Trivial to grief, games can take an indefinite amount of time. The exact proposal you have may be a non-starter, but you could easily adapt that in some way to something that adds time to start under some conditions, or whatever.

Possible, only if the game is extremely poorly designed making it harder than it should to make simple changes. And i've never developed on here so no way for me to know. I have developed for many years on other large mmo's and this type of thing is trivial on a properly designed game. I could make the change in about 1 day and send it down for testing.

I'm just not good enough to compete with the veterans. My ship is fully upgraded so the game thinks I'm good, but I'm useless. What can I do?

Start by learning how to fly a Legion or Warcarrier bomber. This will lower your dependance on aim and give you a chance to be productive while still practicing aiming, evading enemies, and making the best use out of your missile breaks.

Once you get a little better at aiming your lasers, you might be ready to try out gunships, which get killed quickly by elite scouts and strikers, but give you the advantage of usually getting the first shot off against an opponent.

Hey there folks!

 

I just wanted to pop in here really quickly to add a little clarity to this conversation. With the changes that came in 5.9.2 there were no changes made to the match size logic. Starfighter matches have always attempted to make a 12v12 match first and upon failing to do that they create 8v8 matches. The changes in frequency of 12v12 seem to be caused by the larger pool of players that have come from the cross factional queues.

 

We are watching the player sentiment on this. If 12v12 matches being more normal is harmful to the game mode than we will take the time to re-evaluate how matches are created.

 

I hope this help.

Thanks,

Bret

It is only harmful to the game in the way it's being implemented now because teams are not filling up evenly.

The first priority of the matchmaker should be to build an even 8 vs 8 game.

After that, it should try to match the next 2-8 players in the best pairings that it can to fill out the teams evenly.

What needs to be avoided are the "Warzone will end in 30 seconds" endings that have become more frequent as the matchmaker fills up one side and not the other. But with regards to 12 v 12 against 8 v 8 matches, there is nothing wrong with 12 v 12. The people who like to make dominant 4 man groups don't like it because it's harder for them to pervert the outcome by stacking one team. The more people there are to fill out teams and balance their group, the better it is for balance and competitive matches.

One thing that might be a consideration to affect the "race to the satellites" component of a domination match. Eliminate the tensor ability. It only serves one purpose in the game and it's so useless otherwise that people self-destruct their ship after using it at the beginning of the match. Minus the tensor ability, you have more incentive for people to fly scouts at the beginning of a match to get to the satellite quickly and no need for a team to consider starting with the weakest scout just to use its one game-changing ability.

I think there are two separate issues, and neither of them is wholly dependent upon the other. I don't know the best way to solve the backfill problem other than extending the time before a match starts and how long a player has to accept a match. That being said, 90 seconds should be enough time if you are actually in game. People who are AFK should not be queued. Sometimes I get called away emergently and I cannot take the time to un-queue, but most of the time I try to un-queue if I know I'm leaving the keyboard. Its just rude otherwise. However, having a bit more time before the match starts to get those people that got queue notifications after the initial batch went out, that sounds reasonable.

Regarding the 12v12 issue: I am sympathetic to the things RedFantom posted. I would make the observation, though, that often times, there are a few people on either side that backfill into a 12v12, prime time on Satele Shan. That implies that there are probably just a few over 24 people in queue at prime time. Since there is no simultaneous 8v8 game occurring with the 12v12, that means there are between 24-40 people in queue at prime time. But, if there are at least 32, then we could have two simultaneous 8v8's going on. With all that being said, for people who queue for several matches in a row during a play session, do you see enough different faces to suggest that there are between 32-40 people in queue? I don't. Most of the time, the back to back matches I see are the same names, suggesting that there are between 24-32 people in queue. This is even further emphasized by the off peak hours. I hardly ever get even an 8v8 pop between 3 am EDT and 7 pm EDT. That tells me that there just is not a lot of interest in GSF, for better or worse, and I can't imagine the peak hours are much more than 28 or so people. Bret suggested as much during his interview with the BadFeeling podcast guys last week. Maybe its better on Star Forge, but I don't want to leave my guild on SS or pay the cost to move all 16 toons there.

