77 replies to this topic

[Almond Brown]

GalaxyBluestar, on 09 June 2014 - 07:12 PM, said:

LOL people thinking the 2 year delayed CW isn't vaporware compared to SC battlevalues.

stuff

but the OP said himself and I quote.

Quote

Well.. It's not exactly battle value, but it is indeed based on a points system

So is it BV or not ffs?

[Sable]

I think the community has very a very narrow idea of battlevalue. If MWO did use battlevalue and calculated a custom mech with custom weapons, i'm sure there would be ways to game the system like there is in everything else online. 4x3 is a simple way to equalize the weight classes while not restricting what can be brought in. Some people have implied its a simplistic solution yet, aren't the simple solutions usually the best solutions in programming?

[Khobai]

Quote

aren't the simple solutions usually the best solutions in programming?

4x3 is obviously anything but simple, or it would be working by now...

as for it being the best solution, any solution that says a locust and jenner are equal for purposes of matchmaking, is simply not a good solution. a locust is half as good as a jenner, so should not be treated as being the same.

[Sable]

Khobai, on 10 June 2014 - 07:16 AM, said:

4x3 is obviously anything but simple, or it would be working by now...

as for it being the best solution, any solution that says a locust and jenner are equal for purposes of matchmaking, is simply not a good solution. a locust is half as good as a jenner, so should not be treated as being the same.

I actually do think a locust and jenner for the purposes of matchmaking should be treated as equal value. I also think a Kit Fox and Jenner should be treated as equal value because they are all in the same weight class. Just because you see the jenner with more value than a locust should not be ther determining factor on its ability to contribute to a team.

Edited by Sable, 10 June 2014 - 07:39 AM.

[Heiggwinie Halberstadt]

I like the idea of battle value match making. ELO can be used to add to a mech's battle value.

[BARBAR0SSA]

They have over 44 million and haven't released ANYTHING but a fancy tech demo of you hangar. They showed a bugged to hell dog fight demo that they had a hard time ironing out bugs in but it's ok they have a possibly BV system that has never been tested.

If the SC community is anything like MWOs they are going to have a lot of angry people.



Regarding the whole 4x3 and BV in MWO.


BV would work IF we didn't require 12v12 matches. PGI has stated that anything less that 24 players in a match would require a lot more hardware and cost. With everyone already on rants about money grabs and prices etc. You can be certain they wouldn't dare invest in new hardware and hope to offset it with transactions. I honestly think that our own community is the cause of why we didn't have the BV type structure of 16v8 with total tonnage of 700T per side etc.

[Sprouticus]

Wingbreaker, on 10 June 2014 - 04:06 AM, said:

According to battle value systems:

X² smaller platforms sporting small weapons are often as valuable as X large platform sporting a traditional set of weapons.

The constant example is Savannah Masters vs Atlas:

215 vs 1897

So by BV, 8 (or 9) savannah masters are equal to a single atlas. This is laughable, even at best. We all know what's going to happen to something that has 9 medium lasers locked onto it. Even if the Atlas can tear through a savannah master per round, the likely chances of it taking down all 9 prior to having its weapons stripped, or the possibilities of TAC, mean that it has a very small chance of winning a supposedly 'balanced' fight.

And frankly, the situation gets worse if you merely switch the Savannah masters to the SL variant. Mind you, they now have two small lasers per, but their BV sinks to 131, meaning that you now have 14 Savannah Masters versus a single atlas.

This is why BV systems are broken. In BT, It does not account for the mathematical realities that BT involves. It assumes that bigger is better, and assigns values by the top end damage that a weapon can do.

It will always be easy to break.

Except you will never end up with 12 mechs vs 4 mechs in MWO. At its most basic level the MM gives you 12 mechs vs 12 mechs. If it were implemented it would be a combined BV on each side +/- some % to ensure that drops actually occur.

AssaultPig, on 10 June 2014 - 04:39 AM, said:

Assuming unbalanced teams weren't allowed, BV would likely exacerbate the problem we already have: a few so-called 'meta' builds dominating all matches. BV might even makes those builds more prevalent, since the more players were running them the quicker you'd find a match if you dropped in one yourself.

If you want to play 12 lights/mediums against four or five assault mechs maybe we're talking, but that isn't the game we have.

Not true. Assuming 2 players launch at the same time with the same Elo, a medium with a meta build may have close to the same BV as a heavy running a non meta build. You would end up with unbalanced teams from a size standpoint but equal in total BV.




the strongest argument against BV was pointed out by a couple of other folks. It would take a LOT of dev time that would best be used for other things. I am a strong proponent of BV, but 4x3 IS simpler to implement and simpler for folks to understand in general.

