I agree with this point and I think that telling players to go play in the lobby is not a real solution.
However, if the DE lobby had feature parity with voobly’s lobby system, would it then be fair to tell players who want to play with very specific settings to go use that instead of the current implementation of the ranked queue? why or why not? Should the devs bother to fix the lobby to make it more appealing to these players? What effect would this have on ranked play and the ranked ladder in general?
That would be fair, I still think an opt-in system would be objectively better for everyone if done appropriately, but voobly lobby system would be such a massive improvement compared to the current situation (and probably not realistic) that there’s probably no need to speculate more.
There are 2 major unknowns: the ability of devs to make a good lobby system, and the fraction of people who are satisfied with the current system. Without knowing more it’s hard to speculate. Both results are possible, it can be better or worse than what we have now. If it’s done well (opt-in system) I see very little downsides. One of the downside would be that it favors variety less than the current system, but that could only happen if players massively ban diverse maps, showing they were opposed against diversity anyway.
The lobby is in a terrible state and should definitely be fixed, whether it’s worth the very low dev time that we have, that I cannot say, I don’t use it often enough
My opinion is that an opt-in system requires a far larger playerbase than we current have.
Let’s take CS:GO as an example, which is a game that uses an opt-in system in exactly the way that is described in this thread, as you may know, de_dust2 is by far the most popular map on CS:GO, but like DE, they have other maps that are often played in many competitions, but many players don’t necessarily bother with them.
My experience with CS:GO has been the following:
Queuing up for de_dust2 will find you a 5v5 match in anywhere from 20 seconds to a minute, incredibly fast.
Queuing up for other maps that are not de_dust2 can take anywhere from 8 to 10 minutes.
This is a massive disparity, consider that as of this writing, there are over 700k concurrent players in CS:GO:
I suppose it’s worth mentioning that CS:GO is a 5v5 game in ranked and not all of the concurrent players are playing ranked, I’m not sure if there’s a way to find those numbers.
I also concede that my experience with CS:GO was years ago, and their wait times may have gotten better over time, my understanding is that the high level CS:GO players tend to like map variety a lot more than AoE2’s top players, so perhaps that alongside the fact that their community is literally 50x larger than ours, their opt-in system is more sustainable, but I don’t think it will do us much good here.
Didn’t even know CS:GO had map selected queues. Didn’t need to know that to know it was a terrible idea. But it’s good to know the idea existed and failed elsewhere so as to save ourselves from making the same horrible mistake if we choose to be reasonable.
People will play the popular settings, not always because they want to, but because it’s perceived that other players will want to, since it’s the popular setting. It’s a self-perpetuating growth spiral that saps players from the other options by virtue of existing and the only way to stop it is to not let it happen in the first place.
If CS:GO got rid of selected map queues for standard play, it’d be riots. Once you give players an option, even if it’s a bad option for the game, taking it back is suicide. I urge DE not to make that mistake.
I wouldn’t say that CS:GO failed with opt-in queues necessarily, in the end, when matches can take upwards of 40 minutes to an hour, 8-10 minutes of waiting really is not that bad, CS:GO has the critical mass of players to pull that off. I was merely using my experience as an example of how extreme the disparity can get between popular and unpopular queues.
In contrast, wait times of 20-30 minutes is what one might expect from a less populated game, when queues start to take up as long as the average match, people will stop bothering with it and just go to the popular queue to find a match in seconds instead, hence the downward spiral you mentioned in one of your earlier posts.
If I had to wait that long to play a game, I’d have to open up another game to play while waiting for my game. Would probably make food or something too. That kind of long queue leads to a different problem - the AFK queue, where a player was so prepared for a long queue that they overestimate how long they have in wait time and do something else, and cost their compatriots either a requeue (AFK detection and barriers in queue) or a disadvantage (playing a man down).
These are all problems any player is familiar with and they become more frequent as the queue time lengthens with the system I’ve been advocating against.
I don’t play CS:GO, but if you queue only for 1 specific map, and that map is unpopular;, that will take long no matter the matchmaking system. That’s no different in DE. The playerbase size is not as important as it sounds, the system still needs to find 9 players in your geographical region and Elo range that only want to play your map
Note, that queuing for only one map (which is what you’re doing in your example) , is a very different situation than queuing for any map except Arabia/Dust2 (which is probably what most people would do in this thread when they think of diversity).
