien00b said in TMM 3v3 and 4v4 coming soon:
I believe 4v4 will be actually played and not abandoned as 2v2. *3v3 would be completely dead. (Only my predictions)
I think I would queue 3v3 more often than 4v4 and I am #7 in total 2v2 games played on leaderboard with 195 games.
I imagine 3v3 will be less stressful than 2v2 as my mistakes are less likely to ruin another persons game. However, T1 phase will be more relevant in 3v3 than 4v4 and I like this game most at T1 and early T2.
-
@thewheelie said in TMM 3v3 and 4v4 coming soon:
Generally spawns that are far away from eachother, like open wonder, lead to way more boring gameplay due to the fact that it only leaves very little space on the map thats contestable. If you compare wonder with hilly the difference is very clear.
Right, I likely could use more examples to exemplify this, but point taken.
We really need more data on how people feel about spawn distributions to make any sort of move.
-
@Morax @BlackYps
Variety is the spice of life. Some people prefer open wonder-style scattered spawns, while some prefer hilly-style clustered spawns. A lot of people like both. Map gen presently provides variety. This is a good thing. Map gen code could theoretically be tweaked to change the spawn distribution to be more or less clustered if desired, but map gen is already more than desirable enough to be included in TMM in its current state.Some people here seem to think that 3v3 TMM would be unpopular and claim there is a lack of decent maps for it.
Some people here seem hesitant to put map gen maps into TMM.
I challenge those relevant people to try a 3v3 map gen only TMM queue alongside the other planned queues, and see how popular it is. It might be the most popular queue. I know I'd want to play it more than the other queues, and I know many others who would as well. -
@Emperor_Penguin given the choice between XvX random and XvX static I would pick the random indeed. Though ultimately I would probably always pick all of them to reduce queue time.
However, could we maybe include 2 different preset random maps in TMM pool (for 10km)?
-
Except people do know how popular a 3v3 queue would be since data exists for how many 3v3 games currently launch. It absolutely will not be the most popular queue.
-
3v3 could be popular, but it lacks map variety imo.
-
@snagglefox said in TMM 3v3 and 4v4 coming soon:
3v3 could be popular, but it lacks map variety imo.
There’s so little maps designed for 3v3 as it’s never been a focus in hosted games nor events. I’m unsure as to why, but to make it work well would require a substantial addition to your map vault.
-
Let's say we would have a 3v3 random map gen queue. What would the benefit be over a 4v4 random map gen queue?
-
@blackyps said in TMM 3v3 and 4v4 coming soon:
Let's say we would have a 3v3 random map gen queue. What would the benefit be over a 4v4 random map gen queue?
Benefit 1 - less likely to have a disruptive person in the game (you + 5 others means there's up to 5 possible disruptors, but in a 4v4, it's you + 7 others)
Benefit 2 - people with weak CPUs will be less disruptive (a 6-player game would tend to have fewer units moving around than an 8-player game)
Benefit 3 - people with weak internet will be less disruptive (someone might have 400 ping with 5 other people but 550 ping with 7 other people, in the first case you basically don't notice the lag, in the second case it's annoying)
Also:
Some people might prefer less crowded games for other reasons, just in terms of the balance between teching up vs. aggression. You can have 1 less eco/air player per team.
It's easier to find just 2 friends to queue with than it is to find 3 friends
-
@blackyps
The existence of a 3v3 map gen only queue does not preclude the existence of a 4v4 map gen only queue.3v3's and 4v4's have some similarities, but each has benefits that the other lacks. A 3v3 is not the same as a 4v4; if they were, we wouldn't be having this discussion...
Personally, I think that we should have map gen only TMM options for every player count that we have TMM for. Regardless, we definitely should have the ability to play 3v3 TMM, and we definitely should have the ability to play map gen only TMM.
If the problem with implementing a 3v3 TMM is a lack of good 3v3 maps, then making a map gen only 3v3 TMM is a sensible way to kill two birds with one stone.I would still hope that other TMM options/queues are added later, and adding a map gen only 3v3 TMM queue wouldn't prevent that.
-
The other question is what rating to use. Make a new 3v3 rating? Then we have like 5 different ratings instead of 4 (1v1, 2v2, 3v3, 4v4+, global). Or does it get folded in with 2v2? Or 4v4? Even 4 ratings is getting out of hand imo.
-
3 ratings is all you really need.
Either:
1v1/2v2+3v3/4v4+
1v1/2v2/3v3+ -
I don't think we can sustain too many different queues. I know that you like to have many options @Emperor_Penguin but more options is not always better.
Yes, you can multiqueue, but this doesn't magically increase the player base. Each launched game will remove players from the other queues. I fear that we will get "dead" queues that are not as popular and over time people don't even bother to queue up for them.
Now you have to explain to new players that queuing up for some queues is a noob trap. Speaking of new players, I would prefer if could understand the matchmaker tab without having to read up 10 minutes of explanations what the different queues do. -
Too many queues is only an issue if people are queuing as a team
By yourself, you can queue for all of them at once. Which makes it less likely there will be a "dead" queue.
Also if 2-player teams are allowed to queue for 2v2 3v3 and 4v4 at the same time, that would make a "dead" queue less likely
-
But you can only launch into one of them. That is my entire point
-
@blackyps
If we end up with some less used queues/options, that's fine. It doesn't need to be a noob trap. We already have the functionality to show how many people are in each queue. It's not hard to understand.Even if we add some more queues/options, it's not that complicated for a new player to understand, and having more good queues/options available would result in more people playing TMM more often.
For example, currently, we only have 1 option for 1v1 ladder. As map gen week helped demonstrate, there is a substantial number of players who would play ladder more (or at all) if map gen only ladder is an option. So, if we added an additional 1v1 ladder option that only plays on map gen maps, then some people who don't currently play ladder would start to play that ladder. Meanwhile, the players who currently play traditional 1v1 ladder could continue to do so. While many people would queue for both, some players would only want to play 1 of those types and would not queue at all if it was not an option. Thus, having both options available would increase the number of players who queue in total and would also increase the total amount of ladder games played. Without a 1v1 map gen only queue option, we're missing out on the entire green crescent in this venn diagram:
PS: I used 1v1 ladder as an example because we have recent info from map gen week on 1v1 ladder, but this concept also applies to having more TMM options.
-
ur missing the pure open palms 1v1 players in ur diagram
Also what is this extrapolation about map gen week, has anyone even looked at the unique users over said week compared to the week prior or after?
-
If you make too many queues and they end up dead then you can easily just remove them. No need to try and predict how many players will be in each queue and if it will be a problem.
