55 replies to this topic

[Raggedyman]

HumpingBunny, on 19 March 2015 - 08:45 AM, said:

I understand that we will probably never have the ability to select what map we would like to drop into, as disappointing as that is. However, I believe the next best thing would be the ability to de-select a map (or a few) from the choices of the match maker. That way, you'd know for sure you would not drop on a select map. Three major points to make:

1 - It would allow players to avoid a map all together. Let's face it, some of us absolutely hate dropping on specific maps. Sometimes landing on a map you hate can kill the match before it even begins. For me, I despise of river city night.

2 - It would allow PGI to collect a lot of information regarding which maps players really hate, and if we're lucky, they'll possibly reduce the presence of those maps in the rotation.

3 - Knowing you will not drop on a specific map, you can better choose your loadout and role of your mech. For example, If I could remove alpine peaks from the possible maps, I would not be concerned about dropping in with a big slow brawler.

Thoughts?

I think the ability to remove one or two 'least favorites' would be really splendid, especially now we are getting a decent range of maps. I think there is also enough range that if you dropped one or two you couldn't min-max to utter efficiency, so there wouldn't be too much advantage to people who go for the whole range.

Basically anything to get rid of Mordor, as I hate that dull-ass map

[warner2]

I would say there is no reason to despise River City Night. Turn your gamma up and it's like River City Day so there is no need to suffer there.

Sadly, there is no easy fix for Terra Therma it's just a crap race to the centre and a gamble when you get there. I'd deselect it in a heartbeat.

Edited by warner2, 20 March 2015 - 05:36 AM.

[EgoSlayer]

Pjwned, on 19 March 2015 - 11:41 PM, said:

There doesn't need to be a very big sample size if forum posts can easily demonstrate how those maps are bad though. There are some rather subjective complaints about maps like the ring in Terra Therma being too important with not enough access points or too many cliffs in Canyon Network or whatever, and then you have more objective complaints about issues like assault lances getting destroyed in their spawn location on River City in Skirmish due to the map being way too small or the all-important moutain in Alpine Peaks making most of the map irrelevant in Assault and Skirmish. Most of the hate for certain maps comes from issues in the latter category, and it shouldn't take a matchmaking experiment to realize that or to put more priority on fixing it; maybe after those maps get a pass (whether by reducing match sizes to 8v8 or changing the actual map or both or something else...) then there can be an experiment to see which other maps still annoy people.

I'm not saying that the forums can't be used for more detailed information about the complaints for the maps. But you are making assumptions that the issues you are listing are what would be the highest down voted maps. The hundreds of thousands of registered accounts that are part time players who don't post on the forums may vote differently. It's just an additional data point with a much larger sample to validate or invalidate the issues posted in the forums.

And map down voting would be better for the game play for people since they would have more say in what they are playing - which is the real benefit to the players. Data mining is a side benefit for both PGI and the players.

Edited by EgoSlayer, 20 March 2015 - 06:53 AM.

[Dimento Graven]

Either map selection should be in this game, OR, the ability to swap 'mechs out when the players see the map they've been assigned.

We ALREADY have that for CW, where it's possible to change your load out based on which "tile" your match has been assigned to, so why not add that capability to pub/group queues to?

So many of you people are incorrectly thinking that "well, in game lore, you really only had one mech and couldn't change it 'cause it was so expensive, blah blah blah", HOWEVER, you ignore the fact that:

Due to those very facts you mention a mercenary (the status I think the majority are playing as in this game have selected) TYPICALLY wasn't stupid enough to accept a mission/contract ill-suited for his 'mech, and TYPICALLY, the person offering the contract wasn't stupid enough to hire someone who couldn't field a 'mech appropriate to the mission.

So, I as a 'mech warrior wouldn't accept a contract where I and/or possibly more importantly my 'mech would almost certainly die being hamstrung by heat, or being in a place where short range weaponry is a handicap for 95% of the battle.

Nor would I as a person looking to hire said mercenary be willing to take on someone in a 'mech that couldn't handle close in city fighting, or would overheat the first alpha...

Map selection and/or 'mech selection should be in this game.

Map selection should EXACTLY like game mode selection, there being a list of maps that can be checked or unchecked as per the desires of the individual.

Map selection in groups would work EXACTLY how game mode selection works now. The group leader gets to choose which maps the group can drop in, just like the group leader decides which game modes the group plays.

