Jump to content

A Way To Balance Weapons And Armor


31 replies to this topic

#1 Tombstoner

    Member

  • PipPipPipPipPipPipPipPipPip
  • Bridesmaid
  • Bridesmaid
  • 2,193 posts

Posted 30 January 2014 - 03:32 PM

O boy another balance thread.... glad someone had the courage to...... First...

Seriously though i have always felt the adaptation from TT to PC remains problematic and lacks one or two degrees of freedom needed to tweak the game.

First is an armor penetration co-factor for each weapon. by default all weapons have a co-factor of zero penetration . i think it would be relative easy to add a penetration value to some weapons. This lets PGI provide more substance to weapon choice and the ability to disrupt undesirable combinations in the context of pinpoint accuracy.

For example the **** laser suffers from too much similarity with regular lasers. if the pulse variant has a penetration value of .2 then 20% of the damage would penetrate the armor and hit internals and maybe cause a critical hit. the exact value of the co-factor to be determined later. since .2 may be too much.

Auto cannons and ppcs would be zero since all the damage is front loaded. lasers having some degree of penetration would make them more desirable since there damage is spread out and typically generate high amounts of heat.

LRM's would be left alone, but maybe give srms a co-factor or .05

The second idea is to give hit locations resistances to penetration or direct damage reduction to specific types of weapons. this may be a bit early in the time line for glazier or reactive armor. but it would be a way to make the CT more durable and disincentives that location.

Lets face this one simple fact - mech designers would never have made such a wide range of mech morphologies if it mattered. under TT rules it didn't. in a skill based game it does and PGI needs something other then heat/ damage to tweak. mech designers would never have made arms or torsos such a large % of the over all design if the CT was so easy to hit. It would flat out be a few meters thick slab of armor with little t-rex arms or 100 tone Urbanmech's on steroids.

Penetration of some damage might posably be a way to make a viable critical hit system.
Damage resistance to penetration would incentives shooting arms/legs torso's first.
Damage reduction by location would also serve the same function of incentivising. it would also increase TTK.

#2 Khobai

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 23,969 posts

Posted 30 January 2014 - 03:56 PM

Weapons ignoring armor would be way too strong unless you buffed internal structure values and item health considerably.

#3 Tombstoner

    Member

  • PipPipPipPipPipPipPipPipPip
  • Bridesmaid
  • Bridesmaid
  • 2,193 posts

Posted 31 January 2014 - 07:04 AM

View PostKhobai, on 30 January 2014 - 03:56 PM, said:

Weapons ignoring armor would be way too strong unless you buffed internal structure values and item health considerably.

not really since the entire system is ablative. first you destroy the armor then the internals. The only time its an issue is when you have more armor points then internal. then penetration could conceivably destroy the section before the armor is gone. even then the damage penetration would need to be rather high for it to be an issue. i'm thinking along the lines of no more then 20% for the best penetrating weapon. that would be the Lg pulse laser or base damage of 9 becomes 7.2 to the armor and 1.8 to internals. most pieces of equipment have 10 hp. again this can be dialed down to .01.

#4 3rdworld

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,562 posts

Posted 31 January 2014 - 07:06 AM

It could work. But this like many other proposed systems would require a complete re-balance of the weapons.

We can't get them to buff Pulse lasers now.

#5 Khobai

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 23,969 posts

Posted 31 January 2014 - 07:10 AM

Quote

not really since the entire system is ablative.


yes really. it would be extremely overpowered since mechs already die too fast and you want them to die faster.

#6 Tombstoner

    Member

  • PipPipPipPipPipPipPipPipPip
  • Bridesmaid
  • Bridesmaid
  • 2,193 posts

Posted 31 January 2014 - 08:34 AM

View PostKhobai, on 31 January 2014 - 07:10 AM, said:


yes really. it would be extremely overpowered since mechs already die too fast and you want them to die faster.

No the time to kill is the same unless you hit ammo and that chance is covered by the critical hit system, also ammo has HP so its not a guaranteed explosion.
.
No mech has less internal points then armor. you cant kill a mech faster this way. unless you apply bonus damage to internal hits. that should be removed. it has no place in this game. PGI added it to create some sort of critical seeking weapon category that never worked well. its also why mg's eat you alive when your armors gone. That system is broken.

Please explain how a mech CT with 50 armor and 50 internal dies faster when hit with 10x 1 ppc shots @ 10% pen, then 10 ppc's with no penetration. answer it dies at the same rate. as long as the damage penetration is less then/ = 50%. i'm talking about valuse of 0 to 25% with the removal of bonus damage to internal hits. if anything it would make mechs die slower... imperceptibly slower via a small nudge in player behavior.

#7 Durant Carlyle

    Member

  • PipPipPipPipPipPipPipPipPip
  • Survivor
  • Survivor
  • 3,877 posts
  • LocationClose enough to poke you with a stick.

Posted 01 February 2014 - 01:38 AM

View PostTombstoner, on 31 January 2014 - 08:34 AM, said:

No mech has less internal points then armor.

This is where you are wrong.

Default tabletop values meant that max armor for a location was double the internal structure value.

