54 replies to this topic

[feeWAIVER]

Hobbles v, on 19 January 2022 - 05:46 PM, said:

A formula would give flawed results because there is too many variables. The relative balance is affected by mech size, variety and number of hardpoints, speed, available tonnage and slots.

For example. Ermicro lasers are great for pirhranas but utterly useless for Dire wolves.

Guass rifles are really good weapons, but not on a firestarter.

Variables like this cant accounted for in simple formulas. Its why changes are tested in actual combat with a variety of mechs

A formula could tell you all of these things.
Math, kids. I don't even....

[ScrapIron Prime]

feeWAIVER, on 19 January 2022 - 06:41 PM, said:

A formula could tell you all of these things.
Math, kids. I don't even....

A wise man once told me that any equation with more than seven variables is no longer math… it’s art.

[feeWAIVER]

ScrapIron Prime, on 19 January 2022 - 07:43 PM, said:

A wise man once told me that any equation with more than seven variables is no longer math… it’s art.

We ain't gonna get to Mars with that kind of spirit.

[ScrapIron Prime]

feeWAIVER, on 19 January 2022 - 08:12 PM, said:

We ain't gonna get to Mars with that kind of spirit.

Sure we will! We just break the problem down into numerous six-variable equations and find a few talented artists to pull it all together.

[feeWAIVER]

justcallme A S H, on 19 January 2022 - 12:42 PM, said:

How does a formula account for gameplay, player behaviour, maps, the things you are trying to kill etc etc.

Thus no formula exists.

If such a formula did exist, abd actually worked, no game would never need ongoing balance.

Because everything is measured. Everything is quantified.
From the size of a mountain, to the speed of a mech, to heat dissipation, to the size of a side torso, to the damage of a laser, to the statistical probability that a player will do X, Y, or Z.
E v e r y t h i n g in reality is measured by some form of math.

And if you're smart enough, you can figure it out.
PLEASE don't ask me for evidence. Go talk to Nightbird about it or something.

Edited by feeWAIVER, 19 January 2022 - 10:07 PM.

[MW Waldorf Statler]

feeWAIVER, on 19 January 2022 - 09:42 PM, said:

Because everything is measured. Everything is quantified.
From the size of a mountain, to the speed of a mech, to heat dissipation, to the size of a side torso, to the damage of a laser, to the statistical probability that a player will do X, Y, or Z.
E v e r y t h i n g in reality is measured by some form of math.

And if you're smart enough, you can figure it out.
PLEASE don't ask me for evidence. Go talk to Nightbird about it or something.

Math have his Problems with dynamic chaotic Situations and each Human has each Daytime a other condition (oh...my Kid is here and touch on my Keyboard), have with each weapon a other skill, have each second other Situations and other tactics , thats why wars gets lost ...Math say the USA wins the Vietnam War,Math sa y the british ,Russia and USA wins in Afghanistan .Math is useless in dynamic chaotic Systems , no Math can say before which way goes a waterdrop on a Handback ,thats why the Biosphere Projects was a Disaster .its give more Factors thats you forget to see as you seeing.

When the Math say ,the Gauss is the Meta weapon ...im can not handle it with his Mechanic ,for me the Weapon is uselesswhile each People has his own physical and Psychologic Handycaps .

Mathematically it is unlikely that life will evolve on earth and that there will ever be anything like man, nor that he will become the dominant life form.

Mathematics gives the narcissistic humanity the feeling of being in control in an uncontrollable world, being able to predict something, and is the more modern form of astrology or belief in gods that mathematics can determine, guide and control our destiny. Mathematics is more gestural science than natural science, yes deeply based on the latter but misused for the former

https://inis.iaea.or...06/20006763.pdf

https://arxiv.org/ft...6/1106.4087.pdf

Edited by MW Waldorf Statler, 19 January 2022 - 11:01 PM.

[YueFei]

feeWAIVER, on 19 January 2022 - 06:41 PM, said:

A formula could tell you all of these things.
Math, kids. I don't even....

No, a formula wouldn't. An algorithm might, but as any graduate of Computer Science could tell you, the computability of certain classes of problems are infeasible or even potentially undecidable. This is not about what computers are capable of. As a professor once told me, Computer Science would exist (and has existed) without computers. There simply exists certain classes of problems that are undecidable.

