32 replies to this topic

[razenWing]

If you deny the existence of meta...

... you are delusional.

--------------

We have over 30, 40 mechs, with 5-6 variants each. But the irony of upper tier combat is, meta mechs and mech builds, due to their effectiveness, are just used way more often in upper tier (1 and 2). As such, diversity is almost non-existence in high tiers.

So despite the vast possibility of diversity, the game is actually quite limited. And that's the true irony in all of this. Now, I am not saying that you can't make your 1 mg 2 ML 1 SRM 1 LRM mech to work. In fact, every once in a while, you might even get a 1k game out of a piece of crap like that. But, it's not as effective as someone that just straight laser boat, srm boat, or AC boat.

In this game, specialization is key. You want to fight good in mid range? Get all weapons that will complement mid range. You want to snipe? Convergence long range high alpha. These are forumlaic. Almost anyone you meet in T1-2 can memorize the builds and mechs right off the bat.

But occasionally, you want just want to compete with other wacky builds that you will see in tier 4-5 with your own wacky builds. And in T1-2, it's just not possible. (Unless you are proton-esque, then you kick *** no matter what, but that's an oddity, not a norm)

So borrowing from Solaris idea with the proposed mech classes (and if I interpret correctly, is bsaed on meta effectiveness), can we implement something like that in QP as well? Or maybe just expand the algorithm that if you are in a mech class below a certain class, you are allowed to be pair with tiers 3-5. Again, this shouldn't have much controversy as that's what you guys are planning on with Solaris already. That way, you guys are not pressured to make meta mechs either. Just release mechs that are funky and unique. You can please both meta chasers and lore fans at the same time without lore fans feeling like their 1 mg 1 srm 1 LRM 2 ML mechs are getting shafted.

Just throwing out ideas.

Edited by razenWing, 16 January 2018 - 08:02 AM.

[adamts01]

That's tough, just because of the vast build diversity. Take the Timby in its prime for example, people were somehow able to build a tier 5 mech out of tier 5 parts. As someone who likes to run fun builds in tier 1 I feel you, I just don't think this is anything that could work. But, on the bright side, there doesn't seem to be a match maker, so just go ahead with whatever build you like!

[Bud Crue]

In order for PGI to provide the system you are suggesting, then PGI would have to admit that there is a meta, identify it correctly, and modify the MM accordingly. They have historically never been able to do this. If we want to be charitable here, and view the Solaris announcement in a positive manner, its that they are finally willing to recognize that some mechs are in fact better than others, and are going to make the attempt to group them together, at least in terms of 1v1 and 2v2 game play. Beyond that however, PGI has in fact repeatedly over the years essentially denied the existence of a meta (see Paul’s commentary in 2015 regarding all mechs starting at the variant level, having a role and equivelant game play value to all others; to Chris’s commentary following skills tree that the widest individual performance difference between chassis is a mere 8%) or incorrectly identified it by balance proxy (even after the initial Kodiak nerfings Russ was still tweeting about what a dangerous power house the Kodiak-5 was, seemingly unaware that it was the 3 that was dominating play at the time).

Alas, Solaris and its tiered chassis divisions is about the best you can hope for. Hell, it more than I dare hope for until I actually see it. PGI has proved time and time again that they simply can’t recognize what the meta is or at least what its actual impact on the game play of the moment is, and hoping that they could now suddenly figure it out and give you the sort of mech class divisions you are looking in 12v12 play just seems beyond their capability. Now excuse me, I have to go play my Timberwolf. PGI thinks its so meta that it still needs negative quirks.

[sycocys]

Solaris should be pretty interesting to see the divisions with PGI's dartboard balancing strategy.

Sometimes its just downright baffling that they haven't implemented some sort of battle value system that ranks on tonnage + equipment/consumables + skills (would even have worked with the prior bad skill system).

It would be a more accurate way to run the MM than any of the systems they've put for so far since everyone has just been more or less lumped into the same bucket excluding only maybe the very least skilled players.

[Burke IV]

I thought i would let a few others get in first blame the skill tree. People said it was going to kill build diversity.

[El Bandito]

People will still min max even if battle value is incorporated into QP. It is probably not worth the effort.