If you queue for many other maps that are not Dust2 you should get matched quickly. To take an analogy with DE, your example would be someone who wants to play only Megarandom in 4v4. It’s expected that this would take a lot of time even with the current system. In fact, it’s not even possible with the current DE system, so the CS:GO system is objectively better because at least it allows the constraint to be expressed.
Besides, I know this will sound generic, but you can’t compare a matchmaking that was tuned for a completely different game. True, CS:GO has a much larger playerbase, but it’s also likely that CS:GO matchmaker is much more selective on Elo differences, ping between players, and has more maps and game modes than DE. With more constraints it’s possible that CS:GO matchmaker ends up with a similar pool size as DE matchmaker.
Incorrect, when I used to play CS:GO, my friends and I started off by queuing up for all maps because we wanted random map selection. The problem we kept running into is that if we didn’t untick dust2 from the map pool, it was the only map we were ever matched on.
Queueing for everything but dust2 wasn’t as bad as queuing up for just a single unpopular map, we never did that because it took too long.
After a while we began to lose a lot more because matches were often unbalanced. We had more even matches on dust2 by virtue of there being a larger pool of players available to match with, so we switched over to that queue instead of trying to bother with other maps.
You can’t just say a game has failed by implementing an option and at the same time admit it would be riot if it was removed. The most likely explanation is that the option works perfectly fine there, 10 min queue time for an unpopular map is fair and even the one who posted the message said it is.
Sounds logical. The matchmaker doesn’t try to enforce variety, so indeed, unticking was the correct solution.
If you tell me you queue with multiple friends, for every map except Dust2, and it still takes more than 5 min in the queue then it means an overwhelming majority of players have selected only Dust2. I can do the math, but it’s likely more than 95%. With such high probabilities the vicious cycles you described will indeed appear. But 1) it’s perfectly understandable that the system favors Dust2 so much with such high probabilities 2) that problem could be easily avoided through better matchmaking implementation (though that has a tradeoff of slightly longer queue times for Dust2 players, which is one CS:GO devs probably didn’t want to take, given 10 min to be matched on other maps is reasonable) and 3) that has no reason to happen on DE except for maps that are extremely unpopular. ie: I think if opt-in was implemented in DE, you would maybe have 30 min wait time if you queue only Megarandom, but only 5 min if you queue only Islands.
That also makes sense, but please realize that’s a necessary tradeoff. You alone spend 5 min longer in the queue, but 5 players that want to queue only for Dust2 are probably spending 1 minute less and able to enjoy the map they want.
@Fano0517 I’ll give it to you that your argument is not completely garbage as I’ve become used to when people argue against opt-in.
I don’t play CSGO so I can’t really give my opinion on it, but I’ll take your word. However I don’t think that the comparison is 100% therefore I don’t believe that what happens in CSGO can be used to reliably predict what would happen in AOE2.
The opt-in system is basically an automated lobby, but instead the player gets encouraged to try different maps because;
-The presentation of a map pool.
-The reduction in queue time that selecting more maps brings.
-The player doesn’t have to host a lobby and wait alone in the hope someone will join.
The MM system with mappool is by design much more likely to allow variety to happen.
Now look at the past, did only arabia get played? No, at least we could also see arena and bf being hosted in the lobby.
Now add to that the different climate we live in, many new players that love to try different things, old players that have become more varied since DE (myself included).
If you combine all this, It really does not seem logical that with increased likelyhood for variety all of a sudden the opt-in system would behave even worse than a traditional lobby. All in all I don’t get this pessimistic view, at least we should take the scientific approach and test how the opt-in system would behave in DE.
That’s not a problem, it’s normal that you have to curate your picks.
Unsportsmanlike behavior and punishment system was discussed in this old thread, too (link is below). Lots discussed in the thread; but within, I proposed an ‘unsportsmanlike behavior’ score to be tracked to help decide the weight of any penalties issued. ALT+F4s could factor into this score…
“…(S)ome things the devs could consider include what age the player drops in, what age the opponent is in, what their score is relative to the opponent, how long the match has been going on, and maybe how often they resign, to help determine how unsportsmanlike the drop was. Some algorithm could generate an “unsportsmanlike behavior score” on the back-end to help decide on punishment. The algorithm could know what their historical unsportsmanlike score is based on past resigns/non-resigns and maybe factor that in, too.”
I now know it’s generally accepted by the community that players can and should be able to drop early with no ramifications, but I feel there should be some threshold for ‘too’ early. I assume ALT+F4 would fall into the ‘too early’ bucket for most players.
