795 replies to this topic

[Joseph Mallan]

East Indy, on 06 April 2014 - 09:15 AM, said:

It won't easier to create illicitly larger premades, since there are currently no weight restrictions.

Nor does the premade limit have anything to do with weight classes; the limit could be implemented under the hood tomorrow.

I see what you're saying but your thread title is incorrect and misleading.



People will notice and players will get banned, and there may be a helpful, neighborhood watch-style list of units/clans so groups larger than four can be easily identified.

Remember, we're dudes playing a game run by other dudes actively monitoring us. It's not like there aren't social controls to prevent cheating.

Why? Its a team game. The TEAM is following the 3/3/3/3 restriction, They are FAIRLY balanced (Equal W/L Records). They are dropping as a 4 man and 3 solo. How are they cheating?

Edited by Joseph Mallan, 07 April 2014 - 03:15 AM.

[Mycrus]

Eglar, on 07 April 2014 - 03:01 AM, said:

http:8//www.reddit.com/r/OutreachHPG/

notice how it's more about game, content and how to deal with things instead of utopic "i want role-warfare, nao." or "I closed my wallet, pgi sucks"

if you put in something not-that-constructive you get downvoted or deleted by mod. that simple.

Nah it's just pgi sucks.

[Flaming oblivion]

I'm pretty new and I welcomed dropping with friends who offered there help to me learning , wasn't aware it was frowned upon and you know something i'll drop with them as often as I can . Because the fact is in a PuG reguardless of whats said in team you get no help or back up , even if you follow them They'll see you tryna retreat from a beating and block your exit. Or see you getting a beating and not help. I'd honestly prefer 1v1s or much smaller battles. People aren't prepared to work together for 12v12 to work . So I suspect this new change will change nothing in the grand scheme of things.

[Craig Steele]

Flaming oblivion, on 07 April 2014 - 03:29 AM, said:

I'm pretty new and I welcomed dropping with friends who offered there help to me learning , wasn't aware it was frowned upon and you know something i'll drop with them as often as I can . Because the fact is in a PuG reguardless of whats said in team you get no help or back up , even if you follow them They'll see you tryna retreat from a beating and block your exit. Or see you getting a beating and not help. I'd honestly prefer 1v1s or much smaller battles. People aren't prepared to work together for 12v12 to work . So I suspect this new change will change nothing in the grand scheme of things.

If you enjoy the game man, you keep on playing it.

This convo has been going on since the game rolled out, Solo PUG's are upset that pre preogrammed teams stomp all over them, Pre Made teams are upset PUG's aren't offering them enough of a challenge, (insert multitude of other reasons and acqusations from both sides here, the threads are numerous).

But everyone is united in blaming PGI. Whatever PGI do they alienate the other side and the circle keeps spinning round.

But if you're enjoying it, keep playing and having fun. Don't stop or change because the Forums are posting one way or the other.

Thats what the game should be about.

See you out there

[Joseph Mallan]

Craig, when we disagree its always respectably... but when we agree it is always awesome! You sure you aren't a StJobe Alt?

[Cybermech]

read first few posts and not guna bother with the rest.
not a fan of this 3/3/3/3 yoke btw.
but to me there is a higher chance of your "solo's" to be put with other assaults.
as we all know that the assaults and heavies are most popular.

so for me this what would most likely happen.
you guys drop, your solo's find a game and then the group has to wait because its in the biggest queue of all.

[Craig Steele]

Joseph Mallan, on 07 April 2014 - 03:39 AM, said:

Craig, when we disagree its always respectably... but when we agree it is always awesome! You sure you aren't a StJobe Alt?

LOL, pretty sure I am no ones alt

[Eglar]

also back to the topic:

Syncing into the same team. It is easier but it also really depends on how many people there are in your current pool. And if there are enough launching while you are searching the chance of syncing goes ridiculously low .e.g. not even worth trying.

[Craig Steele]

Eglar, on 07 April 2014 - 03:45 AM, said:

also back to the topic:

Syncing into the same team. It is easier but it also really depends on how many people there are in your current pool. And if there are enough launching while you are searching the chance of syncing goes ridiculously low .e.g. not even worth trying.

Now I assume this graph represent s the number of players pressing the launch button to find a team within the MM search time parameters? is that correct?