It's all very simple and very logical, and now all the people who fear change can start their pissing and moaning and flaming...

[Nothing Whatsoever]

Whatever could be done, there has to be a tradeoff.

The devs already have a hard time with the divisions between the three game modes for creating matches, and CW has a separate set of challenges.

So, if we do see the ability to select maps, I imagine that creates the need to deny selecting / deselecting game modes and who knows what other possible changes.

[Dimento Graven]

Praetor Knight, on 20 March 2015 - 08:18 AM, said:

Whatever could be done, there has to be a tradeoff.

The devs already have a hard time with the divisions between the three game modes for creating matches, and CW has a separate set of challenges.

So, if we do see the ability to select maps, I imagine that creates the need to deny selecting / deselecting game modes and who knows what other possible changes.

I disagree with your 'imagination'.

The same 'rule loosening' timing mechanism, already in place, can be used to build matches so that it would take no longer than it currently does to build matches.

[EgoSlayer]

Dimento Graven, on 20 March 2015 - 08:12 AM, said:

<snip>
Map selection in groups would work EXACTLY how game mode selection works now. The group leader gets to choose which maps the group can drop in, just like the group leader decides which game modes the group plays.

It's all very simple and very logical, and now all the people who fear change can start their pissing and moaning and flaming...

I have no problems with anything else you wrote, except this. If you have map selection working like game mode selection with hard exclustions you are literally going to have thousands of players who can't be matched together because of conflicting exclusions. Match making would take forever. Bad, bad, bad idea. So it's obviously not that simple or logical to over look this simple fact.
Map exclusions *must* be a soft limiter based on weighting in order to keep the match creation window at something around a few minutes.

Edited by EgoSlayer, 20 March 2015 - 08:25 AM.

[Nothing Whatsoever]

Dimento Graven, on 20 March 2015 - 08:21 AM, said:

I disagree with your 'imagination'.

The same 'rule loosening' timing mechanism, already in place, can be used to build matches so that it would take no longer than it currently does to build matches.

Fine, just remember that depending on what gets implemented, we can either see less competitive matches or more competitive matches.

There was a fairly recent thread asking what did players prefer: http://mwomercs.com/...-player-choice/

[Dimento Graven]

EgoSlayer, on 20 March 2015 - 08:24 AM, said:

...Match making would take forever. Bad, bad, bad idea. So it's obviously not that simple or logical to over look this simple fact.
Map exclusions *must* be a soft limiter based on weighting in order to keep the match creation window at something around a few minutes.

You are incorrect. Match making would take no longer than it takes now. Match making works on soft limiters of elo and weight class.

Those same soft limits would still be in place, and the hard limit of 4 or 5 minutes of waiting for a match before MM just throws you into ANY available match would still be in play, limiting our wait times to the current maximum.

It's still simple, and it's still logical.

[Dimento Graven]

Praetor Knight, on 20 March 2015 - 08:38 AM, said:

Fine, just remember that depending on what gets implemented, we can either see less competitive matches or more competitive matches.

There was a fairly recent thread asking what did players prefer: http://mwomercs.com/...-player-choice/

Considering what I've seen in MM of late, and the fact that Karl Berg is no longer employed by PGI to work on it, I really think we've seen the 'apex' of MM development anyway.

Ultimately, I think what you've stated actually points out a simple, effective solution:

Combine group and solo queues.

"OH NOS!!!! ROFLSTOMP BAZILLIONS WILL BE HAPPININGS!" goes the panicked cry from those who haven't yet thought it through.

What was the original root cause of the initial separation of solo players and pre-grouped players?

The ultimate base of the root cause of that separation was a LACK OF IN GAME VOIP. This resulted in the "team work is OP" stance that the hard core solo players bitched about.

Well, now, we have in game VOIP. Even SOLO players can use the OP "team work" to play as effectively as any pre-made group.

It should NOW be possible to remove the separation of "pre-made" and "solo" players into one single non-CW queue, giving MM more to work with, a SINGLE MM to tune (vs. the current TWO) and it even allows for the here-to-fore impossible "11 man pre-made" (YAY FOR NO LONGER HAVING A 10 MAN GROUP REQUIRE YOU BRING A FRIEND TO JOIN IN ON THE FUN!).

I see nothing but goodness from this.