Burke IV, on 16 January 2018 - 08:47 AM, said:

I thought i would let a few others get in first blame the skill tree. People said it was going to kill build diversity.

Except even before current skill tree people used meta mechs. Remember the Splatpult era? Remember the PPDLF era? Remember the Timbie era? I'd say QP mech diversity is actually larger than before, since we now have far more mech selection than 2012-2014.

Edited by El Bandito, 16 January 2018 - 09:06 AM.

[Sigmar Sich]

What if QP gets same treatment as CW, where underdogs get tonnage bonus? So for example "bad" mechs will be considered by MatchMaker as -5 or -10 tons of their weight, and meta mechs would count as +5 tons, etc.
Unless MM doesnt balance weight. I'm sure it did at some point of time.

P.S. Or you could balance mech chassis, you know... Those who can't monoboat - get more quirks (preferably to a weapontype for which you have least hardpoints). Those with bad hardpoint placement - get more quirks. Those with bad hitboxes - guess what, more quirks. You just need a dedicated gamedesigner with a full-time job, with enough imagination and experience, and freedom of action.

[El Bandito]

Sigmar Sich, on 16 January 2018 - 09:12 AM, said:

P.S. Or you could balance mech chassis, you know... Those who can't monoboat - get more quirks (preferably to a weapontype for which you have least hardpoints). Those with bad hardpoint placement - get more quirks. Those with bad hitboxes - guess what, more quirks. You just need a dedicated gamedesigner with a full-time job, with enough imagination and experience, and freedom of action.

PGI already made such balance effort. Variants with fewer hardpoints have better offensive quirks while mechs with bad hitboxes have better durability quirks. Of course, PGI can do much better job at balancing, starting from base tech.

[Sigmar Sich]

El Bandito, on 16 January 2018 - 09:15 AM, said:

Of course, PGI can do much better job at balancing, starting from base tech.

Exactly.

[adamts01]

El Bandito, on 16 January 2018 - 09:03 AM, said:

I'd say QP mech diversity is actually larger than before, since we now have far more mech selection than 2012-2014.

Just because we have so many re-skins. It's all about big alphas and high mounts. Nothing has changed.

[Almond Brown]

sycocys, on 16 January 2018 - 08:43 AM, said:

snip

Sometimes its just downright baffling that they haven't implemented some sort of battle value system that ranks on tonnage + equipment/consumables + skills (would even have worked with the prior bad skill system).

snip

Your joking right? That can be the only explanation for such crazy talk. A Battle Value system NOW! given 15 separate selectable Mech weights, 88 base chassis, most with 5-6 Variants, 114 distinct weapons and 8 support weapons not to mention Skills and Quirks and the Math gets a tad much really. I don't do Math's well but 15 x 440 x 114 x 8 x 92 (Skills) x ...

OK, just that creates in excess of 500+ million combinations. LOL! BV FTL...

[Water Bear]

Well, it depends on how "wacky" we're trying to get, and how good you're trying to perform. I think you can incorporate a good bit of build diversity and still do great in T1 - I like to think that I do this. But you can't go full on SRM 2 + LRM 10 + 4 medium lasers in a Cyclops and expect much.

In my mind, T1 in this game is not a high bar. I used to play more competetive e-sports style games (Starcraft II, in particular) and I started running into people that were very difficult to beat once I made it to the equivalent of tier 2 in that game (top 8 Diamond, 1v1). I'd go against people in low masters who could match my APM, base building ability, tenacity, etc. Very hard to overcome.

I find that in MWO, with a maxed out T1 bar, at least 50% of people I encounter in a match are lacking a lot of basic skills. In particular, they just can't aim well enough to counter an abusive use of cover. For example, if someone is hill-humping with a Roughneck or a Battlemaster, there's really only one counter to that - aim really, really carefully at the tip top of their mech. Shoot it in the head if you have to. But most people's solution to that problem is to just find another target - which is fine, except that frees up the BLM to drop 40-50 point alphas onto your teams head non-stop throughout a fight. He's the highest value target, so if you ignore him it's gg.

Any way, my point is that you can totally get away with highly sub-optimal builds at the highest level of this game. The standard is pretty low.

