Jump to content

Feedback On Min/max Tonnage For Each Group Size


435 replies to this topic

#41 Saint Scarlett Johan

    Member

  • PipPipPipPipPipPipPipPipPip
  • Hearing Impaired
  • Hearing Impaired
  • 3,349 posts
  • LocationOn the Delta side of Vicksburg

Posted 29 September 2015 - 01:20 PM

I REALLY hope you don't tell lights to bite the pillow (again) by making the minimum tonnage for small groups enough that a pair of lights or a trio of lights is a no go.

Seriously. I can see you lowering the max average tonnage of larger groups to 60 to prevent Thunder Runs and Black Cat Herds, but to prevent a pair or trio of lights?

I mean, really, taking lights already means you're the underdog...

#42 Tina Benoit

    Community Manager

  • Developer
  • Developer
  • 817 posts

Posted 29 September 2015 - 01:21 PM

'Group of 11' row in the table has been removed, as groups of 11 will not yet be possible.

#43 Saint Scarlett Johan

    Member

  • PipPipPipPipPipPipPipPipPip
  • Hearing Impaired
  • Hearing Impaired
  • 3,349 posts
  • LocationOn the Delta side of Vicksburg

Posted 29 September 2015 - 01:22 PM

View PostXX Sulla XX, on 29 September 2015 - 12:19 PM, said:

Question for those wanting no lower limit on small groups. The way Russ wants it to work is to not do weight matching in the matchmaker. Would you be OK with one side getting say 3 2 man groups of lights and the other side getting 3 2 man groups of heavies?


Yes, because the only way lights can even threaten anything else in the group queue is as a pack.

#44 Tarogato

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Civil Servant
  • Civil Servant
  • 6,557 posts
  • LocationUSA

Posted 29 September 2015 - 01:31 PM

I think the minimum tonnages for the small group sizes is a bit punishing. Light mechs are typically the least played and most in demand by the matchmaker, yet under the new system we cannot play two light mechs in a group of two and we cannot play three light mechs in a group of three. Some of the most fun I've had playing this game were when I was in a group of three playing in a light mech wolfpack. We're usually the only three lights on the team and I wouldn't exactly call it an overbearing tactic.

The minimum for groups of two should be 40, the minimum for groups of three should be 75 or even 60, and to bridge the gap the groups of four should be something more like 180. If people *want* to drop undertonned, I think they should have that option.

#45 Russ Bullock

    President

  • Developer
  • Developer
  • 909 posts

Posted 29 September 2015 - 01:32 PM

Thanks for the feedback thus far

#46 Hann Solo

    Member

  • PipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 276 posts

Posted 29 September 2015 - 01:35 PM

Tonnage is too high for 12 man groups. Comp teams will be running 12 Hellbringers and crushing everyone they come into contact with. If you want to even the playing field for smaller groups against larger groups, the larger groups should be further restricted in weight to give them a real disadvantage to make up for them having 12 coordinated pilots. Also the minimum for a 2 man group should be 70 so 2 friends can run lights together.

#47 Peiper

    Member

  • PipPipPipPipPipPipPipPip
  • The Dragoon
  • The Dragoon
  • 1,444 posts
  • Google+: Link
  • LocationA fog where no one notices the contrast of white on white

Posted 29 September 2015 - 01:40 PM

I would agree to this if single players could opt-in to the group queue. No better way to meet new players and pick up recruits. Otherwise, I really don't see what the big deal is about with the current matchmaker. Get to work on something more meaningful like COMMUNITY WARFARE. Like the mech rebalancing, you're spinning your wheels. We need more substance and no one will fret about the little stuff. They'll be more stuff to do, and less time for whiners.

#48 Big Tin Man

    Member

  • PipPipPipPipPipPipPipPip
  • Rage
  • Rage
  • 1,957 posts

Posted 29 September 2015 - 01:41 PM

Ok, playing with an excel sheet and I see this making some very ugly waves. Here's a brief story on how we got here and what to do about it:

The old old problem: no restrictions on group composition. 12 D-DC's? Sure. 12 Embers back when machine guns were good? Yup. OH GOD HERE COMES CLAN WAVE 1. Everything is bad and broken, clans have a huge advantage with the holy trinity.