Even for a game as simple as Go, a turn-based non-real-time game with perfect information and a far more limited decision-space for the players, whilst we have AIs that can beat humans, we don't know and cannot rigorously prove that it is playing optimally. Hell, we don't even know if it's perfectly balanced or if the player who goes first will always win if he plays optimally. What should the correct value of Komi be to compensate for that 1st player's advantage? We don't know.

It is because of that (intractability or undecidability) that empirical testing, experiments, etc., are still necessary.

Now take a real-time shooter with imperfect information and 24 players each in their own unique mechs/builds going at it simultaneously under varying conditions on different maps/modes.

Billions of dollars are being poured into AI research, into new algorithms, into determining the computability of various problems. Let's get real for a second: none of that $$$ is gonna go into researching any kind of algorithms/AI for MWO for the purposes of determining game balance. Matter of fact, if you think you can come up with one, you should be working for DARPA or something and raking in the big bucks by solving problems that actually matter to the real world instead of a video game, like the Travelling Salesman Problem, or solving the P versus NP Problem.

Which means: pbiggz is right. MWO game balance can be informed by and use the weapon stats as a starting point, but you won't solve the problem of game balance from those alone. That has to be done empirically... as it is done in Go to determine what value of Komi is fair.

Edited by YueFei, 19 January 2022 - 11:00 PM.

[feeWAIVER]

Guys, we're talking about weapon balance here. It's math. What are you going on about?
We're talking about Mechwarrior, right?

[feeWAIVER]

Remember when someone was talking about if you shrink an Atlas by X%, then the torso twist would expose Y% of the side torso during a laser vomit alpha. That's a formula that someone smarter than most of us figured out.

So yeah. I'd hope there'd be some kind of guiding formula for dmg v armor v speed v size to achieve some final point of balance.
Unless new weapons come out, there shouldn't be a reason to xml edit forever.
Or are we just gonna keep patching armor and damage in waves until 2 years from now the game is inflated beyond recognition?

I think it's silly that we're at a point that we're buffing blood asps and battlemasters now.
I mean, okay, but what are we doing here?

[MW Waldorf Statler]

Quote

[color=#959595]Remember when someone was talking about if you shrink an Atlas by X%, then the torso twist would expose Y% of the side torso during a laser vomit alpha. That's a formula that someone smarter than most of us figured out.[/color]

thats correct

And what does that say, under what conditions, with what skill and which Weapon from which range ,and what latency, from what angle does someone hit this torso (the Target use Torso Twisting?), where the engine has difficulties distinguishing between front and back armor?here not AI fight against AI with same basic algorithms ..here Players thats have here first FPS game against new and old players with 30+ Years Gaming Experience and others with many Handycaps.

What use is the perfect weapon to someone who still has to stop for every shot and needs 3 minutes to aim at something or plays with a steering wheel and a ping from 300

Edited by MW Waldorf Statler, 20 January 2022 - 12:27 AM.

[caravann]

many games uses fake bullets to ensure that people do not die too quickly.
armor as hp is better than special armor and special hp. lighter mechs have less structure and means armor is higher valued. lighter mechs should consider side torso as part of the arm. a shoot in the arm destroys the side torso.

balance is not key but backup of the team. a weak mech is going to be attractive and player can fool players to run after mechs to end up hunted.

trying to balance is going to end up removing playstyles. when you can not outsmart you get trenchwars.

[MW Waldorf Statler]

when you in Batllefield balanced the Bomber against the Infantry with a AA or other Jets is useless when no people use the AA or have the Skill to fly a Jet.We have here many Lasers ,with many ranges, PPCs with or without Minimum ranges , LRMs,MMMs,SRMs , many ballistic Weapons with very different mechanism ...we can balanced a Nuke against a Sniperrifle, a shotgun, a LMG, a Knife?

we not carry like other FPS games only a primary and Secondary Weaponand have more as 100 Chassies with very differnt Geometrys and sizes and Weapon Space.

so we balanced a giant 20m 50kmh Mech with 3 shotguns and 4 LMG against a mech with 2 Knifes, 3 Mortars and 2 nuke against a little 140Kmh mech with only 2 Shotguns and a single Nuke

Edited by MW Waldorf Statler, 20 January 2022 - 12:51 AM.

[Curccu]

justcallme A S H, on 19 January 2022 - 05:06 PM, said:

We didn't need a formula to know cLPL was overperforming slightly in QP. Now is the little nerf enough, again gotta let it play out for a month or two to see.