Therefore, if 8v8 were prioritized over 12v12, I think people would be waiting much longer for a match to pop. If there are, say, 28-30 people in the queue, and 8v8 is prioritized, only the first 16 people will get a pop. The remaining 12-14 people will then need to wait until that match is completed (12 minutes or so) and wait until the matchmaker decides on the 2-4 people from the recently completed match who get the privilege of another pop. Whereas if 12v12 are prioritized, less people are waiting for a match pop.

And, as other people have already observed, it is easier to make an even match when you have 24 people to work with instead of 16, and the impact of an individual premade is lessened. I disagree with a previous poster's observation that they see fewer satellite control changes … I've seen the opposite. While I have at times posted some agreement with the idea of separate queues for solos and groups, I've always come back to the notion that there just aren't enough people to support that, and the last few weeks of frequent 12v12 and infrequent 8v8 confirm that to me. So, I think the only way to "deal with the premade problem" that doesn't punish people for grouping up is precisely this method, since my suggestions for incentivizing solo-queuing have not gained much support.

In the end, I'm just happy that I see more matches and more even matches at that. Taking 5.5 onward, GSF has become progressively more fun for me and it remains one of my go-to activities overall. With respect to Red and his colleagues, I think pop-frequency, ability to make more even teams when dealing with a larger pool, and reducing the impact of individual premades are important reasons to leave the system as is.

Eliminate the tensor ability.

This is a terrible suggestion. Most matches I flew a type 3 scout, even prior to 5.9.2, I did not tensorcide. In fact, I see very few people tensorciding these days. I think it can be a very useful ship in the right hands. Furthermore, tensoring without having any backup, let alone a beacon bomber following up on your approach to a satellite, doesn't end well for the sole tensor scout, so its hardly overpowered. I see plenty of all three types of scouts in matches now … my debuff bar full of EMP and my deaths at the hands of battlescout pilots are all you need to see to know that scouts are still a-plenty in the meta.

Possible, only if the game is extremely poorly designed making it harder than it should to make simple changes.

No, that's not what I'm saying at all. I'm saying that exact solution would allow for griefing while also putting the game into an indefinite state. It has nothing to do with how hard it may or may not be for Bioware to make the change, and everything to do with how adding to the griefer arsenal might make things bad.

What needs to be avoided are the "Warzone will end in 30 seconds" endings that have become more frequent as the matchmaker fills up one side and not the other.

Yes, this is absolutely unsatisfactory. This needs to be top priority.

But with regards to 12 v 12 against 8 v 8 matches, there is nothing wrong with 12 v 12.

There's plenty wrong with never having 8s. Each map is very different in 8s and 12s. There's plenty of people ticked about it, but they seem to all be ************ on various Discords. Discord is really a negative force for discussions, openness, and privacy.

The people who like to make dominant 4 man groups don't like it because it's harder for them to pervert the outcome by stacking one team.

Did they tell you this? I run in a four man team for most of my games, and this is not my reason for wanting 8s/12s to be closer to 50/50. Can you quote someone who actually has complained about not being able to win because of 12s, or anything to this effect? Or is it just, you want to believe that because you always have a beef with people who play with teams?

The more people there are to fill out teams and balance their group, the better it is for balance and competitive matches.

If this is true, it's only barely so. And I'm not sure it is true anyway.

One thing that might be a consideration to affect the "race to the satellites" component of a domination match. Eliminate the tensor ability.

Lol. Eliminate? Here's how this conversation starts. You, or someone else, comes in and says "hey, we should start tensor on cooldown because it is too good at the start of a match." Then people discuss that, which I don't agree with. Tensor is great for team play, a cool buff, good in general, and very good at the start of a match.