The old old solution: 3/3/3/3 kept big groups from dropping too many heavies/assaults, and forced big groups to take a variety of mechs.

The old problem: MM tried to make each side 3/3/3/3, which lead to delays and long wait times when combined with regional servers, game mode choice, psr restrictions, etc.

The new solution: ditch 3/3/3/3 and rely only on tonnage limits to match groups together.

The problem with the new solution: average weight for groups is too high if you base max tonnage on a mixed group of mechs (i.e. 3/3/3/3 = 795). This average allows grossly OP sets of mechs to drop as it meets the tonnage requirements (i.e. all lighter heavies). This makes for very boring gameplay, all OP meta builds, is against lore, etc etc. Lots of bad things. Mawai wrote it up really well. 9 TBR's + 3 ACH = ggclose. The weights of mechs do not allow for any sort of balance in this manner.

The main issue with only running tonnage restrictions, is somehow capping the number of particular OP medium, heavy and assault mech's in a group.

A new [bad] solution: force max weights for groups larger than 6 to under 50 tons on average. This will incite rage among big groups, but could potentially achieve some sort of balance, while pissing off all of the heavy and assault pilots.

A hybrid idea: change 3/3/3/3 to 8/6/3/3 combined ALONGSIDE tonnage limits. Only use 8/6/3/3 as a restriction for the group to launch, and do not take 8/6/3/3 into consideration for the matchmaker. The tonnage restrictions takes it from there. Why 8/6/3/3? It allows groups some balance if they want to take a three 100 ton mechs, they may not be able to bring heavies, depending on the weight restrictions, so there will need to be more mediums and lights to make up for it.

Edited by Big Tin Man, 29 September 2015 - 01:43 PM.


#49 Cabbage Merchant

    Member

  • PipPipPip
  • Bad Company
  • Bad Company
  • 69 posts
  • LocationUnited States of America

Posted 29 September 2015 - 01:46 PM

View PostXX Sulla XX, on 29 September 2015 - 12:19 PM, said:

Question for those wanting no lower limit on small groups. The way Russ wants it to work is to not do weight matching in the matchmaker. Would you be OK with one side getting say 3 2 man groups of lights and the other side getting 3 2 man groups of heavies?


This post made me feel things.

#50 Kira Onime

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Dragoon
  • The Dragoon
  • 2,486 posts
  • Twitch: Link
  • LocationMontréal, Québec.

Posted 29 September 2015 - 01:48 PM

795 tonnage limit?

6 dire wolves - 6 arctic cheetahs
12 Hellbringers
12 Thunderbolts
12 ebon jaguars
6 Timber wolves - 6 stormcrows.

Totally working as intended.

Edited by Kira Onime, 29 September 2015 - 02:53 PM.


#51 XX Sulla XX

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • 3,094 posts

Posted 29 September 2015 - 01:48 PM

View PostHann Solo, on 29 September 2015 - 01:35 PM, said:

Tonnage is too high for 12 man groups. Comp teams will be running 12 Hellbringers and crushing everyone they come into contact with. If you want to even the playing field for smaller groups against larger groups, the larger groups should be further restricted in weight to give them a real disadvantage to make up for them having 12 coordinated pilots. Also the minimum for a 2 man group should be 70 so 2 friends can run lights together.
Although you will also have bad 12 man groups not crushing every one. I wounder if weight should be lowered according to over all PSR of the group on the client side?

#52 Trevelyas

    Member

  • PipPip
  • Ace Of Spades
  • 27 posts

Posted 29 September 2015 - 01:55 PM

Quote

In order to do this we need to allow the matchmaker to think about less options while making sure it is able to use PSR to match similarly skilled players for fights while also accounting for the power gap between many small groups vs one larger one.