[sycocys]

Almond Brown, on 16 January 2018 - 09:32 AM, said:

OK, just that creates in excess of 500+ million combinations. LOL! BV FTL...

It's really not that hard of math, just basic addition to get a BV.

You just get a sum of the values of the parts and have a baseline battle value for the mech.

Then you slot them into an appropriate grouping.

The amount of combinations is a complete non-factor, simply doesn't matter in the least bit when determining or slotting by BV.

[Spheroid]

People have wanted variant specific pilot ratings for years, years. I can only conclude that the increase in data fields would be prohibitive either in cost or server query performance.

Its not like this topic hasn't been brought up before.

[Tarogato]

sycocys, on 16 January 2018 - 10:52 AM, said:

It's really not that hard of math, just basic addition to get a BV.

Just remember, where there is a BV system, there is a way to exploit it by finding combinations of mechs and weapons that overperform their intended ratings...

I think it's easier to just rate chassis and variants into MM tiers rather than trying to come up with a BV system that actually works. BV is miles more complicated. If a mech is overperforming for its BV... then you have to ask the question, "why?", and fix it... rather than simply knocking it up a tier to where it belongs without affecting anything else.

[sycocys]

Tarogato, on 16 January 2018 - 11:45 AM, said:

Just remember, where there is a BV system, there is a way to exploit it by finding combinations of mechs and weapons that overperform their intended ratings...

It would probably be doable to find a few niche exploiting mechs given their desire to keep the quirk system, but it's not going to be easy to find a combination that flat out outclasses others in a given range when gear, weapons and skillpoints have values accumulated in - even if you kept the weight class separations and the 3/3/3/3 drop setting in QP.

It doesn't solve for player skill, but the elo didn't and the PSR doesn't either, it just puts mechs in matches that will be in close to the same performance levels which for most of the player base would probably do a better job of separating the meta runners/t1 elite types from the newbies/super casuals/lore builds and the people in between without any further systems in place than anything we'd had up to this point.

They simply aren't going to solve the equation for player skill, but it's not that ridiculous much work to create at least far more balanced matches in terms of what's being fielded.

[Sjorpha]

I think there is quite a lot of build and mech diversity in the current metagame actually, a lot more than before skill tree and certainly a lot more than when I started playing (before clans and during the first poptart era.)

[Fleeb the Mad]

Anyone who's played tabletop knows that BV itself was flawed. So was BV 2.0.

That's just because some things cost much more than they were worth and other things less. For example, BV 2.0 greatly inflated the BV value of any mech with jump jets because mobility was a multiplier, to the point where it cost more for a mech to be 4/6/4 (i.e. 64 kph with JJs) than 5/8/0 (86 kph with no JJs).

Some mechs are definitely better than others, but having to assign and balance (!) a point value for not only every chassis, but every weapon and arrangement of equipment would be an order of magnitude more complicated than it is presently, and also about that much more complicated than the original BV system because a real time PvP game adds more variables.

For example, the original BV system didn't have any values for chassis. Only the equipment, armour, structure, etc. on them. So every mech with ivarient now needs a value independent of equipment. Different omnipods need different values, sometimes different again when on a certain CT. Do weapons get increased BV when put on a chassis that has quirks for that weapon? Do you account for the effects of weapons being amplified when boated or paired with complimentary ones? Do you base it off of what's most popular? If you do, do you live with the value being wrong for at least one of the ways people use them?

It's not simple people. Don't treat it that way.

[Zergling]

I have zero confidence in PGI's ability to correctly set mech Tier for Solaris.

They just haven't demonstrated knowledge of the meta in past.

Edited by Zergling, 16 January 2018 - 01:49 PM.

[slide]

In a highly competitive environment there will be only one way to balance the game.

Same mech different pilots.

This can be expanded some to incorporate different roles but it essentially still distills down to only a few mechs/variants that offer the optimum configuration in each role.

It is axiomatic that as competitiveness goes up, build diversity and fun go down, also the game becomes less like BT and more like the arena shooter it is.

I am close to but not in Tier 1, I have no desire to get there. Lower tiers are more fun and more like the game I want to play.