It only serves one purpose in the game

Tensor field is a truly excellent ability that I have plenty of uses for. I routinely use tensor to:

(1) Refill teammates energy in domination after they run someone to ground, or as part of that.

(2) Greatly reduce time to go from satellite to satellite, especially on Denon.

(3) Counteract the two biggest weaknesses of the bomber class, engine inefficiency and poor top speed.

(4) Ensure dominance on node by buff the speed and turning of myself and at least one ally.

(5) Orbit kite on node with a defensively built type 3 scout (probably my generally most common play on domination)

(6) Help teammates who are dealing with gunships off node by providing them with greater evasion stat and line of sight ability.

(7) Get allies who are paused for breath behind a rock back into the battle much faster.

(8- bonus for this broken patch!) Tensor at the start of a TDM that is going to end soon, so that my team can yolo into battle and hopefully get that one precious kill.

and it's so useless otherwise that people self-destruct their ship after using it at the beginning of the match

This is relatively uncommon now. I'm sure there's still some time where it is better to tensorcide, but I don't think it's with a 4 man team. Perhaps a full premade 8v8 or 12v12, but even then, having a tensor on the map really helps.

Minus the tensor ability, you have more incentive for people to fly scouts at the beginning of a match

Scouts get to the node faster whether they have tensor or not. Tensoring a type 2 scout all but ensures that your team will get to the node faster than a team that doesn't tensor, or doesn't run a type 2 scout. Doing the same with a type 1 scout can ensure it, but they aren't appreciably faster than type 2 normally.

the weakest scout

I practically main this ship in domination man. It's not the weakest scout. It has serious defensive and utility capabilities. If my team was going to lose the use of a single scout permanently via imperial decree, we'd choose the type 1 scout. Which is also a good ship- we could just get away with its loss.

This is a terrible suggestion. Most matches I flew a type 3 scout, even prior to 5.9.2, I did not tensorcide. In fact, I see very few people tensorciding these days.

Quoting pyro for truth. It's rare to see a tensorcide now, the type 3 scout is normally flown by someone who wants to fly the type 3 scout. I can imagine a scenario where tensorcide would be optimal, but it's some niche comp that would require really more than 4 in voice versus a team where that matters. Even then, I think you'd be double tensoring and the double left tensor guy would SD and respawn at a beacon. If you aren't running two tensors or don't have a beacon bomber, you definitely wouldn't do it.

The more people there are to fill out teams and balance their group, the better it is for balance and competitive matches.

 

If this is true, it's only barely so. And I'm not sure it is true anyway.

Well, the larger the sample size, the more likely one would see a Normal (Gaussian, bell shaped curve) distribution. Smaller sample sizes are more susceptible to the influence of extremes. If you have a field of 24 pilots, versus a field of 16 pilots, the chances of "highly experienced" (which we now know to be a function of games played in the entire legacy and the highest requisitioned ship on your bar) pilots being evenly distributed across the field is greater … assuming matchmaker is working as intended, and there are no groups on either side, or there are an even # of 4-man teams. Perhaps 8 more pilots isn't enough to see a difference over 16 people, though. Balance is obviously more than merely a function of games played per legacy and mastered ships, but statistically, I believe Hal's comment is reasonable.

Well, the larger the sample size, the more likely one would see a Normal (Gaussian, bell shaped curve) distribution. Smaller sample sizes are more susceptible to the influence of extremes. If you have a field of 24 pilots, versus a field of 16 pilots, the chances of "highly experienced" (which we now know to be a function of games played in the entire legacy and the highest requisitioned ship on your bar) pilots being evenly distributed across the field is greater … assuming matchmaker is working as intended, and there are no groups on either side, or there are an even # of 4-man teams. Perhaps 8 more pilots isn't enough to see a difference over 16 people, though. Balance is obviously more than merely a function of games played per legacy and mastered ships, but statistically, I believe Hal's comment is reasonable.