I can and I will. Just because a setting is popular doesn’t mean it’s good for the long-term, and getting rid of a popular option, regardless of the reason is suicide by patch. It’s not argumentative that selection queue cannibalizes the other less popular options and that is an objectively bad thing.
If the DE devs had any sense in them, they’d can the preferred map option immediately, as they can readily see what it’s done to the map variability when it’s not even a hard filter for map options yet. Take the hit that they earned by inserting a devilishly attractive idea that murders alternate play options, and recognize they’ve steered clear of the worse stretches further onward.
The preferred map option works well in 1v1s.
It could be improved, but there is plenty of map variety of you want it.
I understand in 4v4s it’s lead to trouble, and I suspect the best solution is a rework which removes map bans, but I don’t really want to talk about the multiplayer queue, since I don’t use it much.
SwaggyOP, you don’t appear to understand people like myself. This isn’t just based on this quote, also what you said about ‘just queue for Islands’ and other things.
Which is fine, but please understand that this lack of understanding leaves you unable to propose a solution which would satisfy everyone.
I think the minority I represent is larger than the minority you represent, based on votes and forum behaviour. I could be wrong. Regardless, an opt-in system is a bad compromise, because it it really bad for a significant number of players.
This makes zero sense, no one likes to be forced to play things they don’t like. Maybe you are confused about the meaning of words. I assume you mean that you like to be “surprised”, well you should feel right at home in the opt-in system. If you are open to try everything then just select every map.
Then comes the entitled part, you think that because you like something you get to force others into the same thing. This is just a selfish person talking.
I think the minority I represent is larger than the minority you represent, based on votes and forum behaviour. I could be wrong. Regardless, an opt-in system is a bad compromise, because it it really bad for a significant number of players.
I think you have a particulary bizarre and unreasonable opinion to be honest. You have made no reasonable argument against opt-in so your last statement is just ridiculous.
By the way I’ll never forget that some months ago you gave credit to the idea of auto houses in a designated area, which you followed up by saying “but microsoft wants to attract new players”. This is equally bizarre as the reasoning you’re giving here in this topic. Please keep it serious.
It’s really unfortunate that they chose to add the preferred selection instead of moving to an opt-in system. The opt-in system isn’t so narrow in outcome and fully respects player preference. Still it’s significantly better than what we had before, because it minimizes unwanted outcomes.
I do not claim to know what maps you and others precisely want to play, I claim that what you’re not doing is queuing for only 1 unpopular map. Yet in the CS:GO example, OP is doing exactly that, and thus not representative of a use cases people in this thread defend. Thus, the analogy is a bad analogy.
The examples of “queueing for any map except Arabia” and “queueing for only Islands” are purely illustrative examples, you can replace them with whatever you do and the logic will probably still apply, as long as it’s not “queuing only for one unpopular map”.
If OP says he’s queueing for multiple maps in CS:GO and still getting more than 5 min queue time, that does not seem compatible with what I would expect mathematically, and I would like to see stats or evidence of that, as everything I find on the Internet seems to hint at all maps being played fairly often in CS:GO and no complaints about map variety. Not doubting OP’s experience but I think he had unreasonable expectations of what a matchmaking system can do, and thought the 1 min queue time of dust2 should be the norm, when it was the exception. If you look in other games having ranked queue times of 5-10 min is perfectly normal even for a large playerbase (Overwatch). Besides, In 3v3 / 4v4, the current system doesn’t even do any better than 5 min, so why are we not applying the same logic here ?
(Note: for 1v1, queue times will be much lower than 5 min, I’m referring to 3v3 - 4v4, because the CS:GO example is 5v5. The more players to match, the longer the queue times).
Opt-in can never be a bad compromise, unless it’s poorly implemented.
The argument is mathematical: to maximize player satisfaction, the matchmaker should know all constraints that influence player satisfaction, and currently it doesn’t know which maps you really want to play, so it can only do a worse job at that than if there were unlimited bans.
Thus opt-in can only improve overall satisfaction of players. If can worsen the experience of YYY community, but then it must improve the experience of minority XXX much more. Opt-in is not a compromise imposed by minority XXX. It should be the standard decided by XXX and YYY to maximize the overall satisfaction of the community as a whole.
In other words:
If people who are not satisfied are a minority as you think, then opt-in would result in little to no downside for people who are satisfied (slightly increased queue times - which should be compensated by the fact Alt-F4 problem will disappear), and massive upsides for people who are unsatisfied.
