No replies to this topic

[Russ Bullock]

The purpose of this post will be to primarily discuss the state of the Group Queue and if any major changes will be made to address both wait times and quality of matches. In order to do that we need to first discuss the solo queue so we have something to compare against.

Another main item to keep in mind as we discuss this is to keep mindful that for a smaller online game when compared to the big monster products, we have an awful lot of options for our players. Certainly more than we ought to have if our number 1 priority is the competitiveness of matches weighed against wait times.

These options across solo and group queue consist of the server selection, game mode, PSR, group size, and group weight class composition.

Another thing to keep in mind is that the NA and EU population peaks are very similar and remain at least 2.5x higher than our low population point, which puts further strain on the MM during those low hours.

During the past couple of weeks we have been playing with a lot of settings while trying to allow PSR to settle. This has allowed us to find what we feel is the best MM options for the current feature set and population of the game.

Below I will discuss the current state of the solo queue and then the group queue, in each case I will list the main stats that matter to players the most.

Solo Queue:

Average Wait time: 142 seconds
Average Skill Diff between teams: 153
Teams always match the same weight classes per side no exception.

This currently has created the best feeling public matches in MWO we have ever had, very happy with this and even at the lowest population point of the day with current settings it should hold up fairly well.

We still encourage players to keep multiple servers selected wherever possible and to keep all game modes selected for the best matches. It is still true that even solo queue could be improved noticeably further if the game modes were not hard stop options for players. However in solo queue things work well enough it does not become a huge issue.

I am loathe to do anything to the Solo queue that will degrade it to any degree, I feel group queue needs to solve it's own problems not join or take players from the solo queue.

Group Queue:

We have been doing a lot of playing around with values since both the new servers and PSR came online and we feel we have found settings that work best. Unfortunately for group queue this means fairly heavily loosening of the valves and criteria to keep wait times down. Here is where the current stats of group queue sit:

Average wait time: 190 seconds.
Average Skill Diff per team: 405

At first blush it seems pretty good on the wait time, and the skill difference although more than double the solo queue seems reasonable for group matches. Unfortunately there is more to the group queue story as there are still two other ways matches can be imbalanced. First off is the amount of groups per side is very hard to keep even, we will frequently have 4 groups vs 2 groups and 3 vs 2 and so on. Also currently after waiting 240 seconds it will give up on trying to match the weight classes per side and you can be further uneven by having more heavies for instance on one side. Sometimes this can be made up for by having more assaults on the other but it's quite random.

So here is where we are at and what we propose. You play with these new settings for group queue and see how they feel to you. Basically the question will be is the wait times vs quality of matches acceptable enough to you. If the answer is overwhelmingly yes then we can keep our current feature set of any size groups and the weight class restrictions on how you create them. Perhaps if the wait times are 190 seconds on average you will accept the quality of matches in order to keep your large groups and so on.

If on the other hand you are unsatisfied with the quality of the games, here is what we are going to do.

- Shift back to a maximum group size of 4 or less.
- Each group needs to be created in a 1/1/1/1 fashion.

This will remove a significant amount of jig saw puzzles from the MM. Imagine no longer needing to deal with odd sized groups like 5, 7, or 9. This is helped further by all the smaller groups of 2 and 3 not being made up of primarily one weight class as is the case now where we see a group of 3 heavies.

This would be our first point of action. If we need to go further than we will look at the game modes.

So please provide your comments in the feedback thread. We will be monitoring it and making a decision shortly.

Feedback thread can be found here: http://mwomercs.com/...edbackcomments/