GSF experience is more like an exponential distribution than a normal distribution. You're right though, more players means matchmaker has more to work with, assuming it is not allowed to skip players who had queued longer.

This gives me an idea. Suppose there are 22 players in the queue, composed of 4 new players, 16 average players and 2 aces. I guess matchmaker would launch 11 vs. 11, with 1 ace, 2 new and 8 average on each side, fill to 12 vs. 12 later if more players queue. But wouldn't it be much better to launch an 8 vs. 8 match with the 16 average players and skip the aces and new players? Obviously there would have to be a limit how many times it can skip a player.

Actually, there would be no need to skip new players in a Domination.

GSF experience is more like an exponential distribution than a normal distribution. You're right though, more players means matchmaker has more to work with, assuming it is not allowed to skip players who had queued longer.

 

This gives me an idea. Suppose there are 22 players in the queue, composed of 4 new players, 16 average players and 2 aces. I guess matchmaker would launch 11 vs. 11, with 1 ace, 2 new and 8 average on each side, fill to 12 vs. 12 later if more players queue. But wouldn't it be much better to launch an 8 vs. 8 match with the 16 average players and skip the aces and new players? Obviously there would have to be a limit how many times it can skip a player.

 

Actually, there would be no need to skip new players in a Domination.

I do think that both ground and GSF warzone queues also take into account the length of time you are waiting.

So, we know the following are variables:

1. Are you in a group or not? This is more of a binary value, groups are prioritized over solo
   
2. What is the number of GSF matches played across the player's whole legacy?
   
3. How much requisition is spent on each ship on the player's bar?
   
4. How long has the player been waiting?
   

Let's assume that all 22 players queue at nearly the same time, merely for the sake of simplicity. Also for simplicity, let's assume all players are solo-queueing. I don't know what the actual rating is called, but lets call the result of the combination of "total matches across legacy" and "the amount of the ship with highest amount of requisition invested" as the composite experience score, or "score" for brevity.

I'll go ahead and make a supposition, or an inference maybe, that a player who has a high "score" is reasonably invested in the GSF game mode. They enjoy it, and they care about development changes. I don't think the invested players are ones you would want to skip, as a game designer.

Similarly, we don't want to upset a new player, or rather, we don't want players new to the game mode to be waiting a long time for a pop. That's a big "meh" on the new player experience and isn't likely to attract more blood. "I queued for 30 minutes and never got a pop. I could have done 2+ ground matches in that time, or an Uprising, or a daily area. Why should I waste time on GSF?" During the BadFeeling podcast interview with Bret, he specifically mentioned that he has ideas and plans for improving the GSF learning curve. That tells me that they at least care somewhat about attracting a larger audience. Prolonging queue times for novices with low composite experience scores doesn't help that.

Most importantly, its a self-defeating system. No one is motivated to increase their "score" because it means they won't get pops. No one is able to get out of the novice zone because they get passed over for pops too.

Personally I like the 12v12's better then 8v8's. It makes it harder for a group of 4 to completely dominate a match, if the match maker is working properly.

Or, as has been suggested many times before, how about removing group queue from GSF? Premades have no place in GSF, and have been the single largest contributor to frustratingly lopsided matches since the format was introduced.

Or, as has been suggested many times before, how about removing group queue from GSF? Premades have no place in GSF, and have been the single largest contributor to frustratingly lopsided matches since the format was introduced.

I disagree that premades have no place in GSF. I think that it's always more fun to play with friends whether you're good or not. The problem with matchmaking is the need for GSF to have equal numbers. Each player should have a Pilot rating based on their play history and equal ratings should play against each other regardless of numbers. This could give us matchups of 4 on 8 or 6 on 12 or whatever makes the groups equal in "skill rating" and not in quantity of pilots. The numbers wouldn't be even and some of the matches would still get lopsided but now you're challenging your better players to overcome possibly large number disadvantages and your less skilled players at least get a chance to improve their skills in matches where they have a fighting chance to maybe win if they work together instead of just being cannon fodder.