Russ and friends: please, PLEASE read the following posts prior to making any more changes to the match-making algorithm:

https://www.reddit.c...for_psr_system/

https://www.reddit.c...h_class/cveeajo

Until PSR calculation is changed from an integral into a distribution (by making the system zero-sum) the match-maker cannot use PSR as a metric to reliably match players by similar skill, nor can there be a clearly defined mathematical approach to figuring out what the proper "PSR boosts" that should be applied to groups of size N should be.

To my pal Paul specifically:
If you still have players' latest Elo values archived from before the switch to PSR, I recommend generating a scatter plot of all players (who have played enough matches since January to "even out" their current PSR, maybe 4000+ matches) with Elo on the vertical axis and PSR on the horizontal and examine it. Ask yourself, "is there a strong positive correlation? Is it roughly linear?"
My suspicion is it will have very poor correlation, with players who have PSRs putting them in tier 1 and 2 having wildly different archived Elos (points on the right side of the graph heavily scattered vertically).

#53 Grifthin

    Member

  • PipPipPip
  • Bridesmaid
  • Bridesmaid
  • 98 posts

Posted 29 September 2015 - 01:55 PM

Please consider lowering the Tonnage numbers for groups. As the current proposed numbers stand every single person in your group can take a Timber/Hellbringer/Ebon Jaguar. Please consider lowering it a bit.

Medium mechs should be the most common sight on the battlefield, not 7-8+ heavies every single round.

#54 Kin3ticX

    Member

  • PipPipPipPipPipPipPipPipPip
  • The People's Hero
  • The People
  • 2,926 posts
  • LocationSalt Mines of Puglandia

Posted 29 September 2015 - 02:04 PM

It needs a way to prevent 12 Hellbringers, 12 Thunderwubs, or 6 Dires and 6 Cheetahs.

3/3/3/3 system does a better job at this because its flaws are less severe.

One weakness of 3/3/3/3 is that there is no advantage to not bring the heaviest mech of each class. So like 3 Dires, 3 Timbers, 3 Stormcrows, 3 Firestarters or Raven 2Xs ( now i guess it would be 3 ACHs if you have them).

I suggest keeping 3/3/3/3 and having 4 tonnage limits tailored to prevent bringing the biggest chassis in every slot. So for the assault class, a tonnage limit could be set to 275t .

TLDR: I think tonnage alone has bigger flaws than 3/3/3/3. 3/3/3/3 can be improved with tonnage limits on each class to prevent 3 dires or 3 timbers.

Edited by Kin3ticX, 29 September 2015 - 02:18 PM.


#55 Dimento Graven

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Guillotine
  • Guillotine
  • 6,208 posts

Posted 29 September 2015 - 02:04 PM

I don't see this working, especially considering that THEORETICALLY PSR is supposed to be a factor for MM as well.

Considering that tonnage by itself is actually only a moderate determination of a 'mech's "BATTLE VALUE", using ONLY tonnage as a primary consideration, and then overly restricting the available tonnage to teams so that it's not possible for them to utilize 3/3/3/3 in the ways they've been used to up to this point is surely going to be a miserable, titanic, epic failure on PGI's part.

My suggestion is to make the weight classes reflect a maximized 3/3/3/3 model:


2: 200tons max
3: 300tons max

4: 375tons max
5: 450tons max
6: 525tons max

7: 575tons max
8: 630tons max
9: 710tons max

10: 745tons max
11: 780tons max
12: 815tons max

Maximum MM tonnage allowed per side: 815tons

At least this way you've got something more reasonable to what is going on now while getting teams used to considering an 'optimized' group makeup within those limits, THEN you can bust your asses into implementing an actual BV system and swap us over to that, where a 12 'mech drop can have a maximum BV of, let's say just PURELY FOR EXAMPLE: 10,500.

Seriously, you PGI guys while creating a fairly decent game are coming up with ideas that seem to indicate you have no significant experience with how the actual gaming populace plays this game or their expectations on how it should function.