[mekabuser]

THIS IS THE ONLY FEATURE THAT MAKE ME SPEND MONEY AT THIS POINT.
goodbye bog, goodbye therma.

would gladly pay .
lulz.

[PappySmurf]

Mr.OP if your referring to the solo MM queues??? then I would say no leave it as it is and let PGI work on more maps-fix bugs-and content.

If your referring to the private MM Queue system it needs a total overhaul an needs to go back to MechWarrior4 style MM with a true lobby/launcher system(MSN GAMMING ZONE STYLE) and be free for everyone to use.

P.S in MechWarrior 4 you could pick the map,pick the game mode , pick weather , pick many options (tonnage) ETC. then drop into a match

Edited by PappySmurf, 20 March 2015 - 09:29 AM.

[HumpingBunny]

Thanks all for the feedback. I'll admit this discussion has me thinking beyond what I had originally mentioned in this thread. I'm gathering that at least some improvement can be made regarding map selection, or deselection. I wonder if it'd be all that difficult for PGI to open up the test server with a round or two of some implementation of suggestions in this thread.

[Mirumoto Izanami]

El Bandito, on 19 March 2015 - 09:17 PM, said:

I wanna get rid of Bog from my queue, so bad. Just can't see anything. And can't be sure if all of my shots hit or most got blocked by invisible walls between branches.

I pretty much run Bog in heat vision only. Makes it a touch easier to see mechs behind foliage, though the range klimitation can hurt. At least I tend towards brawling mechs!

Alek Ituin, on 20 March 2015 - 12:40 AM, said:

GET YOUR LOGIC OUT OF HERE, YOU DIRTY HEATHEN!

How dare you suggest something so reasonable!? Who are you to think you can come in here and make these demands that would improve the quality of the game!?!? Do you know how hard it would be for PGI to do this? IT WOULDN'T BE.

*exits stage left*

[EgoSlayer]

Dimento Graven, on 20 March 2015 - 09:12 AM, said:

You are incorrect. Match making would take no longer than it takes now. Match making works on soft limiters of elo and weight class.

Those same soft limits would still be in place, and the hard limit of 4 or 5 minutes of waiting for a match before MM just throws you into ANY available match would still be in play, limiting our wait times to the current maximum.

It's still simple, and it's still logical.

Sadly it's you that is mistaken on all counts.

First - you proposed a hard lock on the map selection:

Dimento Graven, on 20 March 2015 - 08:12 AM, said:

<snip>
Map selection should EXACTLY like game mode selection, there being a list of maps that can be checked or unchecked as per the desires of the individual.

Map selection in groups would work EXACTLY how game mode selection works now. The group leader gets to choose which maps the group can drop in, just like the group leader decides which game modes the group plays.

It's all very simple and very logical, and now all the people who fear change can start their pissing and moaning and flaming...

Game mode selection is a hard lock. If you deselect 'Conquest' you will never drop into a conquest game. Map selection 'Exactly how game mode selection works now' means that if you deselect a map you will *never* drop on that map.

But then you change your proposal and affirm that a soft lock as I stated in my post is the correct option.

Dimento Graven, on 20 March 2015 - 09:12 AM, said:

You are incorrect. Match making would take no longer than it takes now. Match making works on soft limiters of elo and weight class.

Those same soft limits would still be in place, and the hard limit of 4 or 5 minutes of waiting for a match before MM just throws you into ANY available match would still be in play, limiting our wait times to the current maximum.

It's still simple, and it's still logical.

There are no soft locks on game mode, but this soft lock you propose is a release valve. Release valves are sequential, not parallel. Which means that it excludes people from the match maker until a time limit is reached, then it re-evaluates the pool looking for new matches. By it's very definition it extends the time to make a match because it has a time limit before it ignores that criteria that excluded people. This is still a bad idea.

A map weighting factor, like how they implemented the soft lock on game mode that lasted a few days, doesn't add any time to the match creation because it doesn't exclude people based on their map selection. It only takes the microseconds to compute the vote tallies and adjust the RNG weighting on the map selection.

For something you keep saying is simple and logical, you seem to be missing a lot.

Edited by EgoSlayer, 20 March 2015 - 12:25 PM.

[Dimento Graven]

EgoSlayer, on 20 March 2015 - 12:25 PM, said:

Sadly it's you that is mistaken on all counts.

Well, we'll have to disagree then.

