Jump to content

Game Balance - Maximum Return, Minimum Effort


46 replies to this topic

#1 Xandralkus

    Member

  • PipPipPipPipPipPip
  • 344 posts
  • LocationEarth, for the moment...

Posted 07 March 2013 - 02:13 AM

I'm going to outline a series of simple steps that the devs can take to move the game to a more balanced state, with relatively little effort.

Until we can figure out what ECM should do (remember, Sarna is not infalliable), remove it. There are tons of excellent ideas on the forums about how to fix ECM. Electronic warfare should be a deep, engaging, fun mechanic for all players involved, and have a low enough barrier to entry so that first-day-players can participate alongside veterans. When the devs have a working model of how to do this, then it's time to reintroduce ECM - and not one second sooner.

Until we can figure out what SSRM's should do, and how they should behave, remove them. SSRM6 should be balanced against SRM6. This is how you know when it is time to reintroduce SSRM's.

Undo double armor. There is nothing wrong with landing an alpha strike and seriously damaging a mech of similar tonnage. That's the way Mechwarrior is supposed to work.

Increase LRM spread when using indirect fire. If the LRM boats want anything close to full damage output, they should require line of sight.

Decrease jump jet recharge speed. This ends the annoying issue of pop-tarting as a regularly exploitable tactic.

Yeah, we are still going to have balance issues even after this. These changes are not perfect, but they are WAY closer to an ideal balanced state than the game we have right now. Most importantly, these changes will not require any groundbreaking new features.

In fact, at every turn, groundbreaking new features that were supposed to balance the game have failed apocalyptically miserably. I don't care how badly the managerial trolls are pushing for groundbreaking new features and MOAR money. The central premise of business is 'you have to spend money to make money'. Without risk, there is NO reward. By playing it safe and constantly generating new content for a steady stream of MC income, it is fundamentally eroding and destroying the long-term playability of this game. There is no way around it. Managerial trolls, if you want real profits, you have to grow a pair and take the risk.

Devs, take a break from your MC content spamming, go back to the drawing board, and finish designing your game. I really, really don't mind if we have to delay seeing new mechs for a month or three, if it means we can have a balanced game.

Edited by Xandralkus, 08 March 2013 - 01:44 PM.


#2 FuzzyLog1c

    Member

  • PipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 116 posts

Posted 07 March 2013 - 02:19 AM

Can I have some of whatever you're smoking?

Halving armor would eliminate the need to target vital or weakened sections of hostile combatants, effectively breaking the most fundamental tacit of the fundamental Mechwarrior gameplay mechanics.

Compared to other mech games, Mechwarrior distinguishes itself by being all about positioning, heat (resource) management, and landing accurate shots against vulnerable portions of an opponent that is trying to do the same to you. If anything, they should double armor again, not halve it.

If you're getting killed by indirect fire LRMs--which require careful aim, 5-10 seconds of flight time, and most importantly, that their target not move a muscle--and have the sheer nerve to whine about it, you should be ashamed of yourself.

Edited by FuzzyLog1c, 07 March 2013 - 02:45 AM.


#3 Xandralkus

    Member

  • PipPipPipPipPipPip
  • 344 posts
  • LocationEarth, for the moment...

Posted 07 March 2013 - 10:59 AM

Armor would need to be reduced to much less than half in order to make precision fire have no benefit at all. Fire a PPC salvo into a heavy/assault with red armor now, and at best, you might make the internals turn some shade of yellow. Half the armor, and those internals are going to be bright red.

Suppose you miss your desired body part with half-armor. Your decision to shoot them still matters. You still did statistically significant things. Maybe not quite as much as if you hit something already damaged, but at least the utility gained from shooting a non-damaged target will not approach zero.

LRM indirect fire is an exception to the rule "If a player can hurt you, you can hurt them back". That's a pretty important rule in game balance, so if we are ever going to make an exception to that rule, then we need to answer the question "How much does a player deserve to hurt another player who is incapable of retaliating?" Any answer other than zero or very little is the wrong answer.

This dice-roll of no ECM = LRM apocalypse, or ECM = LRMs useless...needs to end. If a weapon requires simultaneously nerfing and buffing, then the problem is not with the weapon, it is with interacting mechanics.