I figure this will last about as long as the abortive 'map voting' you had a few years back...

#56 Elbrun

    Member

  • PipPip
  • The 1 Percent
  • The 1 Percent
  • 28 posts

Posted 29 September 2015 - 02:07 PM

As long as the PSR has win/loss as an evaluation criterion it will be flawed, remove that and weight PSR on shots fired/shots hit, damage received, ecm/tag/narc/bap useage, and other actual SKILL relevant things and it will far less broken. Until then adjusting drop weight limits on queues (public or CW) will still mean very little, a bad group is still a bad group even if the magic PSR says otherwise.

#57 ALKALIN3

    Member

  • PipPipPipPipPipPip
  • Caladbolg
  • Caladbolg
  • 246 posts

Posted 29 September 2015 - 02:10 PM

The minimum weight for a 2 man is a bit steep. It doesn't allow 2 players to play lights together which a lot of folks like to do.

Edited by Aodh, 29 September 2015 - 02:10 PM.


#58 Big Tin Man

    Member

  • PipPipPipPipPipPipPipPip
  • Rage
  • Rage
  • 1,957 posts

Posted 29 September 2015 - 02:11 PM

View PostGrifthin, on 29 September 2015 - 01:55 PM, said:

Please consider lowering the Tonnage numbers for groups. As the current proposed numbers stand every single person in your group can take a Timber/Hellbringer/Ebon Jaguar. Please consider lowering it a bit.

Medium mechs should be the most common sight on the battlefield, not 7-8+ heavies every single round.


The problem is, you can't effectively lower the average group weight enough to change this without breaking balance between weight classes.

@ 60 ton average you can run 2xACH, 5xSCR, 5xTBR, or as far as 4xACH, 1xSCR, 7xTBR. Broken and OP meta.
@ 50 ton average you can still run 7xACH, 1xSCR, 4xTBR, or 3xACH, 9xSCR but it is next to impossible to run two or three assaults on a 12 man with a restriction this tight. Still an OP meta out there, while choking out any other chance to mix up a group.

Edited by Big Tin Man, 29 September 2015 - 02:17 PM.


#59 XX Sulla XX

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • 3,094 posts

Posted 29 September 2015 - 02:13 PM

This is something I have thought about before. The real problem in group play is not so much group size but quality of large groups. A 12 man of T3 pilots is not the same as a 12 man from a top comp team. And its the big drops from the best pilots the is the hardest for the system to match.

So the idea is to handicap group tonnage beyond the proposed system and based on group average PSR. So if ever one in your 12 man is maxed T1 you get a much light drop than a 12 man of T3 players. The weight is set where it becomes a challenge.

Here is how it would work.

-A 12 man of say EMP or SJR forms up. It shows them a lower total tonnage cap than say a 12 man or T3 players. The handicap is large enough to make it a challenge for them unlike now or with the current tonnage plans.

-As the group size of top players gets smaller their tonnage handicap gets smaller.

Answers to concerns.

-Is this a burdon on top players yes. But winning 98% of your matches while in a large group is not much fun for you or for the people getting rolled.

-Would it be harder for top players to practice together. Yes but how much is rolling every one in large groups really helping now?

-Would it just encourage top players to play in small groups yes and I have no problem with that.

#60 Russ Bullock

    President

  • Developer
  • Developer
  • 909 posts

Posted 29 September 2015 - 02:14 PM

Please focus less on how you disagree with the place holder tonnage amounts we listed and focus more on suggesting different ones.

Also I understand the point about keeping 3/3/3/3 in place WITH the new tonnage limits but when you consider that like 80% of groups are groups of 2 and 3 those will not be effected by the 3/3/3/3 aspect anyhow.

Better to work on the tonnage ranges to allow or disallow certain combinations.

So far I have heard good points to allow lower tonnage limits in smaller groups, and potentially a lower limit for the bigger groups.

Would like to see more suggestion on actual tonnage ranges.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users