

Russ Claims To Be Working On Doing Something About The Big Merc Units.
#141
Posted 03 January 2016 - 06:45 AM
I doubt that reducing the amount of loyalty points would change the current behavior. So either the rank is restricted to level 10 (which means a single Mech Bay) or you reduce the rewards like with Tukayyid: if you get a reward once, you can't get it a second time, excluding cbill rewards of course.
#142
Posted 03 January 2016 - 07:12 AM
JaidenHaze, on 03 January 2016 - 06:45 AM, said:
I doubt that reducing the amount of loyalty points would change the current behavior. So either the rank is restricted to level 10 (which means a single Mech Bay) or you reduce the rewards like with Tukayyid: if you get a reward once, you can't get it a second time, excluding cbill rewards of course.
Been saying that since Phase 1 and the plague of locust effect of "unlocking achievements" began.
You want loyalty? incentivize loyalty.
You want to maintain relevance, do not make it a 'single unlock and done' formula, make a currency that is continually spent and replenished.
Give the lowest personal risk but least freedom to loyalists.
Give more freedom and better pay to mercenaries, but also higher risk.
The highest risk and highest rewards go to freelancers.
Do this and you solve so much.
Edited by Kjudoon, 03 January 2016 - 07:14 AM.
#143
Posted 03 January 2016 - 07:30 AM
Long-term prospects could include special house variants of mechs, camo, decals and mech bays(with a cooldown, as mech bays represent the minimum investement into this game).
I don't think allowing hopping factions is a good thing. Mercs(units) and Freelancers(Pugs) should be "used" to balance faction numbers but as soon as you pledge for a house you're stuck with your decision.
Edited by k05h3lk1n, 03 January 2016 - 07:54 AM.
#144
Posted 03 January 2016 - 10:30 AM
The idea that you can fix Community Warfare is ridiculous because the only way to fix this game is for PGI to stop pretending they can 'be faithful' to the tabletop game.
This is not the tabletop game. If this was the tabletop game we'd have infantry, tanks, and air support that would shred the idiot balls as they waddle around the map. We would have missions that forced opponents to actually do something other than every game just being 'who can kill the most with an irrelevant side objective'.
#145
Posted 03 January 2016 - 12:09 PM
Edited by Gremlich Johns, 03 January 2016 - 12:11 PM.
#146
Posted 03 January 2016 - 01:03 PM
Jack Booted Thug, on 03 January 2016 - 02:52 AM, said:
The way you worded that got me thinking, what would stop higher level players dropping into a solo CW que (dropping their unit tags or using alt accounts) and wrecking? 12 vs 12 of mostly solo minded players, a couple good players on one side will just use their bad teammates as meat shields and still farm pulling 3-4K damage matches regularly.
The current PSR system in the solo que at least somewhat helps prevent new or weaker players from facing higher level players, not that it's a perfect system.
And what's to stop sync dropping in the above example?
People asking for a pug que in CW aren't seeing all the angles, and it seems difficult to convince them it's really not that hard to play as a team.
The better players are going to keep winning in any situation. They are better because they play with and against better players, they team up and pool skills and from that get better and do it faster. Pro sports teams are pro because they have a good coach, good training and play with/against other people. Not because they played pickup games at the local high school for longer.
Nothing done to change the size of units will in any way impact players in teams playing as teams stomping players who play like pugs.
What would help is two things -
1. Drastically reduce faction switching times. A lot of ways to do this but you need to make committing to a faction a longer engagement. This allows other factions (and units) to adjust accordingly. That's how humanity has always worked; big groups form, their neighbors all coordinate to deal with them.
2. Reward big group v big group play. Give a bonus % to cbill and LP payout for playing against (and especially winning against) other big/successful units. Playing against SWOL members should pay more than playing against pugs. Beating 228 should pay a hell of a lot more than beating skittles.
It's not even a hard mechanic. You set thresholds for the payout and you normalize a units win % with its size, so big units that win reasonably often are comparable to small units that win a lot. You calculate how much of a bonus you want, for example you get a 10% bonus, win or lose, to play against a -MS- 12man. If it's 6 MS on the other team you get 5%. Make sense? Your faction tag carries a bonus multiplier, you're 1/12th of whatever a 12man of your unit is worth as a multiplier to your OPPONENTS payout.
So 228 gets a 10% bonus for all the matches they play against MS. On a win that can be like 100k or more. At that point you have a strong incentive to play AGAINST the big and/or successful units. Winning is its own reward; you don't get a reward for being big. Indeed, a group like SWOL suffers because by having so many members it pays people to play against it while they have fewer big groups they can play and win against.