#4 StalaggtIKE

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Raider
  • The Raider
  • 2,304 posts
  • LocationGeorgia, USA

Posted 07 March 2013 - 12:24 PM

View PostXandralkus, on 07 March 2013 - 02:13 AM, said:


I'm going to outline a series of simple steps that the devs can take to move the game to a more balanced state, with relatively little effort.

Until we can figure out what ECM should do (remember, Sarna is not infalliable), remove it. There are tons of excellent ideas on the forums about how to fix ECM. Electronic warfare should be a deep, engaging, fun mechanic for all players involved, and have a low enough barrier to entry so that first-day-players can participate alongside veterans. When the devs have a working model of how to do this, then it's time to reintroduce ECM - and not one second sooner.

Agreed. ECM is currently a conglomeration of many things, a few in which are beyond the current time line. I would hate to see it go. They really need to reassess its risk vs reward and input vs output. Perhaps all it requires is a tweak in numbers. This is where it fails, not from being OP, but being too good for its cost. After it has be reexamined it can be added back.

Quote

Until we can figure out what SSRM's should do, and how they should behave, remove them. SSRM6 should be balanced against SRM6. This is how you know when it is time to reintroduce SSRM's.

They're pretty balanced right now. I would prefer to see them require a new lock between each salvo launch. Also a break in los equals immediate target decay. They also should receive the ability to dumb fire under ECM.

Quote

Undo double armor. There is nothing wrong with landing an alpha strike and seriously damaging a mech of similar tonnage. That's the way Mechwarrior is supposed to work.

This is a sticky situation. Due to weapon convergence, cooldown times, increased damage and ammo. A lot of tweaks would have to be made before armor can be simply halved. With near perfect weapon convergence a little bit of team work will have players going down in a couple of seconds. Boating will be even more destructive. The cooldown times would need to be increased. Also a decrease in damage and ammo per tonnage for some weapons: LRM, SSRM and SRM. Not a bad idea, however as stated a lot of other work must go into it.

Quote

Increase LRM spread when using indirect fire. If the LRM boats want anything close to full damage output, they should require line of sight.

Agreed. Firing blindly results in support fire only. As you stated both players should be able to exchange fire for a balanced game. This keep things fair.

Quote

Decrease jump jet recharge speed. This ends the annoying issue of pop-tarting as a regularly exploitable tactic.

Agreed.

Edited by StalaggtIKE, 07 March 2013 - 12:26 PM.


#5 Xandralkus

    Member

  • PipPipPipPipPipPip
  • 344 posts
  • LocationEarth, for the moment...

Posted 08 March 2013 - 12:35 AM

View PostStalaggtIKE, on 07 March 2013 - 12:24 PM, said:

With near perfect weapon convergence a little bit of team work will have players going down in a couple of seconds.


This already happens. A lot. Double armor doesn't slow them up, it disproportionately strongly incentivizes that the team do only one thing at a time.

Part of the reason I advocate the removal of double armor is so that the combat performance gap is narrowed between highly coordinated teams and solo PvPers. At the moment, the ONLY way to destroy an enemy mech in any timeframe approaching that of a conventional shooter (a few seconds) is with multiple people. And, by multiple people, I don't mean a player and a wingman. I mean a cluster of 4+ players focus-firing on a single target.

If we reduce the target engagement time by half, suddenly it is no longer necessary to have three or four teammates backing you up in doing EVERYTHING. Because there will no longer be safety in numbers, this means that one or two mechs is a large enough threat that the enemy will have to respond in some way - either retreat, attack, flank, etc. When the team does not have to dedicate the bulk of its forces to doing any one thing, and when it becomes strategically and tactically advantageous to do multiple things on the battlefield at once instead of just one thing, we end up with a much more interesting game, with much more flanking, charging, baiting, and scouting. The game becomes a lot more casual-friendly when one or two individual players can go forth and whack enemy targets hard enough with any convenient weapons to change the course of a match.

I don't see where any part of that would be bad. That would be a very fun game to play. The only downside is that without respawns, games would likely be over in a few minutes. I personally consider this a very small price to pay in return for granting every player the power to shape the match with every pull of the trigger.

Has anyone else seen how abysmally poorly all the weapons perform in the practice game mode? It feels like our battlemechs are armed with pellet guns, laser pointers, and bottle rockets. The food processor in my kitchen delivers a more satisfying punch when I pull the 'trigger' than this game's PPC's.