Anyone got any data or insight as to what the actual available population is during the 2 minute timeline MM is searching. Is 50 a fair number? (ie, MM is putting together 1 team every minute on average) Or is it closer to the 400 here? (putting together 8 odd games a minute?) or should we be stretching the graph further?

Edited by Craig Steele, 07 April 2014 - 03:53 AM.

[Joseph Mallan]

House Marik may have better data on this as they started Monday night mash ups.

[Magna Canus]

CMetz, on 06 April 2014 - 11:18 AM, said:

The competitive 12-man cue is what it is, and they'll be able to set their rules, etc. in private lobbies however they want with premium time, so why are people concerned with this?

The concerns I see about this is that the private matches the 12-man competitive players set up reap no rewards. Since Arty/Air strike are spam-cookies there, they will have to enter the public que sooner or later to "farm & restock". Or just to grind out mech xyz, or earn C-bills for whatever.

Considering that those who prefer to play in private matches are less than interested in playing with the "general populace" they are going to want to "get in, get it done, and get out" as efficiently as possible. The easiest means of success and maximized earnings is the classical sync-drop.

Currently the MM parameters are less "playable" because they are more vague than the planned Launch mode parameters. Where before you had 3 options to increase your chances (Timing, average ELO, Game mode) as of 29 April you will have 4 (Timing, ELO bucket, Game mode, Weight class distribution). ELO bucket is an improved parameter over average ELO since it is more clearly defined and thus easier to game.

Being able to game 4 parameters instead of 3 (or 2,5 vs. 4 if you consider the refined ELO parameters) is an increase no matter what % they actually end up being. In a worst case scenario it would make it 1/3 more likely that a sync-attempt is successful.

The one big factor that works against this would be a large player base, but I am not sure we have a large player base or not since PGI is keeping their stats to themselves.

[Craig Steele]

Magnakanus, on 07 April 2014 - 03:55 AM, said:

The concerns I see about this is that the private matches the 12-man competitive players set up reap no rewards. Since Arty/Air strike are spam-cookies there, they will have to enter the public que sooner or later to "farm & restock". Or just to grind out mech xyz, or earn C-bills for whatever.

Considering that those who prefer to play in private matches are less than interested in playing with the "general populace" they are going to want to "get in, get it done, and get out" as efficiently as possible. The easiest means of success and maximized earnings is the classical sync-drop.

Currently the MM parameters are less "playable" because they are more vague than the planned Launch mode parameters. Where before you had 3 options to increase your chances (Timing, average ELO, Game mode) as of 29 April you will have 4 (Timing, ELO bucket, Game mode, Weight class distribution). ELO bucket is an improved parameter over average ELO since it is more clearly defined and thus easier to game.

Being able to game 4 parameters instead of 3 (or 2,5 vs. 4 if you consider the refined ELO parameters) is an increase no matter what % they actually end up being. In a worst case scenario it would make it 1/3 more likely that a sync-attempt is successful.

The one big factor that works against this would be a large player base, but I am not sure we have a large player base or not since PGI is keeping their stats to themselves.

There's also the number of 'rolls' from MM to synch.

To implement this system of Roadbeers means one group and every one else solo, as opposed to the current 3 groups maximum (3 x 4 groups). So his 7 man group will need 1x4 and 3x1 allocations from MM to all get into the same game, whereas before it would have been only 2, 1x4 and 1x3.

Also, individual ELO's for individual mechs will drift over time from the "core" which will hinder future synching.

It's far from impossible, but I still don't think I would call it "easy".

[poopenshire]

This whole topic is based upon the idea and thought that:

• there are no good/decent light/medium pilots
  
• a light/medium mech cannot kill a heavy or assault
  
• pilots all have the same skill set
  
• THE OP HAS NO IDEA WHAT HE IS TALKING ABOUT... but I digest...

seriously folks, you need a hobby. if 3,3,3,3 bothers you that much..... PUG or do something else.

[Eglar]

Craig Steele, on 07 April 2014 - 03:52 AM, said:

Now I assume this graph represent s the number of players pressing the launch button to find a team within the MM search time parameters? is that correct?

Anyone got any data or insight as to what the actual available population is during the 2 minute timeline MM is searching. Is 50 a fair number? (ie, MM is putting together 1 team every minute on average) Or is it closer to the 400 here? (putting together 8 odd games a minute?) or should we be stretching the graph further?

