64 replies to this topic

[STEF_]

Totally agree.
That's why BT works well.
The alternative is a continuos nerf/buff without end, completely disorienting, and without common sense (i.e. ridiculous C-erLL duration, ridiculous victor quirks, and so on).
The alternative is a castrating ELO match-maker that, instead of building teams made of similar players, forces you to play with disaster noobs team to tear down your own elo, if it was gone too high.....

And it's quite hilarous that some players don't even know what BV is
......but... does MWO should be BT related, or what?

Edited by Stefka Kerensky, 30 August 2014 - 03:53 AM.

[Vassago Rain]

John1352, on 30 August 2014 - 03:05 AM, said:

The game mechanics of MWO generally make a good light worth ~75% as much as an assault or heavy IMO. Many assaults are better with standard engines than they are with XL ones. It would become hard to actually put a performance based value on mechs.

Not really. I can do it right now.

Spiders: bad.
Badders: bad.
Commandos: bad.
Locusts: THE WORST MECH IN THE GAME.
Jenner: good.
Firestarter: best light.
Raven: one good variant tier.
Cutefox: okay.

And from here we can work out a points system, so that if you bring a stock locust the enemy team doesn't get a tech 2 mastered firestarter.

Edited by Vassago Rain, 30 August 2014 - 03:52 AM.

[Egomane]

JudgeDeathCZ, on 30 August 2014 - 03:29 AM, said:

yea and PGI can not make their own numbers and statistics based on a milions of drops to implement their own Battle Value numbers right?

And you know how much is involved in finding a suitable solution? There is a reason that the tabletop, is now going into the 5th round of trying to attempt to put a fitting number on an unchanging mech. In an ever changing environment, where the loadout of mechs can be changed freely, that is a nightmare.

Aresye, on 30 August 2014 - 02:19 AM, said:

JoJoxy, on 30 August 2014 - 02:02 AM, said:

Why not have real time (or computed daily or whatever) Battlevalue?
Take the average kills, damage, win/loss, etc. for each chassis and apply the tournament scoring system to it.

Voilà, now you have a normalized score based on which you can determine the effectiveness of any chassis.

For example, we know the Clan Heavy bracket is most certainly, without a doubt, Timberwolves, with an average score of 2654. We can also fairly assume the IS Assault bracket is mostly Victors, with an average score of 2374.

Bad idea! Basing it on the value on the chassis alone is stopping at least two steps before we even get to a balance discussion.

[Vassago Rain]

Egomane, on 30 August 2014 - 04:03 AM, said:

And you know how much is involved in finding a suitable solution? There is a reason that the tabletop, is now going into the 5th round of trying to attempt to put a fitting number on an unchanging mech. In an ever changing environment, where the loadout of mechs can be changed freely, that is a nightmare.


Bad idea! Basing it on the value on the chassis alone is stopping at least two steps before we even get to a balance discussion.

Points costs aren't rocket science.

Even GW figured it out once they dropped 2nd edition. Are you saying that CBT's current owners, as well as PGI, are somehow less competent than GW?

It's super easy to put a high cost on good robots, and a low cost on bad robots. Then you add less or more points depending on the equipment. For instance, an XL in a spider is worth more, percentage wise, than an XL in an atlas, but even the XL atlas gets to take a billion tons more gear, so the XL would still add extra points.

But we're well past the point where they'll make any changes like this, so it's moot.

Edit: I'll humor you and point out the most obvious flaw with the 'points are bad'-argument.

Unlike warhammer or CBT, where updates are years apart, MWO gets a patch every two weeks. It'd be super easy to implement a points system, and revise it as things change in the game.

Edited by Vassago Rain, 30 August 2014 - 04:08 AM.

[GalaxyBluestar]

2011's calling it want's it's thread back.

http://mwomercs.com/...r-the-mmo-game/

anyways here's one made in the suggestion area, very interesting and possible. likey likey now.

http://mwomercs.com/...levalue-system/

it's also waaaaaaaaaayyyyy better for:

balancing the noobs and their champion no module no master mechs against fully kitted veterans

still if you're happy with 3 timbers vs your odd atlas and awesome combos solely on elo and 3/3/3/3 whoops queue is jammed relax 3/3/3/3 pffftcrap.