Last time I checked, PPC's were supposed to be lightning guns - literal plasma cannons that shoot streams of ionized gas with absolutely massive electrical potential. They certainly don't feel like it.

Edited by Xandralkus, 08 March 2013 - 01:30 AM.


#6 Frater Sender

    Member

  • PipPipPipPipPipPipPip
  • Star Colonel
  • Star Colonel
  • 809 posts
  • Locationdisapered

Posted 08 March 2013 - 04:35 AM

@Xandralkus
There are many things in Tabeltop-Battletech wich fit not for the reality. Also there are many things in MWO wich fit not for Battletech, because Battletech is an adventure story and MWO is like a simulation. Some people belive the ultimate Mechwarriorgame is a 1:1 implementation from Tabeltop. I see Tabeltop as a trial to convert a adventure story to a playable roundbased game. A simulation must be balanced, otherwise than a roundbased game, so a 1:1 implementation is not desirable, because it would not work.
An example:
In reality there are no coincidences, only unknown reasons for events.
In an adventure book there are coincidences in all situations when the hero normaly had no chance to escape, win or something.
In a simulation there is no place for coincidences! Put all away, graphics, effects and only a spreadsheet remains.

Also a simulation is closer to reality than an adventure book, but the simulation has to tell the story, too. Now we need some of this and some of that and call it balance. If there is to much reality the story cant be brought to the user and if there is to much of the story the game turns to unplayable. So please be careful with your wishes, think more "balanced".

thx for your attention


p.s. sorry for my bad english, corrections are welcome

#7 StalaggtIKE

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Raider
  • The Raider
  • 2,304 posts
  • LocationGeorgia, USA

Posted 08 March 2013 - 06:48 AM

View PostXandralkus, on 08 March 2013 - 12:35 AM, said:


This already happens. A lot. Double armor doesn't slow them up, it disproportionately strongly incentivizes that the team do only one thing at a time.

Part of the reason I advocate the removal of double armor is so that the combat performance gap is narrowed between highly coordinated teams and solo PvPers. At the moment, the ONLY way to destroy an enemy mech in any timeframe approaching that of a conventional shooter (a few seconds) is with multiple people. And, by multiple people, I don't mean a player and a wingman. I mean a cluster of 4+ players focus-firing on a single target.

If we reduce the target engagement time by half, suddenly it is no longer necessary to have three or four teammates backing you up in doing EVERYTHING. Because there will no longer be safety in numbers, this means that one or two mechs is a large enough threat that the enemy will have to respond in some way - either retreat, attack, flank, etc. When the team does not have to dedicate the bulk of its forces to doing any one thing, and when it becomes strategically and tactically advantageous to do multiple things on the battlefield at once instead of just one thing, we end up with a much more interesting game, with much more flanking, charging, baiting, and scouting. The game becomes a lot more casual-friendly when one or two individual players can go forth and whack enemy targets hard enough with any convenient weapons to change the course of a match.

I don't see where any part of that would be bad. That would be a very fun game to play. The only downside is that without respawns, games would likely be over in a few minutes. I personally consider this a very small price to pay in return for granting every player the power to shape the match with every pull of the trigger.

Has anyone else seen how abysmally poorly all the weapons perform in the practice game mode? It feels like our battlemechs are armed with pellet guns, laser pointers, and bottle rockets. The food processor in my kitchen delivers a more satisfying punch when I pull the 'trigger' than this game's PPC's.

Last time I checked, PPC's were supposed to be lightning guns - literal plasma cannons that shoot streams of ionized gas with absolutely massive electrical potential. They certainly don't feel like it.

Perhaps you are right. I however still believe that weapon convergence, damage, cool down and ammo count would need to first be addressed before such a drastic change.

Edited by StalaggtIKE, 08 March 2013 - 06:50 AM.


#8 IrrelevantFish

    Member

  • PipPipPipPipPipPip
  • Survivor
  • Survivor
  • 208 posts

Posted 08 March 2013 - 06:56 AM

Halving armor is a terrible, terrible, terrible idea. Yes, focus-firing would become a non-issue (though why you'd want that is beyond me), but so would positioning and tactics. The game would become a matter of who has the biggest alpha and who can fire first.