Yes it is and it calculates a single player syncing with a full 4 man lance and disregarding the elo match-maker. I am not sure if I am correct about this but it was said that there will only be 1 premade per team for the new matchmaker which means that With around roughly 50 people in the que you will have about 21% chance of syncing into the SAME team as a 4-men pre-made if you take a mech-chassis that excludes theirs. (for example 4-Man takes 3xAssaults+1xHeavy and you take 1xlight).

While this might not be the most up-to-date information last year (September 2013) a few friends and me made new accounts and tried to sync over and over again. Out of roughly 600 matches, we estimated the amount of "people in queing up at the the same time" at prime time (6-9pm PST) at 24-288 (Average of 224.3) and otherwise it was around 24-120 (Average of 88.9) - Back than, the elo buckets were much tighter than now.

I just updated the same spreadsheet to calculate the current sync drop probability. While I understand that some of the formula is inaccurate it is in my opinion accurate enough to describe the chance on you to sync with your 4-mens and that there is also the elo matchmaker which is not accounted in many occasions.

Edited by Eglar, 07 April 2014 - 04:28 AM.

[Mawai]

Hi,

It is an interesting idea. However, it is based on the same flawed assumption as the basic idea of 3/3/3/3. Your situation can ONLY happen if there are exactly 3 of each weight class dropping all the time AND there are sufficient groups without being too many.

Matches can NOT be formed without either
1) 0 or 2 groups
2) 3 of each mech class

If this is not exact all the time then there will be BOTH group and weight class queues. Any sort of queue will immediately prevent the issue you have outlined from happening since the queues will likely be first in-first out in order to be fair.

In fact, if assaults are more common than any other mech in player preference (which seems to be the popular opinion though I don't have data for it) ... then your 3A/1H group could be waiting a long time competing with the single player assault queue while your buddy in the light or medium (which may be a less desirable weight class) will drop in a match long before your group.

The only time this kind of sync drop might work given the new constraints is the first minutes after the game comes up.

I agree that there are issues with the 3/3/3/3 plan ... but the one you point to won't happen in any realistic distribution of mechs entering the queues except the situation where everything is perfectly equal.

[Bobzilla]

Roadbeer, on 06 April 2014 - 08:59 AM, said:

Ok, I've said this in a couple other threads, but I figured I might as well put it up front, and also create a new topic so it's easy for me to find when the inevitable "I told you so" links need to come around.

Lets take what we know, and explain how easy it's going to be to game the system.

1.Each team with have 3 of each weight class, I don't think I really need to explain what that means.
2. Each player will reside in an Elo "bucket" for each weight class, and there are 3 buckets.


So, because my friends and I all play together, it's safe to assume that for the most part, we'll reside in the same bucket, and with the stats page, it'll be pretty safe to assume which 'bucket' we reside in for each weight class. It's not an exact science, but with a little guesstimating, it should be fairly easy to accomplish.

So, with that information, lets say I have 6 other friends on, making a 'group' of 7.

Knowing that each team will only have one group per side on it, we'll form a 4 player group leaving the other 3 as "solos".

The 4 player group will take maximum weight, because well, why would we want to give that firepower to PUGs. So the group will be 3A and 1H.

The "solos" will now take whatever mech they want, though it'll be safe to assume that they'll take the maximum weight still open, so "Solo" 1 & 2 bring heavy and "Solo" 3 will bring a medium or light.

Because Matchmaker is now pulling from weight classes, and with everyone launching at once, The group will be assigned to a match, and with MM looking to fill the remaining weight slots, and with our "solo" players meeting that criteria, I'd say the odds are pretty good that our "solos" will be filling those slots. At least, I give this a higher degree of probability than we have right now.

So, what does all of this mean?
The example I used was just with 7 people, the more people you have, the odds of a successful sync actually go up, ultimately, in the interest of making 'fairer' matches, they've actually created a system that is infinitely easier to game.

Like I said, it'll take a little guesstimating and trial and error at first, but I bet by the end of the second week, seeing 10 player "premades" in the public queue is going to be quite common.

Just leaving this here for posterity, do with it as you will.