Why cSPL nerf? Those still seem to be pretty rare sight (from what I have seen) outside of light mechs.

caravann, on 20 January 2022 - 12:20 AM, said:

many games uses fake bullets to ensure that people do not die too quickly.
armor as hp is better than special armor and special hp. lighter mechs have less structure and means armor is higher valued. lighter mechs should consider side torso as part of the arm. a shoot in the arm destroys the side torso.

balance is not key but backup of the team. a weak mech is going to be attractive and player can fool players to run after mechs to end up hunted.

trying to balance is going to end up removing playstyles. when you can not outsmart you get trenchwars.

Say what?

[MW Waldorf Statler]

Balancing...nerf the Lights...nerf the Assaults, nerf the UAC, nerf the RAC,nerf the LRM, nerf the SRM, nerf Groups, nerf teamplay killing ego players..nerf and balance all what me give not a Advance against the others

[Hobbles v]

feeWAIVER, on 19 January 2022 - 11:13 PM, said:

Guys, we're talking about weapon balance here. It's math. What are you going on about?
We're talking about Mechwarrior, right?

Its not just math. Because the weapons go onto these other things called mechs. Those mechs are given different loadouts, even identical chassis will have different weapons. Those different mechs will almost certainly use different skill trees.

Then add on that those mechs get different maps or even spawn locations every round.

Piloted by differing people some whom might be good most days, but some days are drunk, others who are terrible all the time unless you give them a specific weapon to use.

Then theres the psychological effects certain play can have, it might not be mathematically a lot of damage output but the frustration they cause can make players tilt and screw up in the best mechs.

Example the new ecm urbie with light PPCs. Defending hellbore in seige mode, from the mountains. It is not doing a ton of damage. But it is drawing a ton of attention and frustrating the enemy because with its ECM, high mounts, nimbleness, jump and burst fire Ppcs, it is extremely hard to counter snipe. The frustration it causes cant be quantified.

The point is, so many of the variables are subjective and cant be quantified, but they can easily be felt.

Example 2. The recent screen shake nerf to LRMs. I dont think LRMs needed a nerf at all, they are actually pretty weak IMO, im sure a lot of low tier players would disagree and find LRMs to be very scary.

But as far as nerfs go, i think the cauldron chose the right nerf. They kept the weapon just as effective at killing as it was before, but reduced screen shake allows for easier counterplay shooting back and may be less frustrating for people getting peppered by them running for cover. This type of nerfs effects cant be measured mathematically as the results are subjective. For me who has no problem with LRMs almost nothing changed. For the tier 5 guy getting hammered and feeling hopeless, ot probably improved his game.

[1453 R]

feeWAIVER, on 19 January 2022 - 11:39 PM, said:

Remember when someone was talking about if you shrink an Atlas by X%, then the torso twist would expose Y% of the side torso during a laser vomit alpha. That's a formula that someone smarter than most of us figured out.

Cool. That doesn't inform balance. Or at best it gives the people responsible for balance one of the data points they need.

How common are energy-centric 'Mechs? What's the average laser alpha? What's the highest effective alpha? What's the Lulzmeme number? Which numbers should be accounted for? How long is the Atlas taking to respond to being attacked and start twisting? What range of "respond to being attacked" values is considered reasonable and fair? How durable is an Atlas supposed to be in the first place? What's the target value for Living Longer vs. Killing Faster?

There comes a point, in this exponentially increasing sea of variables, where "eyeball it, tweak, put it into the wild and see, then repeat" is the most mathematically efficient way to do the work. Because trying to solve for every last conceivable variable in MWO at once for a single perfect, mathematically flawless Proof is so time-consuming and arbitrary as to be effectively impossible.

There's a reason all the most successful live-service game developers, the ones with games people actually like, iterate rapidly rather than building out a mathematically "balanced" game once and then never touching it again.

feeWAIVER, on 19 January 2022 - 11:39 PM, said:

So yeah. I'd hope there'd be some kind of guiding formula for dmg v armor v speed v size to achieve some final point of balance.
Unless new weapons come out, there shouldn't be a reason to xml edit forever.
Or are we just gonna keep patching armor and damage in waves until 2 years from now the game is inflated beyond recognition?