#9 Xandralkus

    Member

  • PipPipPipPipPipPip
  • 344 posts
  • LocationEarth, for the moment...

Posted 08 March 2013 - 01:29 PM

View PostFrater Interitum, on 08 March 2013 - 04:35 AM, said:

@Xandralkus
There are many things in Tabeltop-Battletech wich fit not for the reality. Also there are many things in MWO wich fit not for Battletech, because Battletech is an adventure story and MWO is like a simulation. Some people belive the ultimate Mechwarriorgame is a 1:1 implementation from Tabeltop. I see Tabeltop as a trial to convert a adventure story to a playable roundbased game. A simulation must be balanced, otherwise than a roundbased game, so a 1:1 implementation is not desirable, because it would not work.
An example:
In reality there are no coincidences, only unknown reasons for events.
In an adventure book there are coincidences in all situations when the hero normaly had no chance to escape, win or something.
In a simulation there is no place for coincidences! Put all away, graphics, effects and only a spreadsheet remains.

Also a simulation is closer to reality than an adventure book, but the simulation has to tell the story, too. Now we need some of this and some of that and call it balance. If there is to much reality the story cant be brought to the user and if there is to much of the story the game turns to unplayable. So please be careful with your wishes, think more "balanced".

thx for your attention


p.s. sorry for my bad english, corrections are welcome


I...am really not sure what you are trying to say. I already advocate fierce deviation from TT canon in the name of fun, player choice, and game balance. You don't bring a walrus to a dog show, and you don't bring tabletop mechanics to a shooter or a simulator.


View PostIrrelevantFish, on 08 March 2013 - 06:56 AM, said:

Halving armor is a terrible, terrible, terrible idea. Yes, focus-firing would become a non-issue (though why you'd want that is beyond me), but so would positioning and tactics. The game would become a matter of who has the biggest alpha and who can fire first.


Currently, the scaling between weapon Alpha-per-Ton and other statistics such as DPS per ton, damage per heat, DPS per heat stable tonnage, and abstracted ease of firepower application is FAR off. A sacrifice of 20% DPS per ton or damage per heat should yield a 5-10% increase in Alpha per ton, not a linear 20% increase - because alpha per ton is SO MUCH MORE USEFUL than the other stats.

This balances high-alpha builds in low-armor situations by taking away their cookies of OPness.


View PostStalaggtIKE, on 08 March 2013 - 06:48 AM, said:

Perhaps you are right. I however still believe that weapon convergence, damage, cool down and ammo count would need to first be addressed before such a drastic change.


I could not agree with you more. Changes to convergence (namely, making torso-mounted weapons non-converging) would buff player fun in a radical and awesome way. Additionally, the 'problem' of boats is solved by granting weapons broad generalist roles with equal parts parity and deviation in derived stats such as alpha per ton, DPS per ton, damage per heat, and stuff like that. You might be interested in a post I made a few months back, with spreadsheets showing idealized weapon balance: http://mwomercs.com/...20#entry1477020 . Imagine a world where Splatcats have no absolute statistical advantage over any canon build.

If the devs made the changes I suggested in my opening post, the game would still need heavy and thorough balancing work. They would not be 'done' by any stretch, just from an hour or two of removing and tweaking stuff. The changes I suggest are only the easiest ones to implement, with a high ratio of game balance created to effort required. There ARE things the devs will need to do that will take weeks, even months of coding, before game balance begins to approach perfection.

They need to do the easy stuff first.

Edited by Xandralkus, 08 March 2013 - 01:50 PM.


#10 Frater Sender

    Member

  • PipPipPipPipPipPipPip
  • Star Colonel
  • Star Colonel
  • 809 posts
  • Locationdisapered

Posted 09 March 2013 - 01:34 PM

@Xandralkus
I ment you are falling in extremes and you have to stay more in balanced zones.
No game wich starts as a adventure book has the whole bandwith from book to the pro-reality-simulation as a working area.
Every leveler you use to move affects the hole game. You want to go closer to the tabeltop-rules so this game will work in a total different way. This may fit to your wantings but there are very much, in future paying, "non-battletech-pro´s" wich want a game with not so much "toughtfulness".
In this regard, especialy americans aren´t very capable of suffering. Anyway, it's pretty tough on the border if it does not fly and has no 3rd personsight.
So dig it in....kick it.... kiss your concept good bye, or what ever you use to call it.