Side note:

Whomever suggested chassis based Elo is close to the mark. IMO Elo should be 3 levels:

• Variant Elo (if >25 drops in the variant within last 6 months)
  
• chassis (if >25 drops in the chassis within the last 6 months)
  
• Weight class (if >25 drops)

This system fits the current configuration, simply adding some extra values to the Db and a small additional check when determining a players Elo for MM purposes. The Dev time would (seem) to be small and the impact would be quite large IMO. It would also emulate the value of some chassis/variants over others.

Edited by Sprouticus, 10 June 2014 - 08:14 AM.

[R Razor]

Battle Value would have to take into account load out information on each mech as well.........in short, every single component, weapons, modules, engines etc, would all have to have a value assigned to it..........the chassis would have a base value based on size, hard point quantity and location and maximum engine size allowable.......then the components would be added to that base value to give you a value of the mech AS CONFIGURED.

[Heffay]

The problem with BV is it would naturally escalate into everyone bringing the highest BV mech they could. Let the other chump take the weak mech; everyone wants to be the hero and next thing you see it'll be all assaults.

Elo works. People who don't like Elo don't understand Elo.

[Cerlin]

So comparing mwo matchmaker to a mythical SC matchmaker that does not actually exist....That is a totally pointless exercise. Once it is out I will happily compare the two. At least stick to things that are on the same level such as the uncreated meta games or the gameplay so far.

So if you want to argue for a battle rating, pick a game that actually has one and that is actually using it, not just theoretical. Such a game would be warthunder, and even there it is VERY contentious. Basically the Developers gave a battle rating to all the tanks and planes, BUT it is dynamically changed by players performances in each chassis. What this means is the more popular vehicles end up being much more lowly tier (such as Russians and Americans) while the less popular factions that tend to have better players (Germany, UK, Japan) have much much higher battle ratings. This has lead to a situation where many old vehicles fight much more modern ones. The community generally views this as a failure and wants another system to determine how the ratings are set.

I gave this example because its a living one, that while existent is highly controversial and makes most people angry most the time. I think Battle rating sounds nice in theory, but it is all about implementation and execution. HOW do you give a mech a battle rating? How do upgrades and weapon changes effect this? Does this mean if you have a weapon that does better than another, it raises your Battle rating? How about by chassis? Variant? Weight class? Who makes these choices?

Are you starting to see the problem? It was easy when you played tabletop with a couple friends, but once you take in all the variables that can exist, it gets a lot harder.

I am not a huge fan of 3-3-3-3 but I think it is a simple solution to a difficult problem. As a pilot that loves all weight classes with medium being my favorite, I would much prefer this system to the current one...Which leads most my mechs to be a sub-par choice.

Also think about this, the clans are coming, they are going to mess up ANY rating system you have...Who decides the Battle rating difference between clans and IS?

Edited by Cerlin, 10 June 2014 - 08:41 AM.

[Sable]

Battle value is subjective anyway. A meta build is only effective because people found a certain combination of weapons to be that meta. Which would mean that battlevalue would need to be adjusted to accout for it's subjective value. And a subjective value would change constanly. They would need someone to actively adjust battlevalue as the meta shifts. I'd rather have them focus on producing more content than wasting time forever adjusting battlevalue. Class matching is a much longer lasting strategy for addressing the issue. And i am in full favor of the design choice they made.

[MoonUnitBeta]

Almond Brown, on 10 June 2014 - 06:41 AM, said:

but the OP said himself and I quote.

So is it BV or not ffs?

Haha. Can't say if it is or not yet. The difference is that we don’t know enough about star citizen's matchmaking to say it’s the same as battlevalue. It’s a grey area still for Star Citizen, so it’s a grey area here.

[FupDup]

Kassatsu, on 10 June 2014 - 12:13 AM, said:

I would, considering it would get me another 1+ teammates to rush headlong into the enemy team's wall of bullets and die a horrible death instead of using our numbers to flank and overwhelm them.

Imagine a massive team of locusts, vs whatever equivalent value is given to the enemy team. Sure you'd probably lose, but the sheer numbers would cause chaos in the enemy ranks and more than a few to scatter so you could take them out one at a time. The rest would wisely group up (assuming they didn't ALL scatter or leave their slower friends behind), and win the match, but it would be fun.

People tend to forget that games exist to be enjoyed, not to min/max 720noscope esports all day long. Sadly, "people" includes the developers and publishers.

Where did you get the idea that teams could have a different number of players than the other team? This thread is about just adding an external point value to each robot. It never said anything about being able to have a higher or lower player count than the other team.

Sprouticus, on 10 June 2014 - 05:21 AM, said:

--
In a pure BV respect they would be quite valuable because I can get more out of a low BV medium than a meta medium and thus provide more BV to the rest of the team, giving myself a greater chance to win.

"Meta medium" is almost an oxymoron.

