State Of Match Making - Feedback/comments
#81
Posted 03 September 2015 - 03:09 PM
Shift to 1/1/1/1 groups of 4 man MAX. I think that'll certainly help some.
Some other feedback:
Ever since PSR, I've had nothing but 12-0 stomps. It's getting really quite annoying. What's up with that? It's either full-win or full-lose, almost no close-call matches.
#82
Posted 03 September 2015 - 03:11 PM
FiglioDiBatman, on 03 September 2015 - 02:41 PM, said:
I will think on it again - MAYBE if it pulls just a very few it might be okay. I do not want to effect the quality of the solo queue.
#83
Posted 03 September 2015 - 03:11 PM
#84
Posted 03 September 2015 - 03:12 PM
Triordinant, on 03 September 2015 - 03:05 PM, said:
I'm hoping this has something to do with the commitment the company has to creating a simulation of mercenary combat ops. Many of us see unit combat as the goal and the public groups as a proof of concept for the MechSim.
#85
Posted 03 September 2015 - 03:12 PM
Edited by Appogee, 03 September 2015 - 03:13 PM.
#86
Posted 03 September 2015 - 03:13 PM
Night Thastus, on 03 September 2015 - 03:09 PM, said:
Well, TBH the score is typically a good indicator of a good/bad game, but I've seen plenty of 12-3 or 12-5 games that might seem like a stomp that were actually really well fought and great games. MWO matches typically land slide. Once one side is up by a couple and has favorable positioning, it quickly escalates.
#87
Posted 03 September 2015 - 03:13 PM
Problem 1. Groups saturate weight classes, leading to longer waits.
Problem 2. PSR matching, coupled with attempted 3/3/3/3, leads to longer waits. Sometimes stomps.
Problem 3. Using locale and mode to pare down available participants leads to longer waits.
Problem 4. Group sizes can lead to longer waits (i.e. if everyone's in even sized groups, and an odd sized group enters the queue).
Looking at the above, I don't think 2 is resolvable other than adjusting the release valve. That gives them 1, 3, and 4 to play with. What about the following: (bear with me, just spitballing)
Solution to Problem 1. Incentivize weight class spread, i.e. the entire group gets a cbill bonus if they avoid saturating any one weight class, or incentivize underutilized weight classes.
Solution to Problem 2. Adjust release valve. Nobody *wants* it, but nobody wants long drop delays either. Stomps are going to happen, it's the nature of combat. Look at tug-of-war - sometimes one team pulls the other off their feet. Sometimes it's very even, then all at once it skews in one direction. Sometimes it's back and forth to the end. That's just how it works.
Solution to Problem 3. Remove mode selection, leaving only locale selection - or maybe even have a release valve where it "prefers" your selection, but a release valve can put you on a different locale for *public* games. I know people don't prefer some game modes, but you can either be waiting, staring at a spinning robotic cat ******** (try to un-see it!!) or you can be playing - albeit in a game mode not of your choosing, but playing! Shooting! Having... fun!
Solution to Problem 4. Fix the size of groups. Not necessarily to "up to 4", but perhaps increments of 4, i.e. 4,8,12. An 8 man can easily find a 4man. A 4 man can find 2 more 4mans. No more worrying about that odd 9man that needs a 3man to join with. Alternatively, allow solo pugs to backfill the queues - I know it's not ideal, but with VOIP puggers can at least attach to one of the groups. Incentivize "backfill" players with a healthy cbill boost.
No, I don't have the best ideas - I am aware of that, but I'm trying to be constructive - I hope everyone else does the same. This is a game. It's a recreational activity. The moment it becomes more than that, it's defeating the purpose of recreational entertainment. Have a good day all, and GL,HF. - Fiero
P.S. - I'm not "white knighting", but I think that PGI's done a great job of at least "doing things". Instead of sitting back and letting this continue - here's Russ, engaging the public. Take a moment and realize - they are *trying*. I don't always agree with their decisions... but if *everything* always went your way you'd get bored pretty quickly.
Edited by Fierostetz, 03 September 2015 - 03:19 PM.
#89
Posted 03 September 2015 - 03:14 PM
I always thought the 2-10 was too hard on MM
Edited by Chemie, 03 September 2015 - 05:01 PM.
#90
Posted 03 September 2015 - 03:15 PM
Edited by xX PUG Xx, 03 September 2015 - 03:57 PM.
#91
Posted 03 September 2015 - 03:16 PM
- Don't enforce 1/1/1/1, they cant really work in lances at all.
- Remove odd groupings above 4 (so no 5/7/9/11).
- Don't enforce max groups of 4 (although I would like it). Sync dropping would definately come back, and everyone coming from steam will feel the pain and be dissapointed right at the beginning.
- Dont touch solo queue, its perfect.
- Embrace the LFG more! Its too hidden. Make it open from the start, so people have a lobby!
- Add options to only play 4 or 8 man matches, not always 12.
- Always match group-combinations exactly. So 4+6 can only play against 4+6. OR use PSR wisely.