Edited by GalaxyBluestar, 30 August 2014 - 04:16 AM.

[Procyon Alpha]

So why again are you against a system that other games have used quite successfully? Almost any game that has some kind of balancing system that uses numbers. You logic is flawed, as to it is too hard to implement, that is horse hockey.

Sorry we can't all bury our heads in the sand, and pretend to tote the PGI line the a numerical system of balance is just too hard. Oh and don't tell me the clans are just different and it will all work out in 12 v 12, that is a mathematical impossibility.

GalaxyBluestar, on 30 August 2014 - 04:13 AM, said:

2011's calling it want's it's thread back.



http://mwomercs.com/...r-the-mmo-game/

anyways here's one made in the suggestion area, very interesting and possible. likey likey now.

http://mwomercs.com/...levalue-system/
Yeah I voted for that system a while back. Personally it is coming to a head with CW and the clan mechs. Something needs to be addressed soon.
it's also waaaaaaaaaayyyyy better for:

balancing the noobs and their champion no module no master mechs against fully kitted veterans

still if you're happy with 3 timbers vs your odd atlas and awesome combos solely on elo and 3/3/3/3 whoops queue is jammed relax 3/3/3/3 pffftcrap.

I totally agree, and I agreed then. But with the Clans and CW this needs to be addressed soon, the glue is becoming undone and we don't want it to fall apart at the seams.

Edited by Yeshua Kerensky, 30 August 2014 - 04:22 AM.

[JudgeDeathCZ]

Egomane, on 30 August 2014 - 04:03 AM, said:

And you know how much is involved in finding a suitable solution? There is a reason that the tabletop, is now going into the 5th round of trying to attempt to put a fitting number on an unchanging mech. In an ever changing environment, where the loadout of mechs can be changed freely, that is a nightmare.

Too bad we have 2 yrs of MM which obliviously do not work and nothing changed...

[Seeker Kirov]

Tonnage doesn't work as a balance mechanism, and by extension neither does 4x3. And I thought ELO was designed specifically for chess players. BV works pretty well in TT. Not perfect, but there is no perfect balance system. It certainly works better than what we currently have. However to implement it wouldn't you need to restructure the game to be much, much closer to the TT rules? Heat, damage, etc. - you would absolutely have to get rid of this absurd weapon convergence we have. TTK would go up, matches would take longer, people would get lower C-bill/XP yields per hour played. I don't think PGI would do it, though I would love to see it. It might finally get the weight of drops to be more in line with what the setting depicts - Medium and Heavy 'Mechs are the ubiquitous workhorses of the regiment while Lights and Assaults perform supporting roles.

[Appogee]

BV isn't going to fix the problem of being stuck in matches with incompetent teammates.

Yesterday I was placed in a match where a team mate had an Atlas D-DC (a high BV Mech) who was shooting his AC20 at targets 800m out and then didn't even move an inch for the entire 30 seconds it took for the enemy team's incoming fire to put him down.

Elo is an attempt to give players a Derp Value. However, while the MatchMaker continues to try to average low Elo pilots with high Elo pilots we may as well be dropping randomly.

Edited by Appogee, 30 August 2014 - 05:24 AM.

[Hoax415]

Straight tonnage would be better than what we have now.

In fact tonnage with a positive modifier if the mech is Clan might be "good enough".

The problem with something that simple would be that either the summoner would cost so much that you hurt your team (say the tonnage boost for clan is +20%-30%) or the Mad Cat would still be amazing (say the tonnage boost for clan is +5%-10%).

Could you make a better system? Yes.

But reality is that right now we have no system. You see 3x IS non DDC assaults versus a pair of DWF and its just a huge handicap because of RNG.