Both internal structure and armor values have been doubled in MW:O (I know armor has, and I think internal structure has too). That means that internal structure is still only half as strong as maxed armor values.

Your armor penetration idea could easily result in games where lots of 'Mechs die without being stripped of armor.

#8 Joseph Mallan

    ForumWarrior

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • FP Veteran - Beta 1
  • FP Veteran - Beta 1
  • 35,216 posts
  • Google+: Link
  • Facebook: Link
  • LocationMallanhold, Furillo

Posted 01 February 2014 - 06:20 AM

View PostDurant Carlyle, on 01 February 2014 - 01:38 AM, said:

This is where you are wrong.

Default tabletop values meant that max armor for a location was double the internal structure value.

Both internal structure and armor values have been doubled in MW:O (I know armor has, and I think internal structure has too). That means that internal structure is still only half as strong as maxed armor values.

Your armor penetration idea could easily result in games where lots of 'Mechs die without being stripped of armor.

Define LOTS, TAC was a rare occasion on TT and if the chance to score a TAC here was set at 1%... Would make it happen much less often.

#9 Mr 144

    Member

  • PipPipPipPipPipPipPipPip
  • 1,777 posts
  • LocationWisconsin

Posted 01 February 2014 - 07:28 AM

View PostTombstoner, on 31 January 2014 - 08:34 AM, said:

No mech has less internal points then armor


Internal Structure HP = 1/2 of maximum armor value for that location.

Internal structure HP does not change. It is always based on the maximum armor points available.

The sole exception is the head, which has 15 internal HP as opposed to 9 (18 max available armor).

#10 Khobai

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 23,969 posts

Posted 01 February 2014 - 08:19 AM

Quote

No mech has less internal points then armor.

Uh all mechs have less internal points than armor. Mechs have half the internal structure as they do max armor. Hence why I said you'd have to increase internal structure considerably for armor penetration to work.

#11 Levi Porphyrogenitus

    Member

  • PipPipPipPipPipPipPipPipPip
  • Mercenary
  • Mercenary
  • 4,763 posts
  • LocationAurora, Indiana, USA, North America, Earth, Sol, Milky Way

Posted 01 February 2014 - 08:23 AM

TtK is way too fast right now, and you want to add more ways to speed things up?

And then you want to band-aid your first change by adding a hokey DR mechanic to CTs only?

TtK is more a function of weapon cycle times (double, triple, or more of TT values) and perfectly precise group fire than it is anything else. PGI shouldn't change any other mechanics until they do something to extend TtK, preferably by doing a cooldown pass on most weapons and adding circumstantial penalties to precision weapon fire.

Plus, TACs are yet another element of RNG, and RNG is typically bad game design, especially in something that has hopes to becomes an e-sport in the future. As for guaranteed armor penetration, save that for special munitions in the future (AP ammo, for instance). Special armor types as well. They should come eventually, and in the mean time we don't need to add half-measures that break the current balancy system.

#12 Joseph Mallan

    ForumWarrior

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • FP Veteran - Beta 1
  • FP Veteran - Beta 1
  • 35,216 posts
  • Google+: Link
  • Facebook: Link
  • LocationMallanhold, Furillo

Posted 01 February 2014 - 08:24 AM

I don't think it is to fast Levi, in fact from my computer chair it takes to long.

#13 Joseph Mallan

    ForumWarrior

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • FP Veteran - Beta 1
  • FP Veteran - Beta 1
  • 35,216 posts
  • Google+: Link
  • Facebook: Link
  • LocationMallanhold, Furillo

Posted 01 February 2014 - 08:32 AM

View PostKhobai, on 01 February 2014 - 08:27 AM, said:

TTK is definitely too fast. You should not be able to kill an Atlas with a Jagermech in less than 20 seconds. No one in their right mind would build a multi-billion dollar mech that could be destroyed that easily.

They are multi million dollar war machines and we do it right now. F-22 Raptor $150,000,000.00 per jet! Abrams 4.3 million per tank.

Edited by Joseph Mallan, 01 February 2014 - 08:37 AM.


#14 Khobai

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 23,969 posts

Posted 01 February 2014 - 09:32 AM

Quote

They are multi million dollar war machines and we do it right now. F-22 Raptor $150,000,000.00 per jet! Abrams 4.3 million per tank.


1) Youre not building an Atlas for less than a billion dollars in today's currency. And no military in their right mind would want an upright combat vehicle that can be instantly destroyed. The only way it's viable is if you have insanely strong composites and armor technology, like Battletech does. MWO just doesnt accurately represent the survivability of an Atlas.

2) the airforce refuses to deploy F-22s in real combat, because theyre like $300,000,000 each, and are so flimsy and unarmored that they can be shot down by a .50 cal machine gun. The F-22 was a complete failure which is why the entire program was cancelled.

Edited by Khobai, 01 February 2014 - 09:36 AM.


#15 Joseph Mallan

    ForumWarrior

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • FP Veteran - Beta 1
  • FP Veteran - Beta 1
  • 35,216 posts
  • Google+: Link
  • Facebook: Link
  • LocationMallanhold, Furillo

Posted 01 February 2014 - 09:36 AM

View PostKhobai, on 01 February 2014 - 09:32 AM, said:


1) youre not building an Atlas for less than a billion dollars in today's currency.