If people who are not satisfied are in fact the majority, then that would mean we had been enforcing the majority to play with rules they didn’t want, in order to please a minority,. Then, indeed, opt-in will result in significant worse experience for said minority, but with the massive benefit of the majority regaining interest in the game. This will in turn increase player count and start various positive cycles, as well as solve the Alt-F4 problem. I think this is enough to compensate queue times and say that people who like map variety should be pleased as well.
The goal of a matchmaker is not to satisfy everyone, it is to maximize overall satisfaction. Which is why it is done in CS:GO. It is impossible to satisfy everyone. The current DE system doesn’t satisfy everyone either.
You assume incorrectly. Maybe you should have read a little more closely, or maybe I should have made an even longer explanation?
I’ll try to explain more explicitly: If there are only 2 maps in the map pool which I like (which most recently happened a couple of cycles ago), I’d prefer to play 80% of my games on those maps. I still prefer the current match system, even if it means I play only 60% of my games on those maps, because of the reduced waiting times. It’s a mutually beneficial social contract: some games I’ll play on someone else’s preferred map, and sometimes we’ll play on my preferred map, so that we can both be playing instead of waiting.
The effect could be a lot more pronounced on the team ladder, if the ladder was designed that way: some BF-preferring players would find it acceptable to have to play the occasional game on Team-Islands if in return they were able to play BF more often. I am not talking about surprise.
Of course, many players would find this unacceptable, and currently the match-making system is geared towards that group of players, and as a result almost all team games are played on Arabia or Arena. (Again, only have this from hear-say).
This is not a paradox, I am not talking about surprise, and it does make sense.
I see what you mean, however it doesn’t make sense to me because in return for the “saved” time in queue you’re now spending much longer in a map you don’t like. You could also wait a few mins longer in queue and play what you like. If I’ve tried a map and know I dislike it, then I don’t even want to start it. And if I do want to play it, it will be when I’m feeling like it. For example in the rare occasion that I want a chill game because i’m tired, I might want to do a BF game. But no way when I’m excited to play and open up aoe I’ll accept BF to be my first match of the day.
I guess we differ in opinion here.
However to expect others to play maps they don’t prefer because you don’t want to wait the appropriate amount of time for your desired map is not reasonable. That you’re willing to play unwanted maps in return for this is respectable, but not something you can expect from others.
Yes, we are different types of players in that sense. And I don’t see how a single queue could satisfy both types of players fully.
I do agree you should have the option to queue up without having the chance of randomly getting a BF game. Anyone who tells you to “go to the lobby browser” is a douche.
I do agree, hence my support for proposals to add some opt-in ques. (I would also support other compromise solutions.)
I hope you realise that proposals to remove the current system (which is almost 100% ideal for players like myself) isn’t reasonable either.
When the current system takes away my basic right of control over maps which was there for 20 years, then finding a better solution is not only reasonable but the just thing to do. DE came in, took something away, that was not their right.
Compare that to what you’re asking, you want to have shorter queue times which is a luxury. We’re talking about different magnitudes here.
I don’t bring up the argument as much that the current system makes getting arabia much tougher because I think this is a rather small argument in light of bigger issues that the system brings. But for example If I queue in 4v4 alone I could wait up to 1 hour before getting arabia.
A single system I suppose you mean, well opt-in gives you the option to play whatever you want. It’s just that you also want something which you cannot expect from others.
You can’t go in a restaurant and expect your whole table to order the same thing so the chef can do batch cooking to shave off a few mins.
I appreciate this.
Splitting it will make the current system worse and opt-in will not live up to it’s potential.
I have multiple reasons to believe that you could get decent queue times for less popular maps in an opt-in system, it could even allow slightly more maps (eg; 12) since queueing for more doesn’t significantly hurt, which would increase the potential for enjoyable maps.
The opt-in system will even entirely avoid you to have to play maps you don’t enjoy, the worst that can happen is that you select multiple maps and you’ll have to play your least favorite map within your own selection.
Maybe someone only likes 1 map, and it’s not popular. We’ll then they are by definition in the vast minority. In the current system they would get alt-f4’d. Luckily there seems to be a correlation beween players that enjoy variety and the less popular maps. At last in this case one must realise that that’s life, if not many want to play with you then you’ll have a hard time finding a match. Just as ironically it’s now a huge hassle to guarantee the most popular map because of this warped system.