Quote

First - you proposed a hard lock on the map selection:

Game mode selection is a hard lock. If you deselect 'Conquest' you will never drop into a conquest game. Map selection 'Exactly how game mode selection works now' means that if you deselect a map you will *never* drop on that map.

Exactly how it should be, unless you, or your group leader, selects a particular map, it should never come up in the rotation.

Quote

But then you change your proposal and affirm that a soft lock as I stated in my post is the correct option.

You're horribly confused. I never affirmed that soft locks based on maps should EVER be implemented.

A user choice is a user choice, and it should be inviolate.

Quote

There are no soft locks on game mode, but this soft lock you propose is a release valve. Release valves are sequential, not parallel. Which means that it excludes people from the match maker until a time limit is reached, then it re-evaluates the pool looking for new matches. By it's very definition it extends the time to make a match because it has a time limit before it ignores that criteria that excluded people. This is still a bad idea.

A map weighting factor, like how they implemented the soft lock on game mode that lasted a few days, doesn't add any time to the match creation because it doesn't exclude people based on their map selection. It only takes the microseconds to compute the vote tallies and adjust the RNG weighting on the map selection.

For something you keep saying is simple and logical, you seem to be missing a lot.

I never proposed a soft lock, that's where you are confused.

And you compound that by an extremely bizarre and f'd up explanation of what you think is going on now I guess, and by your explanation seem to indicate you've never followed Karl Berg's posts on the subject.

There is a hard limit to how long the MM will attempt to put you in a match. My understanding is that it's 4 or 5 minutes.

AT NO POINT will a non-CW match EVER exceed the 4 to 5 minute hard limit on placing you in a match. If there are absolutely no matches to put you in, at the end of that 4 to 5 minute time period, you'll get the "Failed to Find Match" message, and can restart searching.

DURING that 4 to 5 minute time period, the soft limits of weight class and elo distribution are lessened, somewhere near the end of that 4 to 5 minute period, as I understood from what Karl Berg, the programmer who worked for PGI and did quite a bit of the work on the MM told us, that it drops those soft limits, hence you can end up with matches where the limitation of 3/3/3/3 isn't true, and why you can sometimes end up in a match where you 11 new guys with little skill on your team and the other side ends up with 12 people of extreme skill.

Your concern is that some how adding hard limits of what maps the users can be placed in will cause MM to increase the 4 to 5 minute search limitation.

No, that's not going to happen, at worst, you won't find a match once you hit that 4 to 5 minute threshold and you'll have to either drop again, or change your criteria and drop again.

Now, obviously with the advent of in game VOIP, we really don't need to separate out the "solo" and "pre-mades" into their own queue. The only reason we separated them in the first place was that it was difficult for pugs to coordinate due to the potential disperate VOIP servers that 12 different people could be on. As it is now, with in game VOIP, we can all communicate and facilitate team work.

So, let us dispense with the two separate queues, create one large queue, reduce the number of active MM's to only ONE, and in so doing reduce overhead, and effectively combine MM player base, AND, allow for everyone to select maps.

It will work.

[EgoSlayer]

Dimento Graven, on 20 March 2015 - 12:47 PM, said:

Well, we'll have to disagree then.

Exactly how it should be, unless you, or your group leader, selects a particular map, it should never come up in the rotation.

You're horribly confused. I never affirmed that soft locks based on maps should EVER be implemented.
<snip>

It will work.

So 'Match not found' would be the norm then. Genius. It won't work with hard locks.


EDIT: And kindly please show where my description of the soft locks is incorrect. As I have been following the Karl Berg threads and listened to streams with Karl and NGNG.

Edited by EgoSlayer, 20 March 2015 - 01:17 PM.

[Pjwned]

EgoSlayer, on 20 March 2015 - 06:53 AM, said:

I'm not saying that the forums can't be used for more detailed information about the complaints for the maps. But you are making assumptions that the issues you are listing are what would be the highest down voted maps. The hundreds of thousands of registered accounts that are part time players who don't post on the forums may vote differently. It's just an additional data point with a much larger sample to validate or invalidate the issues posted in the forums.

I'm not saying that though, I'm saying such glaring issues need to be fixed REGARDLESS of any map downvoting, and then maybe afterwards more subtle issues can be fixed after looking at some map downvoting data, though preferably not at all because downvoting maps is a crap shoot.