- Open up all game modes in CW so 12 mans can do what they want there.
Edited by TexAce, 03 September 2015 - 03:20 PM.
#92
Posted 03 September 2015 - 03:17 PM
Shredhead, on 03 September 2015 - 03:08 PM, said:
Is this about appraising a game mode or unit continuity?
First, I'm sure the players regularly in Community Warfare agree with you a little, since it's far from perfect, but not entirely — or else they wouldn't play.
Second, what if PGI offered, through CW, the same or a similar experience with game modes as in solo queue and group queue, but for premade groups? You'd have game modes you like plus the ability to drop with 5 or more. Sound fair?
#93
Posted 03 September 2015 - 03:17 PM
Suko, on 03 September 2015 - 02:55 PM, said:
Very roughly but looking at Match Maker Command Center right now:
Perhaps 22% of the games are group queue and about 75% of those games are groups of 4 or less.So around 5-6% of MWO matches have groups of 5+
#94
Posted 03 September 2015 - 03:18 PM
Fierostetz, on 03 September 2015 - 03:13 PM, said:
Solution to Problem 1. Incentivize weight class spread, i.e. the entire group gets a cbill bonus if they avoid saturating any one weight class, or incentivize underutilized weight classes.
I like this idea. Incentivizing things is always the best way to adjust behavior. Forcing behavior with rules might cause the player hemorrhaging no one wants.
#95
Posted 03 September 2015 - 03:20 PM
Russ Bullock, on 03 September 2015 - 03:11 PM, said:
I will think on it again - MAYBE if it pulls just a very few it might be okay. I do not want to effect the quality of the group queue.
Limiting group sizes to 4 was the single worst decision ever made for this game, so many units were slaughtered after that change because they could no longer play together. Let solo players mix with group players at their own leisure. It's the only decision you should make here.
#96
Posted 03 September 2015 - 03:20 PM
Russ Bullock, on 03 September 2015 - 03:11 PM, said:
I will think on it again - MAYBE if it pulls just a very few it might be okay. I do not want to effect the quality of the group queue.
I would think this should be implemented with some sort of delay. "Try X seconds to match up a group queue game purely with groups, after X seconds start bringing in Solo puggers that have opted-in to fill the gaps".
I would also think that there would be greater control over the difference in skill levels when the MM gets to pick and choose single players to fill gaps, instead of being forced to fill gaps with groups that may be all over the place in terms of skill level. This would if anything improve the quality of the game by bringing the skill levels closer, and VOIP exists now so the advantage of communication is shared by all in the game, regardless of them being in a group or not.
I may be way off since I'm just some chump with a keyboard but there's my .02
#97
Posted 03 September 2015 - 03:21 PM
pbiggz, on 03 September 2015 - 03:20 PM, said:
Limiting group sizes to 4 was the single worst decision ever made for this game, so many units were slaughtered after that change because they could no longer play together. Let solo players mix with group players at their own leisure. It's the only decision you should make here.
We didn't have CW back then though...
#98
Posted 03 September 2015 - 03:21 PM
Jman5, on 03 September 2015 - 01:59 PM, said:
- Group size 2-4: 1/1/1/1
- Group size 5-8: 2/2/2/2
- Group size 9-12: 3/3/3/3
Russ Bullock, on 03 September 2015 - 02:52 PM, said:
Well this is precisely what 1/1/1/1 is across the board.
So your just saying to keep all group sizes as it is now but put the 1's throughout.
This could be a viable first option but it sounds many dislike that aspect as much as anything.
What I'm suggesting is to allow premade groups of any size like we have now, but create a dynamic chassis restriction based on the premade size. So a small premade group has to follow 1/1/1/1, but a larger premade group gets 2/2/2/2 or 3/3/3/3.
This helps alleviates the mech mismatch problem.
Whatever you decide, I think at the end of the day you're going to have to make an unpopular decision for the sake of faster and more accurate matchmaking.
Perhaps if you want to encourage smaller groups, you could add some sort of cbill reward for dropping in smaller groups. So you create a 120,000 cbill reward for victory that gets split evenly to the premade every time they win a match. The smaller the group the bigger slice of the pie the player gets.
Edited by Jman5, 03 September 2015 - 03:28 PM.
#99
Posted 03 September 2015 - 03:21 PM
Russ Bullock, on 03 September 2015 - 03:11 PM, said:
I will think on it again - MAYBE if it pulls just a very few it might be okay. I do not want to effect the quality of the group queue.
I would think that this is the way to go. If I solo queue, I'd much rather be a solo in the group queue. I do make use of the in game voip etc, so the group play is something that a lot of us are very familiar with. I would think you would want to limit it to higher tier however.
#100
Posted 03 September 2015 - 03:21 PM
Monitor 1001, on 03 September 2015 - 03:18 PM, said:
I like this idea. Incentivizing things is always the best way to adjust behavior. Forcing behavior with rules might cause the player hemorrhaging no one wants.
From my perspective, forcing people to do things makes them complain. Positive reinforcement gets them treats. Which one quells the complaining more efficiently
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users