I'm actually of the belief that 4v8 is not playable as long as any of the 8 can find their boost key, but I don't dispute your overall idea. The problem is that it is jumping the gun.

Right now we don't even have skill based matchmaking- we have a matchmaker that will roughly try to pair long term legacies against each other, and will *probably* handle groups sorta right (counting them as 4 copies of the most advanced legacy). This is an improvement over before, when the fabled matchmaker was literally just "YOLO WHATEVS", apparently.

Your first step would be to start writing some code that will make a reasonable guess as to skill level. Probably you want to base this off of wins/losses, but there's plenty. The point is, if you have some guess about this, you can then start actually matchmaking against a number or an ordered set, presumably with your willingness to make an unfair match going up the longer the longest-waiting player has been chilling in queue. Once we have that, we can start guesstimating how smart it is, and start talking about 9v7s or 12v10s or even 9v6s or whatever.

I disagree that premades have no place in GSF. I think that it's always more fun to play with friends whether you're good or not. The problem with matchmaking is the need for GSF to have equal numbers. Each player should have a Pilot rating based on their play history and equal ratings should play against each other regardless of numbers. This could give us matchups of 4 on 8 or 6 on 12 or whatever makes the groups equal in "skill rating" and not in quantity of pilots. The numbers wouldn't be even and some of the matches would still get lopsided but now you're challenging your better players to overcome possibly large number disadvantages and your less skilled players at least get a chance to improve their skills in matches where they have a fighting chance to maybe win if they work together instead of just being cannon fodder.

The matchmaking system implemented in 5.9 has not dramatically improved the fairness of GSF matches. That's a fact. The biggest factor in lopsided, unbalanced matches has been premades, not the evening of teams. That also is a fact. While the discussion throughout this thread has been about keeping teams even, which is also a problem in GSF, whether or not they manage to fix the queue system to address this won't prevent lopsided matches from happening due in large part to.....premades. I've been in hundreds of matches where the teams were even, but one had a premade of vets in VOIP and the other did not, which ended in a miserable loss for the team without the premade. I've also been in matches where the teams were uneven, and the one with fewer people but that also had a premade still won. The advantage of a premade in GSF cannot be overstated, and has a far greater impact on the outcome of the match than a premade does in regular WZs.

So while I completely understand the argument of wanting to queue up with friends for PvP, unlike ground PvP (where premades are fairly commonplace), GSF doesn't have enough interest to balance that out. In other words, there aren't scores of guilds queueing up premades for GSF. And anyone who's being honest here would admit that the biggest detriment to GSF is how rarely you actually get a match that's even, fair, and fun for all participants - not just the Epeeners in a premade cackling in Discord as they obliterate the other team with little or no competition.

But by all means, keep the status quo in GSF. It's clearly working to make it a popular and engaging PvP format.

I've been in hundreds of matches where the teams were even, but one had a premade of vets in VOIP and the other did not

Vae Victus

The advantage of a premade in GSF cannot be overstated

TEAM GAME REWARDS TEAMS

NEWS AT 11

Show us on the doll where the premade touched you.

Also, the matchmaker keeps stacking recognizable veterans on a side - whether there's a group in queue or not - because the measure it uses to rate a player, i.e. legacy number of games, is absolutely terrible. Interestingly, I haven't seen the matchmaker put two groups on the same side of the map yet. So if you'd like to be more proactive, form a group and use the GSF Discord to take on whatever group is giving you trouble. You might even make friends along the way!

Definitely excited about the custom matches, because it will allow events or even friendly competition the opportunity to create balanced or even intentionally unbalanced matches (for training, extra challenge, etc) in the future. What we've got is still better than single faction matchmaker, in that people can no longer stack one faction when they start to lose. The pilot rating definition just needs to be tweaked to work properly; something that is on the developers, not groups of friends who want to play together.