First of all: uhhhh, yes? This month-by-month iteration is exactly the correct process. Note that the changes have been getting smaller in scope every month; we got four weapons changes this month as opposed to four pages of weapon changes, and each change was quite small indeed compared to the Wild West that was the spring patches. 'Mech quirk fixes are being reduced in scope as well, and fewer chassis are getting extra passes. As the game evolves (however much it can, at this point) and things shift over time, the balance team will account for that. "XML edit forever" is absolutely the goal, thank f@#$ing god we're doing it, and it only took us how many years to get to this point?

Second of all - if there was a 'guiding formula' one could follow for Perfect Balance, that formula would have a solution. That solution would be The Best Most Optimal 'Mech in MWO, and using anything else would be an actively bad idea. A "guiding formula" indicates a solved game, and solved games are no fun for anybody. If a game is solved, it's time to shut the servers down and put it away.

feeWAIVER, on 19 January 2022 - 11:39 PM, said:

I think it's silly that we're at a point that we're buffing blood asps and battlemasters now.
I mean, okay, but what are we doing here?

Why not? Blood Asps are easy to hit and pick apart, and Battlemasters aren't much harder. The Battlemaster, in almost all formats, is an energy-centric assault 'Mech, which is a classification that just doesn't work well in MWO. Even then, it's worse at that job than many other assault 'Mechs. Battlemasers are drastically inferior to quirked Awesomes as PPC showboats, the new 7D Stalker eats their lunch insofar as Sphere assault-weight laser spam goes, and the missile-based Battlemaster builds have always been memes. What's the Battlemaster good at, precisely? It's defensively mediocre for its weight and offensively it's got little going for it that other 'Mechs don't do better.

Yeah, Blood Asps have those skyhook shoulder hardpoints, but they're not anymuch smaller than Dire Whales and they've got almost no defensive benefits. Blasps have remarkably few useful builds - you get Gausslasers or autocannon blap, and while those are serviceable they're also not unique to the Blasp. The critters also didn't receive any significant buffs, other than the non-ECM CTs getting a RaDerp quirk to make piloting non-ECM Blasps semi-palatable. They certainly weren't buffed so heavily as to becoming domineering, and if somehow, by accident, they were?

Well, hell. That's what rapid iteration is for!

[SharDar]

Hobbles v, on 19 January 2022 - 05:46 PM, said:

A formula would give flawed results because there is too many variables. The relative balance is affected by mech size, variety and number of hardpoints, speed, available tonnage and slots. For example. Ermicro lasers are great for pirhranas but utterly useless for Dire wolves. Guass rifles are really good weapons, but not on a firestarter. Variables like this cant accounted for in simple formulas. Its why changes are tested in actual combat with a variety of mechs

It seems the mechs should be designed for the weapons and not the weapons designed for the mechs. Providing parity between weapons doesn't need to factor in which mechs they would be used in. Mech design would need the same kind of balancing to make sure that the available weapons don't favor a particular mech too heavily.

[1453 R]

SharDar, on 20 January 2022 - 08:42 AM, said:

It seems the mechs should be designed for the weapons and not the weapons designed for the mechs. Providing parity between weapons doesn't need to factor in which mechs they would be used in. Mech design would need the same kind of balancing to make sure that the available weapons don't favor a particular mech too heavily.

Problem: all the 'Mechs were designed forty-odd years ago for a completely different system. Ship has long, long since sailed on that one.

[ScrapIron Prime]

1453 R, on 20 January 2022 - 08:46 AM, said:

Problem: all the 'Mechs were designed forty-odd years ago for a completely different system. Ship has long, long since sailed on that one.

And be thankful we don't have hardpoint SIZES like they implemented in MW5 (and multiple titles before that).

[pbiggz]

justcallme A S H, on 19 January 2022 - 05:06 PM, said:

Yep could for sure be useful when doing initial designs before you get onto the field to play and see how it all feels/works in the game itself.

Right now for MWO - not needed. Most of the values have been, apart from the initial bulk changes, just tweaks here/there based on mostly gameplay/feel and community feedback.

Some changes of course we math/test in lobbies/training grounds out pretty hard and in depth some probably too much. That isn't needed in every case though.

We didn't need a formula to know cLPL was overperforming slightly in QP. Now is the little nerf enough, again gotta let it play out for a month or two to see.

We're in agreement on that just to be clear.

You dont use a hammer to drive in a screw. Different tools for different problems.