What I see happening is just like you describe. Premades of 3 assaults and 1 heavy then others syncing. So all syncs will be searching for mediums and lights. While you attempt to sync drop all your lights and mediums will fill the gaps of the other sync drops in waiting plus all the assault puggers in waiting. This assumes that elo becomes less of a factor the longer the wait.

[Fooooo]

Koniving, on 06 April 2014 - 06:51 PM, said:

For shits and giggles, who rememebrs the Community Warfare plan? No, not the first one that got totally thrown out the window. Or the second one. The third one in the Launch party. Anyone?

Dropships and logistics. Totally gonna be worthless in a 3/3/3/3 system.

snip!....

Sorry for the snip, twas a good post.

Pretty much EXACTLY what I had imagined for CW.


Being able to send only limited amounts of stuff depending on certain conditions. (does your faction have any of X mechs left or about to come off the prod line from Y planet you hold ? Do you have enough dropships to send that much of a force ? Is it worth sending that much in the first place ?)

The asymetrical sides........just like your examples....... 1 team went heavy so has no reserve mechs and only 9 players vs the other sides ligher team, however they have 12 players and 1 reserve.

It all adds to the overall strategy in the CW game for corps and factions imo by a ton.


Having bland matches like we have right now, but as 3/3/3/3 in CW isn't really going to impress me very much.


For the pubs without CW involved yea ok it might be alright....however I'd rather have multiple decks........so 3/3/3/3 then next match might be a 1/4/2/5 etc etc.....at least that would spice it up a little not knowing what deck your match is..........(for pubs).........


As to the OP, yea it does make it a little easier to abuse....unless there is a ton of players online at the same time from what I understand...... (like 10k+ online at 1 time......which I doubt at this present moment is even close to achievable.... ) and then you still need a decent amount in the bucket your in for it to make it not easier to abuse....so yea its easier....

Edited by Fooooo, 07 April 2014 - 04:37 AM.

[Mawai]

Eglar, on 07 April 2014 - 03:45 AM, said:

also back to the topic:

Syncing into the same team. It is easier but it also really depends on how many people there are in your current pool. And if there are enough launching while you are searching the chance of syncing goes ridiculously low .e.g. not even worth trying.

Does this figure take into account that anything other than 3/3/3/3 weight class distribution will create queues? That these queues will prevent sync drops because folks will be pulled from the queues likely in an ordered fashion over which the players have no controls.

i.e. sync dropping no longer means entering the game at the same time ... it means entering queues for each weight class and group at the same time ... and since the queues are different lengths ... there will be a different wait for everyone before entering a match.

[Eglar]

Mawai, on 07 April 2014 - 04:31 AM, said:

Does this figure take into account that anything other than 3/3/3/3 weight class distribution will create queues? That these queues will prevent sync drops because folks will be pulled from the queues likely in an ordered fashion over which the players have no controls.

i.e. sync dropping no longer means entering the game at the same time ... it means entering queues for each weight class and group at the same time ... and since the queues are different lengths ... there will be a different wait for everyone before entering a match.

No, it does not. It only compares the odds of people matching the 3/3/3/3 rules vs people who pick a "random" mech.

[Magna Canus]

Eglar, on 07 April 2014 - 03:01 AM, said:

http://www.reddit.com/r/OutreachHPG/

notice how it's more about game, content and how to deal with things instead of utopic "i want role-warfare, nao." or "I closed my wallet, pgi sucks".

if you put in something not-that-constructive you get down-voted or deleted by mod. that simple.

You don't see stuff like:
Official apology from PGI
Competetive Gamestreams
PGI posts
Pre-release Mech Announcements
Smurfy updates (from dev)
Buildcrafting Next Hero Mech (which will probably be the next one released)
Or many top-elo players (they all don't want to come here) posting here. And no not saying that top elo players should define the game.

on these forums you open the "Game Balance" subforum and see haters everywhere, yea very insightful. If I was a pgi dev I wouldn't bother reading all this either.

LOL, simply LOL.

It's like never going to a certain place again because your angry ex is there and you can't face the fact that you screwed up big time. The coward will run and hide, the brave will be the bigger man and face the music.

Well at least they are not bribing people with MC and special benefits to white knight for them. Now that would be really low. So would secretly moving posts they don't like somewhere else because they are "undesired".