Which then leads you to sometimes feeling like "screw this guy piloting his Highlander or Awesome or even worse Dragon or Locust" that guy isn't trying hard enough to win. He's hurting the team. Which is poor and has been poor and with the clan mechs will get much much worse.

PGI pls. Wake the hell up.

[Wingbreaker]

You are bad at math.

BV is bad at math.

BV was created with the idea that bigger is better.

Battletech mathematically rules smaller in larger numbers are always better.

Fourteen Savannah Masters do not equal one Atlas.

But according to Battle Value, they do.

Edited by Wingbreaker, 30 August 2014 - 05:41 AM.

[Hoax415]

You may have thought that was a clever post but it really wasn't.

What we have now is terrible. When you have a terrible system you don't insist that replacing it with anything but perfection is a waste of time.

Also 14 Savannah Masters can be much more meaningful than an Atlas. It depends on how large the battlefield is and the win conditions / scenario. The Atlas can only be in one place and it can't get from place to place very quickly.

Imagine on Alpine conquest, would you rather have 4 Locusts or a DDC? The DDC is the wrong choice.

[M E X]

Egomane, on 30 August 2014 - 04:03 AM, said:

...
Bad idea! Basing it on the value on the chassis alone is stopping at least two steps before we even get to a balance discussion.

Actually GUNNERY and PILOTING skills are a HUGE factor in calcualting a BV !

That means IF they would use a BV system instead of a ELO ( on player skill only ) based system the matchmaker should be able to make better pairings.

GUNNERY could be calculated from the accuracy of your shots, if you miss a lot it will be a bad ( HIGH ) value and if you hit with nearly every shot is should go down to 1( or even lower when you use real values instead of integer )

PILOTING could be calculated from your ability to avoid hits with torso twists, going into cover, ....

In Classic BattleTech the BV is usually calculated for a average pilot with a Gunnery/Piloting of 3/3 or 4/4 which is then modified by the Gunnery and Piloting skill of the pilot of the mech, which usually is in the range from 1 ( elite ) up to 8 ( n00b )

Please read http://www.sarna.net/wiki/Battle_Value , http://www.sarna.net/wiki/CBT_Tables , ... , and try http://www.masteruni...nfo/Tools/Skill yourself, for information about HOW BV works ....

Wingbreaker, on 30 August 2014 - 05:41 AM, said:

...

Fourteen Savannah Masters do not equal one Atlas.

But according to Battle Value, they do.

Actually you need MUCH LESS THAN 14 Savannah Masters for killing a single Atlas !
http://www.sarna.net...Savannah_Master
And if you would compare the TONNAGE, you would get 20 Savannah Masters for a Atlas ;P

Each Savannah Master is a small 5-ton hovercraft with a single medium Laser, moving at up to 216 kph.

Edited by M E X, 30 August 2014 - 06:03 AM.

[Wingbreaker]

Hoax415, on 30 August 2014 - 05:50 AM, said:

When you have a terrible system you don't insist that replacing it with anything but perfection is a waste of time.

When you have a terrible system you don't insist on replacing it with a WORSE system.

[Jonny Taco]

Egomane, on 29 August 2014 - 11:15 PM, said:

BV = BattleValue

A numerical system implemented to balance the tabletop game. It applies values to things like movement, excess heat, amount of ammunition and more. It's easy to manipulate and I therefore believe it's not a good choice for a mechwarrior game.

Assuming that all BV systems are equal is not really a very good argument against it... You've already highlighted some of the canon implementations shortcomings which is a step in the right direction towards fixing them... Furthermore, cannon BVs do not at all need to be mirrored 1 for 1 to BT... If BV is to be introduced, the values should be created completely independently from the table top... They also should not be balanced around usage which is one of the dumbest ways to balance anything as usage of something is not simply defined by using the best thing all the time... Some people have different tastes in speed, look, weapons, ect ect ect... BV (which btw is not specific to only BT) should be entirely based around a mechs capability, not it's popularity...

IF the dev team actually wants people to run a variety of load outs and mechs, there really is no other way to do it... Example being tbr vs smn... With the current system the smn is going to have to be buffed to relatively canon breaking levels in terms of hard points/internals.... OR!!! the MM could value a tbr more than a smn and retain canon loadouts...

