29 replies to this topic

[Prezimonto]

Asmudius Heng, on 27 January 2014 - 08:52 PM, said:

100% agree with roland.

If the awesome was worth 'effectively' 70 tons or less because it was being used so little and the highlander was worth 100 tons, you might see people reconsidering the mechs that have the most advantageous set ups.

If you were an ace commando pilot say and your mech could be brought for the effective tonnage cost of 15 tons because no one else played it you could save your team lots of tons for other mechs and if you do well you are bringing a lot more to the table than the stated 15 tons.

If PGI changes something balance wise and people drop using a high end meta mech then the less usage will be reflected very quickly and it will become cheaper to field.

This is the best way to have a manageable BV system that can change as the game evolves IMO

And people like me that see a {Scrap} mech as a challenge (while still recognizing it as a bad mech) would have a field day and new challenges!

Mechs with JJs and ECM will then cost more most likely reflecting how powerful that equipment is also

You could also easily incorporate other factors into a market based BV system once community warfare come out: each mech, upgrade, weapon, and equipment could receive a factor based on availability. So not only will the meta on pugs affect your BV, but your success on the battle field... Want lower BV Victors... take that planet that's got a factory for them.

Other than up front coding cost there is no downside to a market based BV system over tonnage limits on a global scale. On a personal scale there is the issue that the "best" builds will be expensive and will increase in BV until they're unsustainable... meaning you'll have to adapt. That's going to be highly unpopular with a certain crowd.

[MischiefSC]

BattleValue wouldn't work. For one thing it doesn't translate to MW:O - LB10X is NOT superior to AC10, ERPPC not vastly better than PPC, etc. XLs even not that much better than standard.

Tonnage also isn't a direct match. Wispy in a Jenner is more dangerous than Wispy in an Awesome. Even in similar tonnages a Jag is more than 5 tons more dangerous than a Dragon, far more so in some peoples hands.

The better option is to track Elo by weapon and chassis. Your Elo gets adjusted up or down based on your specific performance in mechs with that loadout. You a rockstar with PPCs and JJs? Your Elo is higher when you're carrying them. You a killing machine in just about anything you put in your Jag DD? It shows and dropping in your DD means you're weighted higher in Elo for the Matchmaker.

Now for that extra kick you provide a synergy bump for premades. You win more when you drop with player A, B and C? When you guys are together your respective Elos are higher. This better accounts for Elo performance in premades without having to have a separate premade Elo.

Then you combine that with weight limits.

BOOM. Instead of general values for gear you've got the same thing as trying to assign a value to gear and weapons and combine it with the persons Elo to get a new value you generate a unique value for each player based on their performance with weapons and chassis and use it to make a unique sort of BV for that player and the gear they're dropping with in that match.

Some players suck with assaults, throwing them in with people way better than them is bad for everyone. Some people are the raised middle finger of death in a light. Throwing them in with less skilled folks because they're in a Jenner is cruel as well.

Gentlemen, we can rebuild it. We have the technology. We can make it better than it was. Better...more balanced...faster.

*queue music*

[Goose]

OR: We could could track Elo per 'Mech.

Give it a few days, things will sort out …

[ShinVector]

Goose, on 27 January 2014 - 10:56 PM, said:

OR: We could could track Elo per 'Mech.

Give it a few days, things will sort out …

Don't think this will solve the weight imbalance issue with pre-made groups though.

[Roland]

MischiefSC, on 27 January 2014 - 10:49 PM, said:

BattleValue wouldn't work. For one thing it doesn't translate to MW:O - LB10X is NOT superior to AC10, ERPPC not vastly better than PPC, etc. XLs even not that much better than standard.

Tonnage also isn't a direct match. Wispy in a Jenner is more dangerous than Wispy in an Awesome. Even in similar tonnages a Jag is more than 5 tons more dangerous than a Dragon, far more so in some peoples hands.

My battlevalue system actually WOULD work, as it would specifically account for those things you mention there.

Of course it wouldn't specifically account for Wispsy, but it would make his Jenner fairly high value since it's one of the strongest light configs currently.

[3rdworld]

MischiefSC, on 27 January 2014 - 10:49 PM, said:

BattleValue wouldn't work. For one thing it doesn't translate to MW:O - LB10X is NOT superior to AC10, ERPPC not vastly better than PPC, etc. XLs even not that much better than standard.

Tonnage also isn't a direct match. Wispy in a Jenner is more dangerous than Wispy in an Awesome. Even in similar tonnages a Jag is more than 5 tons more dangerous than a Dragon, far more so in some peoples hands.

The better option is to track Elo by weapon and chassis. Your Elo gets adjusted up or down based on your specific performance in mechs with that loadout. You a rockstar with PPCs and JJs? Your Elo is higher when you're carrying them. You a killing machine in just about anything you put in your Jag DD? It shows and dropping in your DD means you're weighted higher in Elo for the Matchmaker.