2) the airforce refuses to deploy F-22s in real combat, because theyre like $300,000,000 each, and are so flimsy and unarmored that they can be shot down by a .50 cal machine gun. The F-22 was a complete failure which is why the entire program was cancelled.
B-2 Stealth Bomber 737 Million dollars.
F-14 38 million
Bradley 5.6 Million (compared to a Spiders 2.9 million)

The Abrams(4.6 Mil) is on par with a 65 ton Thunderbolt's 5.4 million C-bill price tag.

Edited by Joseph Mallan, 01 February 2014 - 09:41 AM.


#16 Khobai

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 23,969 posts

Posted 01 February 2014 - 09:41 AM

Quote

The Abrams(4.6 Mil) is on par with a 65 ton Thunderbolt's 5.4 million C-bill price tag.


Dollars and Cbills arnt the same currency. Building a mech costs WAY more than a tank. Darpa spent $10,000,000 on their new version of the Big Dog mech which is a fraction of the size of the mechs in battletech.

Quote

B-2 Stealth Bomber 737 Million dollars.


The B-2 isnt used in everyday combat. Its a strategic weapon thats held in reserve. The Atlas is a frontline combat machine... no one would deploy a frontline combat machine that can be destroyed in seconds that costs hundreds of millions or even billions to produce/develop. The US has tons of expensive toys but we rarely if ever actually use them.

The point is, it doesnt make a whole lot of sense for Atlases to get punked in 15 seconds by a Jagermech. If thats the kindve nonsense thats been going on in the Inner Sphere its no wonder theyre living in the dark ages.

Edited by Khobai, 01 February 2014 - 09:46 AM.


#17 Joseph Mallan

    ForumWarrior

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • FP Veteran - Beta 1
  • FP Veteran - Beta 1
  • 35,216 posts
  • Google+: Link
  • Facebook: Link
  • LocationMallanhold, Furillo

Posted 01 February 2014 - 09:47 AM

View PostKhobai, on 01 February 2014 - 09:41 AM, said:


Dollars and Cbills arnt the same currency. Building a mech costs WAY more than a tank. Darpa spent $10,000,000 on their new version of the Big Dog mech which is a fraction of the size of the mechs in battletech.

Who's Darpa? So using present day manufacturing you are trying to justify your position? Currency is currency, unless you can find a Official C-bill=X dollars. Did yo know that a prototype engine for Fords can cost hundreds of thousands of dollars if not millions to develop but end up costing around $2,000 per production engine.

#18 Khobai

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 23,969 posts

Posted 01 February 2014 - 09:49 AM

Quote

Who's Darpa? So using present day manufacturing you are trying to justify your position?


DARPA is the US department of defense thinktank. They develop a lot of the next-generation weapons for the US military.

Quote

Did yo know that a prototype engine for Fords can cost hundreds of thousands of dollars if not millions to develop but end up costing around $2,000 per production engine.


Absolutely. Even if mech production was refined to an artform, its still going to cost the equivalent hundreds of millions if not billions to produce an Atlas. Its magnitudes more complex than a tank which is still essentially just a box on wheels with a gun on top. The only way mechs could work as viable combat vehicles is if you had insanely strong armor to justify them having such exposed profiles.

Its really dumb to see an Atlas roll into a fight and die 1v1 to a Jagermech or Cataphract less than 20 seconds later. Whats the point of an Atlas if it cant take a decent amount of hits?

Edited by Khobai, 01 February 2014 - 10:00 AM.


#19 mania3c

    Member

  • PipPipPipPipPipPip
  • The Scythe
  • 466 posts

Posted 01 February 2014 - 09:50 AM

if we are really talking about "real" things..noone would ever spend penny to make walker war machine.. any same sized tank would steamroll any walkers no problem..just saying..

Edited by mania3c, 01 February 2014 - 09:51 AM.


#20 Joseph Mallan

    ForumWarrior

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • FP Veteran - Beta 1
  • FP Veteran - Beta 1
  • 35,216 posts
  • Google+: Link
  • Facebook: Link
  • LocationMallanhold, Furillo

Posted 01 February 2014 - 10:07 AM

View PostKhobai, on 01 February 2014 - 09:49 AM, said:


DARPA is the US department of defense thinktank. They develop a lot of the next-generation weapons for the US military.



Absolutely. Even if mech production was refined to an artform, its still going to costs hundreds of millions if not billions to produce an Atlas. Its magnitudes more complex than a tank which is still essentially just a box on wheels with a gun on top. And in most cases a tank is superior by today's standards. The only way mechs could work as a viable combat vehicle is if you had insanely strong armor to justify them having such exposed profiles.

Prototype Car costs in the ballpark of 10x the market value of a production model. It will costs hundreds of millions to billions to develop but not to produce. I work in prototype Khob, production costs are way less than prototype. It takes us 6-10 hours to make a prototype Camshaft at $33 an hour for the man hours, an I don't know how much our salary inspectors make. So one cam runs between $800-$1000 per cam. You can buy one at AutoZone for abut $200!





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users