Edited by Frater Interitum, 09 March 2013 - 01:37 PM.


#11 Xandralkus

    Member

  • PipPipPipPipPipPip
  • 344 posts
  • LocationEarth, for the moment...

Posted 09 March 2013 - 10:34 PM

View PostFrater Interitum, on 09 March 2013 - 01:34 PM, said:

@Xandralkus
I ment you are falling in extremes and you have to stay more in balanced zones.


This is about game balance. Anything short of near-perfection is unacceptable. You can call this an unbalanced viewpoint, but we don't need to sacrifice game balance in exchange for something else. Nothing in a multiplayer game is that important.


View PostFrater Interitum, on 09 March 2013 - 01:34 PM, said:

Every leveler you use to move affects the hole game.


I'm well aware of that. Perhaps if we had a 99.9% balanced game, individual precision tweaks to weapon and battlemech behavior might be appropriate. We are not there yet, nor are we anywhere close. It is the very nature of game design for mechanics to interact with other mechanics, but all of those possible permutations and interactions can be calculated and extrapolated. If the devs had actually done this, we would not be in the mess we are in now.

I encourage you to think deeply upon the game mechanic changes I suggest, instead of merely recoiling in shock because it would 'fundamentally and radically alter the game'. I am well aware that it would do that, and I am deeply and intricately aware of exactly how it would affect the game. It would make it a lot more fun.

View PostFrater Interitum, on 09 March 2013 - 01:34 PM, said:

You want to go closer to the tabeltop-rules so this game will work in a total different way.


Actually, I want to do the opposite - I want to fiercely deviate from TT canon, because it is pure lunacy to believe that a tabletop game can somehow transition over to a shooter/simulator. We MUST make sweeping, radical changes to combat if this game is going to attract and retain players. The game is not any 'less' battletech, just because we happen to change some numbers around in a spreadsheet.

However, not every single deviation from TT canon is automatically 'intelligent game design'. Giving mechs the ability to precision-teleport around the battlefield would be a deviation from TT, albeit a ludicrously stupid one. Double armor falls in the same category.

The only people who would actually be 'hurt' by these changes are the hardcore TT fanatics, who would QQ because the mech they designed in tabletop wouldn't be precisely the same in MWO. That crowd is a market niche - a microscopic market niche, incapable of monetizing MWO alone.

Edited by Xandralkus, 09 March 2013 - 10:37 PM.


#12 IrrelevantFish

    Member

  • PipPipPipPipPipPip
  • Survivor
  • Survivor
  • 208 posts

Posted 10 March 2013 - 12:14 AM

Dude, not only are you suggesting that they scrap their entire gameplay system and start from scratch, you're also suggesting that MWO be a Mechwarrior game in name only. That is not only unreasonable, that is just plain wrong. Yes, TT rules need some adapting in order to fit in a real-time format, but just chucking them out entirely would make this the gaming equivalent of the Starship Troopers movie.

And so, even though I intend no disrespect, I can't stop myself from asking that you please please please GO AWAY AND DO NOT POST ON THIS FORUM EVER AGAIN!!!!!!!

Edited by IrrelevantFish, 10 March 2013 - 12:17 AM.


#13 Cyke

    Member

  • PipPipPipPipPipPip
  • 262 posts

Posted 10 March 2013 - 07:44 AM

Doubled armor values and pinpoint accuracy (even from a running 'Mech) are directly tied into one another.

In TT, where weapons were in most cases not accurate enough to choose what enemy location to hit, lower armor values are fine, because the damage will be spread over a few different parts of the enemy 'Mech.

The reason Piranha had to double armor values over TT BattleTech is because of the pinpoint precision of weapons, allowing us to focus all our firepower from lasers, autocannons or PPCs into a single hit location (notably the enemy's center torso).

The only way MWO can half armor values back to TT levels, is if direct fire weapons were made less accurate. This would be downright annoying and would lower the skill ceiling, and I do not agree with this. However, reducing weapon accuracy from a *running* 'Mech sounds all right to me.
Pinpoint accuracy when you're standing still, slight deviation (cone of fire) while you're moving at up to 65% throttle, and worse accuracy (but not too horrible) when your 'Mech is at a dead run.