Also, "providing more BV to the rest of your team" doesn't quite work that way. You're making it sound as if your team only gains from the exchange. It's both a gain and a loss. Sure, the rest of your team got slightly better robots, but now your robot is now worse than it used to be. You wouldn't be able to contribute as effectively as a better mech (i.e. if you used SHS on a Swayback you would be nowhere near as useful to your team as a DHS Swayback).

Whether the exchange would have a "net" gain or loss in effectiveness would depend on the exact case, because both teams just having equal BV doesn't give them equal effectiveness. Some distributions of "x" BV are inherently more effective than other distributions of "x" BV, despite both of them being the same "on paper." One example I've seen (in this thread) is that a Mad Cat Prime versus an equal BV force of Urbanmechs would pretty much always lose, even though BV is supposed to say that they're "equal."

Edited by FupDup, 10 June 2014 - 09:24 AM.

[N0MAD]

Heffay, on 10 June 2014 - 08:28 AM, said:

Elo works. People who don't like Elo don't understand Elo.

Elo doesnt work in MWO, Elo in a game like this should be based on an XP point system.
A Pilot with a W/L ratio of 50/50 with 2 years xp in MWO should not be pitted with a Pilot with a 50/50 W/L that has been playing 3 weeks as eg.
If Elo was based on Player XP then a simple tonnage rule should be all thats needed to match up teams, roughly same experience roughly same hardware team wise.

[Eddrick]

I don't see BV working very well for this game for several reasons.

#1. If BV is static. It would be balanced around the skill level of the Devs and internal testers. Do you want the game balanced around THAT group?

#2 If BV is dynamic. It would be balanced around the group that is most represented in the player base. A number of people believe, most of the players are not very good at this game. Do you want the game balanced around THAT group?

#3 Can't use the origenal BVs. Because, this is a differant kind of game and the origenal BV was broken anyway.

BV would have problems no matter how it was made. Because, it can't support all groups.

[Name140704]

Wingbreaker, on 10 June 2014 - 04:06 AM, said:

According to battle value systems:

X² smaller platforms sporting small weapons are often as valuable as X large platform sporting a traditional set of weapons.

The constant example is Savannah Masters vs Atlas:

215 vs 1897

So by BV, 8 (or 9) savannah masters are equal to a single atlas. This is laughable, even at best. We all know what's going to happen to something that has 9 medium lasers locked onto it. Even if the Atlas can tear through a savannah master per round, the likely chances of it taking down all 9 prior to having its weapons stripped, or the possibilities of TAC, mean that it has a very small chance of winning a supposedly 'balanced' fight.

And frankly, the situation gets worse if you merely switch the Savannah masters to the SL variant. Mind you, they now have two small lasers per, but their BV sinks to 131, meaning that you now have 14 Savannah Masters versus a single atlas.

This is why BV systems are broken. In BT, It does not account for the mathematical realities that BT involves. It assumes that bigger is better, and assigns values by the top end damage that a weapon can do.

It will always be easy to break.

I love your post, reminds me of why I don't bother with NGNG anymore.

Take off your blinders and consider what they have kept directly from BT.

Savannah Masters...give me a break. "Because bad example that does not apply to this game!"

[Almond Brown]

MoonUnitBeta, on 10 June 2014 - 08:59 AM, said:

Haha. Can't say if it is or not yet. The difference is that we don’t know enough about star citizen's matchmaking to say it’s the same as battlevalue. It’s a grey area still for Star Citizen, so it’s a grey area here.

So how did we end up hear again?

P.S. The funny thing about the Forums and its participants is that it is very obvious "everyone" has free time on their hands. Sadly, those same people will not allow PGI time to add and fix things based on how the participants make use of it. It is f'ing brutal and yet we argue over **** that does not yet even exist in another game on these Forums. I think just doing that here shows a true lack of any type of class at all.

Wonder how those same folks would feel if strangers came into their homes and started telling them how it stunk and looked like a shithole. No worries right, it is just their opinion...

[oldradagast]

FupDup, on 09 June 2014 - 10:44 PM, said:

How many people who are in favor of ButtValue are volunteering to pilot a low-BV mech into combat against a high BV one? Yes, that's right, "bad" mechs would still be able to be placed against "good" ones, unless you have every mech in a given match have the same average BV rather than both teams having the same total.

Part of why this isn't an issue in TT is because you can and generally do control more than one unit at a time. It doesn't matter that my Panthers are outright inferior to any heavy or assault mech in existence, because I'm not sitting in the cockpit of one. I'm watching them from above, giving orders to them and other mechs.

But in MWO, all you get is one mech, period. This isn't like playing Starcraft where you can send a huge army of Zerglings against a single Ultralisk or something. This is like playing Starcraft and sending ONE Zergling against an Ultralisk. Not many people would enjoy that, methinks...

A good point.

In tabletop, nobody cares if the light mech that dies 2 rounds in was having fun - it's just a unit under your command. But in a video game where we each pilot 1 unit, that suddenly becomes an issue, which means what makes for good balance and game play changes.