Now for that extra kick you provide a synergy bump for premades. You win more when you drop with player A, B and C? When you guys are together your respective Elos are higher. This better accounts for Elo performance in premades without having to have a separate premade Elo.

Then you combine that with weight limits.

BOOM. Instead of general values for gear you've got the same thing as trying to assign a value to gear and weapons and combine it with the persons Elo to get a new value you generate a unique value for each player based on their performance with weapons and chassis and use it to make a unique sort of BV for that player and the gear they're dropping with in that match.

Some players suck with assaults, throwing them in with people way better than them is bad for everyone. Some people are the raised middle finger of death in a light. Throwing them in with less skilled folks because they're in a Jenner is cruel as well.

Gentlemen, we can rebuild it. We have the technology. We can make it better than it was. Better...more balanced...faster.

*queue music*

I wouldn't use a per item ELO if it were me. I would simply modify each chassis by an "effectiveness modifier". Which would be used to make bad chassis very "cheap" to the matchmaker, and good chassis more expensive.

This would then be used in combination with a persons ELO to come up with the total value for that person in that mech to the matchmaker.

In your example, wispsy in a Jenner would for sake of argument have a value of 80. Wispsy in an awesome may have a value of 60. Because one of the major issues with the matchmaker you touched on. Players in crappy mechs are matched against equally skilled players in great mechs.

Edited by 3rdworld, 28 January 2014 - 06:32 AM.

[BlackIronTarkus]

Roland, on 27 January 2014 - 09:09 PM, said:

The system would autobalance... Not only would the results be better than anything manual balancing could accomplish, but it would be WAY LESS WORK on PGI's part.

It'll never happen though.

I know right, its too much of a good idea for PGI to use it. >

ELO is a mess proven to be ineffective, and tonnage limit doesn't solve much other than having less assaults on the battlefield. I mean it would be great to have a lot less assaults on the battlefield, but people would rage at other people for wasting tonnage on a mech such as the AWESOME because its {Scrap} even compared to an CATAPHRACT or a JAGERMECH which weight less.

Organised group with the only limitation of tonnage limit would simply use the most effective mechs and make the battlefield even more generic.

With tonnage limit dont dare waste any tonnage on anything other than;
Lights: Jenner, spider, ECM raven
medium: Hunch, shadowhawk, centurion
Heavies: Jaggermech, cataphract
Assault: Stalker, victor, highlander, atlas.

Thats it. Bring anything else and you are hampering your team.

But instead with battle value, a cataphract 3D would cost a lot more than an AWESOME which in turn make the AWESOME desirable. Its a poor chassis, but it doesn't cost much battle value which mean someone else can bring a higher battle value mech to the battlefield. For another exemple: A highlander with jumpjets, PPC and AC5 would cost more battlevalue to the team than an ATLAS without an ECM.

Indeed Battlevalue MUST be displayed on everything, on weapons, on modules, on variants with for a total that is also displayed for the player to make meaningful choices.

[Prezimonto]

MischiefSC, on 27 January 2014 - 10:49 PM, said:

BattleValue wouldn't work. For one thing it doesn't translate to MW:O - LB10X is NOT superior to AC10, ERPPC not vastly better than PPC, etc. XLs even not that much better than standard.

Which is why you seed the a market based BV system over a month of usage statistics first, then let it run free. Who cares about TT values anyway? They're totally borked compared to MWO. Market based BV FTW!

[ReXspec]

Mystere, on 27 January 2014 - 07:40 PM, said:

Well I surely hope this does not happen. Just imagine all the balancing arguments that would create. They would be endless.

I think this is one of those times in which the KISS principle might be relevant.

And yet TT Battle Value works just fine. It's not perfect, but it's a helluva lot better then the system that is in place right now.

Prezimonto, on 28 January 2014 - 04:49 PM, said:

Which is why you seed the a market based BV system over a month of usage statistics first, then let it run free. Who cares about TT values anyway? They're totally borked compared to MWO. Market based BV FTW!

No they're not... there is a reason the core TT BV system has been used for for nearly 25 years. Now, what PGI has in its place is truly borked.

You may be on to something with seeding the BV's in the market, though...

[Lukoi Banacek]

ReXspec, on 27 January 2014 - 06:35 PM, said:

Current Battle Value is simply based on Weight Class and ELO. If they based Battle Value on the TT system, there would be a lot less lob-sided matches.

Pure BV does not take player skill into acct at all. Are you sure you want to base the MM on arbitrary point scales that are based on TT values for the weapons, capabilities and quirks of a mech (like torso twist, speed of twist before engine modification, hitbox desing and distribution) rather than a BV system based on the stats of all of those items as they are actually used in game?

BV is higher for LPL than LL.......would you honestly consider LPLs a better weapon than LL right now?

Unless there is something I missed your option is too simplistic and based on the wrong arbitrary mathematical values.

Edited by Lukoi, 29 January 2014 - 06:02 PM.