This also adds an additional layer of tactical decisions to gameplay, if top speed means diminished accuracy.

I want to clarify that this doesn't mean faster 'Mechs have worse accuracy; it will be tied to your throttle setting. 0% throttle for pinpoint accuracy, up to 65% for slight fire cone, 66-100% for the worst fire cone.

I don't agree with doubled armor, but any opinions on reduced accuracy when a 'Mech runs at top speed?

#14 Xandralkus

    Member

  • PipPipPipPipPipPip
  • 344 posts
  • LocationEarth, for the moment...

Posted 10 March 2013 - 10:50 AM

View PostIrrelevantFish, on 10 March 2013 - 12:14 AM, said:

Dude, not only are you suggesting that they scrap their entire gameplay system and start from scratch, you're also suggesting that MWO be a Mechwarrior game in name only. That is not only unreasonable, that is just plain wrong. Yes, TT rules need some adapting in order to fit in a real-time format, but just chucking them out entirely would make this the gaming equivalent of the Starship Troopers movie.

And so, even though I intend no disrespect, I can't stop myself from asking that you please please please GO AWAY AND DO NOT POST ON THIS FORUM EVER AGAIN!!!!!!!


Changing TT into a functioning mechwarrior game is going to require a LOT more than a patch and a bandaid. Dice were at the very core of the game, fundamentally supporting the vast array of mechanics comprising TT. Take that out, and the whole thing collapses and implodes. That is the mess that we have now.

We're not going to find the perfect mechanic to replace the dice - not in a shooter/simulator. What we can do, however, is dismantle and redesign TT around shooter mechanics. Will we have to 'invent' a new game? No, not completely from scratch, but we are going to have to invent - and that's not a bad thing.


View PostCyke, on 10 March 2013 - 07:44 AM, said:

I don't agree with doubled armor, but any opinions on reduced accuracy when a 'Mech runs at top speed?


This is...interesting. Lasers would be hard to do this with though - do the beams merely point off-center, or would the beams constantly shake when moving? Both of these outcomes have their own nuances and imperfections. Off-center stationary beams can be brought 'back' on target, and constantly shaking beams would give lasers a significant disadvantage at damage application at range, when moving quickly.

In order to make sniping while moving a viable tactic (because some players will still want to do it), I'd like to see fully articulated limbs have something like a 50% reduction in speed-induced inaccuracy, compared to non-articulated limbs and torso-mounted weapons.

Edited by Xandralkus, 10 March 2013 - 10:58 AM.


#15 Pht

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,299 posts

Posted 10 March 2013 - 12:54 PM

Maximum return for minimum effort ...

http://mwomercs.com/...different-idea/

Use the combat system in real time minus the parts that simulate the skill of the 'mechwarrior - probably 90% or more of your work would be done for you.

#16 Cyke

    Member

  • PipPipPipPipPipPip
  • 262 posts

Posted 10 March 2013 - 01:58 PM

View PostXandralkus, on 10 March 2013 - 10:50 AM, said:

This is...interesting. Lasers would be hard to do this with though - do the beams merely point off-center, or would the beams constantly shake when moving? Both of these outcomes have their own nuances and imperfections. Off-center stationary beams can be brought 'back' on target, and constantly shaking beams would give lasers a significant disadvantage at damage application at range, when moving quickly.

In order to make sniping while moving a viable tactic (because some players will still want to do it), I'd like to see fully articulated limbs have something like a 50% reduction in speed-induced inaccuracy, compared to non-articulated limbs and torso-mounted weapons.
Hmm, that's a very important point you've brought up that I haven't given any thought. Weapons with their damage spread over a "duration", namely lasers, interact in a complicated manner with non-perfect accuracy.
Perhaps the inaccuracy due to running-induced instability could be constant and cyclic, rather than random? For example, the firing platform simply pitches back and forth in a steady rhythm based on the BattleMech's gait? Perhaps tuned to an appropriate amount, a predictable instability pattern allows skill-based compensation (which is good for competitive play), but no matter how skilled a player is, stopping briefly to take a shot still makes it easier to make a precise shot. Viable?