THAT would fix this issue way better than trying to 'cap' or limit big units. Splitting the queues would be just turning CW into new pug/group queue game modes.
#147
Posted 03 January 2016 - 01:29 PM
MischiefSC, on 03 January 2016 - 01:03 PM, said:
2. Reward big group v big group play. Give a bonus % to cbill and LP payout for playing against (and especially winning against) other big/successful units. Playing against SWOL members should pay more than playing against pugs. Beating 228 should pay a hell of a lot more than beating skittles.
It's not even a hard mechanic. You set thresholds for the payout and you normalize a units win % with its size, so big units that win reasonably often are comparable to small units that win a lot. You calculate how much of a bonus you want, for example you get a 10% bonus, win or lose, to play against a -MS- 12man. If it's 6 MS on the other team you get 5%. Make sense? Your faction tag carries a bonus multiplier, you're 1/12th of whatever a 12man of your unit is worth as a multiplier to your OPPONENTS payout.
So 228 gets a 10% bonus for all the matches they play against MS. On a win that can be like 100k or more. At that point you have a strong incentive to play AGAINST the big and/or successful units. Winning is its own reward; you don't get a reward for being big. Indeed, a group like SWOL suffers because by having so many members it pays people to play against it while they have fewer big groups they can play and win against.
THAT would fix this issue way better than trying to 'cap' or limit big units. Splitting the queues would be just turning CW into new pug/group queue game modes.
Hmmm, you might be on to something. Basing payouts on the quality of opponent could be an option. Not that PSR is a perfect indicator, but face a team of lower tier / or new players, lower rewards. Play a team of quality opponents, higher payout for every kill even on a loss.
Might need a seperate CW PSR that is calculated in a different manner though, not sure if using solo que PSR would work.
In short, a bounty system, which I believe has been brought up before. I would really like a bounty system.
#148
Posted 03 January 2016 - 01:31 PM
Just having a unit's win% should be enough to give them a multiplier, as there are small unit's like KCom for example who have a higher win% than most big unit's and almost play as many games as their entire unit. A KCom 12 man should be more rewarding to kill than a SWOL 12 man because of their higher win%.
Edited by l)arklight, 03 January 2016 - 01:34 PM.
#149
Posted 03 January 2016 - 02:13 PM
l)arklight, on 03 January 2016 - 01:31 PM, said:
Just having a unit's win% should be enough to give them a multiplier, as there are small unit's like KCom for example who have a higher win% than most big unit's and almost play as many games as their entire unit. A KCom 12 man should be more rewarding to kill than a SWOL 12 man because of their higher win%.
I agree, a high value target (bounty) system would help in games where you just have a lance of high tier CW players with skittles facing the same or another team of pugs. Something to indicate to players on the battle screen or drop screen so they know who the high value targets are. Could help skittles in a target rich environment focus down the biggest threats and reap the most rewards, and maybe give them a better chance of winning.
#150
Posted 03 January 2016 - 02:55 PM
#151
Posted 03 January 2016 - 05:34 PM
Jenovah, on 03 January 2016 - 04:17 AM, said:
And yet no one stumbles upon the answer right in front of your faces---
Why does everyone believe that successful merc groups must be ripped down? Because you can't form one, so you have to destroy that which you cannot create or lead?
Or maybe you create your own group under a faction and step up the competition.....
But I guess salting all over the forums is easier.
The problem isn't the size of the merc units, but how much influence they bring to a faction when they switch.
One example I might present is:
- MS and 228 each were in Falcon/Wolf. Each of those factions grew greatly by their influence.
- MS and 228 change sides to Steiner and FRR. Now, Falcon and Wolf are in decline, to the point that Clan Wolf have almost not planets left under their control.
- Back in the day, Marik and Kurita had an alliance together, and a wormhole appeared in Combine space. Together, we worked on a project called "Purple Dragon Island", with the intent to bring another faction's efforts to the Clan front for attack lanes. Well, MS joined Marik, and attacked the Combine, expanding Marik control within our "ally's" territory, which was not what we wanted. Then, after angering the Combine, MS went back to Ghost Bear, and attacked the Combine while the Combine (now angered at us for attacking them) went after Marik.
Merc groups that faction jump so much can really influence the map. Sometimes in disruptive, or at least interesting, ways. Depending upon perspective
I'm not advocating on unit size restrictions, as I feel that would be bad for the game. However, the simple already proposed voting system for Loyalists to control attack lanes solves large influence Merc groups, forcing them to attack only the indicated targets of their employers. It provides mercs with their continued play abilities, while giving Loyalists a reason and some manner of control over "their hired mercs". Win win?
#152
Posted 03 January 2016 - 06:15 PM
Quote
That was "Roadbeer Island".