There really seems to be allot of opinions based more around fear of change than anything constructive.

Edited by lartfor, 30 August 2014 - 06:10 AM.

[kailii]

Perfect Balance is an unachievable goal, but all we need is sufficient balance.

A BV based system is too complicated IMO, and basing it on in-game statistics will never work (just take 20 jocks and have them all drop in the same Mech X for 10 times going kamikaze - wrecks that Mech X's stats).

How about some baby steps instead, one after another...
It could look like this (just a suggestion to get the discussion started):

Phase 1:
3x3x3x3 is dropped in favour of 2x2x2x2. The 8 missing Mechs are added in pairs of the same weight range.
Takes minimal developing time - more than an hour and i'd fire the programmer.
(Doesn't help with balancing, i just thought it's a nice idea giving some variety and helps with the "50% heavies in queue" bottleneck)

Phase 2:
Start with 2x2x2x2 and add Mechs based on tonnage. Does not require 24 Mechs in total.
Balances a little bit, IS vs. Clan Mechs is not taken care of. Again, minimal developing time.

Phase 3:
Include Electronic Warfare. Count the numbers of ECM (maybe even TAG etc.) Mechs and try to balance them out.
(2 ECM Mechs in one Team vs. 3 in the other Team is no problem, 0 vs. 2 is)

Phase 4:
Keep the tonnage system, but also rate the Mechs.
Now this is tricky, how about basing it not on Mech chassis but weapons and tonnage?

1) Every weapon gets a base value, say 10 for a PPC and 8 for a LL.
2) A PPC in an Awesome causes more havoc (on average) than one in a Spider, so the Awesome PPC is valued 10 + 20% = 12 points. The added value can be based on Mech Type, tonnage or actual armour for example.
3) Further modify this value based on efficiency. This might include ELO, Ghost Heat, whatever.
4) Add a weapon modifier. C-ERLAS OP? Don't nerf the weapon (or only a bit), instead add to its value.
5) Add 137 more steps if necessary, but keep it simple to adapt to gameplay changes.

Edited by kailii, 30 August 2014 - 05:59 AM.

[Hoax415]

No you don't understand a system where a Kit Fox is the same for matchmaking as a Locust is clearly superior to anything else. Wingbreaker says so with zero evidence or argument or data to back him up so it must be true.

[Wingbreaker]

M E X, on 30 August 2014 - 05:50 AM, said:

[center]Actually you need MUCH LESS THAN 14 Savannah Masters for killing a single Atlas !
http://www.sarna.net...Savannah_Master
And if you would compare the TONNAGE, you would get 20 Savannah Masters for a Atlas ;P

Each Savannah Master is a small 5-ton hovercraft with a single medium Laser, moving at up to 216 kph.

Using the SL variant. It's just a mental exercise to show people how imbalanced the system actually is when they insist that it's the glorious savior of all mechwarrior.

[Jonny Taco]

kaili, i just wanted to highlight your point about dropping the game back to 8v8. 8v8 is easier for the match maker, more in line with the vast majority of maps, and imo provided a much better gameplay experience allowing exceptional (good or bad) pilots actions to have a much heavier weight on the outcome of the game...

Better yet! Some maps should be 4v4, some should be 8v8, some should be 12v12.

[CMDR Sunset Shimmer]

You know, the primary "Simple" fix to this is to actually make the matchmaker work like it should...

You know, matching players VS other players of similar "skill"

Now I quote Skill because I'm refering to ELO score more than anything...

I'm going to use a simple system. of 1-10.

If I'm say, an ELO of 5 to start with, I should be fighting in groups that consist of ELO's 4-6.

This puts everone on a "similar" skill curve, and if you take out more higher level pilots you advance in rank to the next bracket.

ELO is a **** system for this kind of game anyway, but given that it's what we have... it should be utilized to match similar skill levels, not tossing new players in with the top tier players to "Balance" the team.