Also, now that you mention it, I do agree that arms should have better stabilization (compensation for running-induced inaccuracy) compared to torso-mounted weapons. Aside from it being realistic and true to fiction (which are not valid reasons), it also further enhances the value of arm hardpoints over torso hardpoints, and further diverges the value of the various chassis/variants in the game, which is something that I believe adds positive complexity to the game.
It might also slightly reduce the viability of "boating", which seems to be something that is generally perceived as negative.

#17 Xandralkus

    Member

  • PipPipPipPipPipPip
  • 344 posts
  • LocationEarth, for the moment...

Posted 11 March 2013 - 10:51 AM

View PostCyke, on 10 March 2013 - 01:58 PM, said:

Hmm, that's a very important point you've brought up that I haven't given any thought. Weapons with their damage spread over a "duration", namely lasers, interact in a complicated manner with non-perfect accuracy.
Perhaps the inaccuracy due to running-induced instability could be constant and cyclic, rather than random? For example, the firing platform simply pitches back and forth in a steady rhythm based on the BattleMech's gait? Perhaps tuned to an appropriate amount, a predictable instability pattern allows skill-based compensation (which is good for competitive play), but no matter how skilled a player is, stopping briefly to take a shot still makes it easier to make a precise shot. Viable?

Also, now that you mention it, I do agree that arms should have better stabilization (compensation for running-induced inaccuracy) compared to torso-mounted weapons. Aside from it being realistic and true to fiction (which are not valid reasons), it also further enhances the value of arm hardpoints over torso hardpoints, and further diverges the value of the various chassis/variants in the game, which is something that I believe adds positive complexity to the game.
It might also slightly reduce the viability of "boating", which seems to be something that is generally perceived as negative.


Oddly enough, a 'view bob' when moving might work quite well, since moving the view would also move the reticule back and forth in a predictable motion. This also still allows 'pinpoint' accuracy, since the weapons will still go where the reticule points - but the aforementioned hit-and-run sniping maneuvers would become much more skill-intensive. The torso crosshair would stay in the center of the screen, moving with every step of the mech - and the arm reticule would move on the screen, cancelling out half of the movement-induced inaccuracy.

The very limited tracking speed of the torso prevents one from easily 'undoing' the movement-induced inaccuracy, and the fact that the arm reticule would actually move around prevents easy, near-instantaneous acquisition and pinpoint fire.

Limb articulation could even be done on a per-chassis basis. For example, Dragons and Hunchbacks could have a 50% reduction in movement inaccuracy - but a mech like the Cataphract may only have a 35% reduction, since its limbs are much less articulated. The Catapult, Stalker, and YLW arms would not count as articulated - or perhaps they would have vertical articulation, but no horizontal articulation.

Boats seem to rely heavily upon non-articulated limbs...so at least this would serve to reduce its prominence.

Now if we could just get the devs to stop making mechs and MC items for a month or so, and actually fix the game...

Edited by Xandralkus, 11 March 2013 - 10:57 AM.


#18 TheRulesLawyer

    Member

  • PipPipPipPipPipPipPipPip
  • 1,415 posts
  • LocationChicagoland

Posted 11 March 2013 - 01:43 PM

The most important thing about undoing double armor would be restoring class balance. Right now lights are hard to hit. That's fine, but if an assault connects with an alpha it should do bad things. Not just just turn their armor yellow. Plus it screws with the whole long vs short range balance. Yet I know it'll never happen.

#19 Cyke

    Member

  • PipPipPipPipPipPip
  • 262 posts

Posted 11 March 2013 - 01:46 PM

Xandralkus: Having different amounts of arm stabilization on different chassis (that counteracts hypothetical running-induced instability) sounds interesting as well.

Interestingly, if it was done the way you describe, there's already a ready yardstick for how accurate various chassis' arms would be: the presence of upper arm actuator and lower arm joint actuators.
You may have already noticed that some 'Mechs have more free slots available in their arms compared to others, because some 'Mechs have their slots "taken up" by additional actuators.

So it actually comes together very nicely that the 'Mechs with fewer free slots in their arms (due to joint actuators taking up the space) would also have more stable aim for the weapons mounted in those arms.

#20 sC4r

    Member

  • PipPipPipPipPipPip
  • Survivor
  • Survivor
  • 475 posts
  • LocationSlovakia

Posted 12 March 2013 - 12:04 PM

until we figure out how to perfectly balance the game and make everyone happy... remove the game

HELL YEA!!!





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users