Good times... weird times. Also should serve as a warning to the Antares Scorpions.
Edited by Kjudoon, 03 January 2016 - 06:16 PM.
#154
Posted 03 January 2016 - 07:51 PM
Jack Booted Thug, on 03 January 2016 - 01:29 PM, said:
Hmmm, you might be on to something. Basing payouts on the quality of opponent could be an option. Not that PSR is a perfect indicator, but face a team of lower tier / or new players, lower rewards. Play a team of quality opponents, higher payout for every kill even on a loss.
Might need a seperate CW PSR that is calculated in a different manner though, not sure if using solo que PSR would work.
In short, a bounty system, which I believe has been brought up before. I would really like a bounty system.
Exactly. It doesn't have to be hard - you want it very simple. You just need some % metric based on win/loss AS A UNIT AVERAGE and TOTAL PLAYER/MATCHES.
So you get big units that do a lot of drops along with units that win a lot of matches (even if not as big) offering the highest bounties to play against.
This would drive multiple small units to play *against* big units and pay out best to the best units/biggest units to play against each other.
You don't want something as complex as bounties, that's going to get exploited. You just need to pay big/good units more for playing each other. Even pugs, everyone. If you play against a big/good unit or even just members on the other team of a big/good unit, you make more. Win or lose. Otherwise you get pugs and solos shifting (as they did with the Clans) to play on the same team as the big/good units because just dropping with them paid more than dropping against them.
Make sense? That would do it. Almost un-exploitable and it motivates all the people you want to motivate in the right way to do the right thing.
#155
Posted 04 January 2016 - 12:49 AM
make it so Mercs may not decide how a faction will attack. They may defend to their hearts content and attack planets given the okay by faction loyalists (to represent that they are 'working' for that faction) and encourage pseudo loyalty by steadily increased rewards the longer that they stay under that contract like a %gxp or cbill boost that resets if they leave.
There big merc units don't have a big sway on the attack path decisions, everyone gets to keep the rewards we've kown and love, and who would have thunk it but it makes sense contextually.
#157
Posted 04 January 2016 - 01:25 AM
Tesunie, on 03 January 2016 - 05:34 PM, said:
Merc groups that faction jump so much can really influence the map. Sometimes in disruptive, or at least interesting, ways. Depending upon perspective
I'm not advocating on unit size restrictions, as I feel that would be bad for the game. However, the simple already proposed voting system for Loyalists to control attack lanes solves large influence Merc groups, forcing them to attack only the indicated targets of their employers. It provides mercs with their continued play abilities, while giving Loyalists a reason and some manner of control over "their hired mercs". Win win?
This could work, however, "faction jumping" is simply going where the cbills are, no?
Want to ensure certain groups are working for you and not against you? Up the price.
Getting roflstomped by mercs from your enemy? Entice them with better payouts.
Not getting into the "lore" perspective, but to treat the merc player base as 2nd rate gamers and trying to "tax" them into non existence with R&R, additional "costs" to play while also controlling their every move with no recourse seems a little....... hmmmm.... let's not get into politics, eh?
But if the blow were softened, or offset, by not restricting the faction moving (assuming that either the duration or objectives of a contract were met) and the merc player base could go where the cbills are and unit coffers were a thing, this could definitely work.
As soon as a merc unit completed their assigned objective or duration on a contract or both, we could then renegotiate the terms which would be lucrative at the least.
#158
Posted 04 January 2016 - 01:44 AM
With this restriction the larger groups would have to split up and fight for different factions as they could not sustain themselves with one faction.
The loyalist would obviously compete to attract the best merc units, but also could pitch lower rewards to smaller units.
#159
Posted 04 January 2016 - 02:15 AM
NextGame, on 04 January 2016 - 01:10 AM, said:
What kind of warning? That Marik are largely a bunch of pretentious roleplayers?
LOL... So salty.
You wouldn't take the warning even if explained.
The result of the Roadbeer Island experiment was also due to players who believed they controlled it all.
But hey, you know better. Do what you're going to.
Edited by Kjudoon, 04 January 2016 - 02:15 AM.
#160
Posted 04 January 2016 - 03:07 AM
Kjudoon, on 04 January 2016 - 02:15 AM, said:
You wouldn't take the warning even if explained.
The result of the Roadbeer Island experiment was also due to players who believed they controlled it all.
But hey, you know better. Do what you're going to.
Salty about what? The only salt I'm seeing here are a bunch of people whining because others don't want to play by their rules and instead want to move around the factions.
i.e. generally pretentious roleplayers who want control over others game experience.
Edited by NextGame, 04 January 2016 - 03:08 AM.
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users