Quote

And map down voting would be better for the game play for people since they would have more say in what they are playing - which is the real benefit to the players. Data mining is a side benefit for both PGI and the players.

Yet it would make matchmaking worse for everybody.

[Pjwned]

Dimento Graven, on 20 March 2015 - 09:19 AM, said:

Considering what I've seen in MM of late, and the fact that Karl Berg is no longer employed by PGI to work on it, I really think we've seen the 'apex' of MM development anyway.

Ultimately, I think what you've stated actually points out a simple, effective solution:

Combine group and solo queues.

"OH NOS!!!! ROFLSTOMP BAZILLIONS WILL BE HAPPININGS!" goes the panicked cry from those who haven't yet thought it through.

What was the original root cause of the initial separation of solo players and pre-grouped players?

The ultimate base of the root cause of that separation was a LACK OF IN GAME VOIP. This resulted in the "team work is OP" stance that the hard core solo players bitched about.

Well, now, we have in game VOIP. Even SOLO players can use the OP "team work" to play as effectively as any pre-made group.

It should NOW be possible to remove the separation of "pre-made" and "solo" players into one single non-CW queue, giving MM more to work with, a SINGLE MM to tune (vs. the current TWO) and it even allows for the here-to-fore impossible "11 man pre-made" (YAY FOR NO LONGER HAVING A 10 MAN GROUP REQUIRE YOU BRING A FRIEND TO JOIN IN ON THE FUN!).

I see nothing but goodness from this.

Solo queue matches do not suddenly become coordinated with the addition of VOIP, it simply makes it a bit easier (in some cases) to relay some information, and that's it. You're wrong that the core issue of solo vs group was the lack of in-game VOIP, it has always been "these coordinated groups of players who probably know each other and how they play tend to have a big advantage over a team of 12 random people." You also obviously have the ability to complement your groupmates' mechs (e.g spotter + LRMs) while depending on certain builds to just show up in solo queue is a crap shoot which means yet another advantage for any groups that drop against random players.

Groups of players do not belong with random players who do not have a group, it's really difficult to argue against this.

[Dimento Graven]

Pjwned, on 20 March 2015 - 03:12 PM, said:

Solo queue matches do not suddenly become coordinated with the addition of VOIP, it simply makes it a bit easier (in some cases) to relay some information, and that's it. You're wrong that the core issue of solo vs group was the lack of in-game VOIP, it has always been "these coordinated groups of players who probably know each other and how they play tend to have a big advantage over a team of 12 random people." You also obviously have the ability to complement your groupmates' mechs (e.g spotter + LRMs) while depending on certain builds to just show up in solo queue is a crap shoot which means yet another advantage for any groups that drop against random players.

Groups of players do not belong with random players who do not have a group, it's really difficult to argue against this.

Oh, so just because 12 people belong to the same unit they're suddenly super powered or something?

No, don't be foolish.

It was absolutely the fact that these pre-mades could enjoy a level of communication unavailable to the solo pugs that is the core of the difference in play style, not that they all have the same unit tag, or theoretically get to play with one another more often.

I pug equally as much as I play grouped, and the unit I'm with happens to be one of the largest units in the game, and I can tell you that solo queue or pre-made queue the side that win more often are the ones where people communicate with each other and you don't have a majority of individuals doing their own thing.

It has zero to do with how often you play with the 11 other people you happen to drop with and more to do with what those 11 other people understand is the current strategy, which enemy is target priority, and where they should be moving to. ALL of these things more easily communicated via VOIP than text chat.

And yes, since in game VOIP has been made available I've been on teams where it has been 6 groups of 2 vs. 1 group of 12, and the group of 12 was ROFLSTOMPED due to the level of communication over VOIP. I've also been on solo queue teams where, again, it's very easy to see which side is communicating and directing the team and which isn't. The 12-0 stomps in the solo queue I've participated in regularly occurred when the team was using in-game VOIP, and the 0-12 losses I've experienced it was near dead silent.

This for me proves that the level of communication in game is now sufficient that SHOULD YOU CHOOSE TO USE IT, any group of experienced pugs can perform as effectively as any similarly experienced group of unit players.

What I've seen in this thread is a fear of change, "Oh no's it will be different and I will have to wait longer", no, not true.

"Oh no's it will be different and I will lose more often," probably not true, though given how some people have expressed their understandings of how the game functions, I can't imagine them winning all that much.