Jump to content

A little bit of Math on weapon effectiveness (feel free to discuss) .... Caution wall of text.


  • You cannot reply to this topic
38 replies to this topic

#21 Kartr

    Member

  • PipPipPipPipPipPipPip
  • Knight Errant
  • Knight Errant
  • 560 posts

Posted 13 April 2012 - 03:02 PM

Its a very good look at the numbers part of effectiveness, however I'll have to echo what Orzorn said. Your table doesn't take into account how the damage is applied. Medium lasers have almost double the effectiveness of AC/20s in your graph, but that doesn't account for the fact that 4 medium lasers will do damage to 4 different locations (TT) or do their damage across multiple areas in MWO because of how they apply their damage. AC/20 on the other hand will do all 20 pts of damage to a single location resulting in a much more devastating weapon that takes down enemies faster.

#22 Pyotr

    Rookie

  • 8 posts

Posted 13 April 2012 - 03:18 PM

View PostKartr, on 13 April 2012 - 03:02 PM, said:

Its a very good look at the numbers part of effectiveness, however I'll have to echo what Orzorn said. Your table doesn't take into account how the damage is applied. Medium lasers have almost double the effectiveness of AC/20s in your graph, but that doesn't account for the fact that 4 medium lasers will do damage to 4 different locations (TT) or do their damage across multiple areas in MWO because of how they apply their damage. AC/20 on the other hand will do all 20 pts of damage to a single location resulting in a much more devastating weapon that takes down enemies faster.


A good way to account for that may be to also give the probable damage to, say, the CT or head, as well as standard deviations from this (which would be different for different firing rates, if I remember my statistics correctly).

#23 Hollister

    Member

  • PipPipPipPipPipPip
  • The Nimble
  • The Nimble
  • 321 posts
  • LocationFlorida

Posted 13 April 2012 - 03:26 PM

View PostMotionless, on 13 April 2012 - 02:52 PM, said:

I think ponies have rotted your brains out.


Ponies do not rot brains, they enlighten ones self to love and tolerance.

If the mechlab truly offers custom paint schemes. Love and Tolerance will be the names i put on the barrels of my A/C's to remind me that I am purifing the evils of the Innersphere with Righteous Hellfire.

Edited by Hollister, 13 April 2012 - 03:27 PM.


#24 Nighthound

    Member

  • PipPipPipPipPip
  • Stone Cold
  • Stone Cold
  • 146 posts
  • LocationGermany - Düsseldorf

Posted 13 April 2012 - 03:30 PM

Here you go.
Posted Image

I had to turn around Damage per C-Bill (C-Bills per Damage) because those were some ridiculously small numbers.

#25 Gun Bear

    Member

  • PipPipPipPipPipPipPipPip
  • 1,016 posts
  • LocationGarrison duty on some FWL Planet and itching for action.

Posted 13 April 2012 - 03:38 PM

I guess knowing the math has its advantages but I still prefer knowing where to shoot.

#26 Hollister

    Member

  • PipPipPipPipPipPip
  • The Nimble
  • The Nimble
  • 321 posts
  • LocationFlorida

Posted 13 April 2012 - 03:46 PM

I can not seem to remember if they have said anything about needing to buy ammo with C-Bills. I would assume you would need to. Which if that is the case, missiles seem like a real money pit at the moment to keep stocked. I would also assume all unused ammo after a battle would be kept and put back into your supply for later use.

Edited by Hollister, 13 April 2012 - 04:40 PM.


#27 Nighthound

    Member

  • PipPipPipPipPip
  • Stone Cold
  • Stone Cold
  • 146 posts
  • LocationGermany - Düsseldorf

Posted 13 April 2012 - 03:50 PM

View PostOrzorn, on 13 April 2012 - 02:13 PM, said:

This table provides a good general idea of weapon effectiveness, however, parts of it do, of course, fall flat.

One major issue is that the model does not distinguish between front loaded damage and split damage. ACs do front loaded damage (although, depending on how they implement fire rates, some will do DPS, while others will do burst), lasers (at least MWO's current implementation) are damage-over-time weapons (they do their damage in minute amounts, probably less than one damage, but many of those "ticks" happen within a single second, over time), and LRMs are split-damage (that is, they do their total damage in packets that may or may not all connect, meaning a missile weapon does anywhere from 0 to its rated value in damage, depending on how many missiles actually hit).

Regardless, while those concepts are important to keep in mind when interpreting the data, I don't think the model itself needs to make any sort of distinctions.

View PostKartr, on 13 April 2012 - 03:02 PM, said:

Its a very good look at the numbers part of effectiveness, however I'll have to echo what Orzorn said. Your table doesn't take into account how the damage is applied. Medium lasers have almost double the effectiveness of AC/20s in your graph, but that doesn't account for the fact that 4 medium lasers will do damage to 4 different locations (TT) or do their damage across multiple areas in MWO because of how they apply their damage. AC/20 on the other hand will do all 20 pts of damage to a single location resulting in a much more devastating weapon that takes down enemies faster.


Yes, I know and I believe I stated as much in my Overviews .... those are pros and cons and tastes .... very hard (read impossible) to quantify that stuff to put it into a formula!

View PostPyotr, on 13 April 2012 - 03:18 PM, said:

A good way to account for that may be to also give the probable damage to, say, the CT or head, as well as standard deviations from this (which would be different for different firing rates, if I remember my statistics correctly).


Something like this?
Head 3%
Arm 14%
Leg 11%
Side Torso 14%
Center Torso 17%
TAC 3%

#28 Yeach

    Member

  • PipPipPipPipPipPipPipPip
  • 1,080 posts

Posted 13 April 2012 - 04:44 PM

Why are we using Solaris 7 dueling rules?

from Sarna net
http://www.sarna.net...Mech_Duel_Rules
Game Experience
The higher level of detail offers more tactical options but the game gets more complex and time consuming as well. Consequently, it is typically only suitable for fights between very small numbers of combattants (usually 1 on 1).
Because of the modified heat recording system and the limitation on fewer actions per turn, players must consider their moves very well. They have to decide when the time is right to fire, for the 'Mechs run a lot hotter and the respective weapons may not be used in the following turn(s). In addition, the scale shift and quadrupling of the weapon ranges (in game hexes) ensures that weapons with a long [color="#ba0000"]minimum range[/color], such as LRMs which have a minimum range of 24 hexes under these rules, are difficult to use reasonably.

Together this leads to a distinctively different performance of some weapons and of whole BattleMech designs. Many designs that shine in one ruleset perform poorly in the other.

This further muddies with the regular Battletech rules.

#29 Nighthound

    Member

  • PipPipPipPipPip
  • Stone Cold
  • Stone Cold
  • 146 posts
  • LocationGermany - Düsseldorf

Posted 13 April 2012 - 06:06 PM

View PostYeach, on 13 April 2012 - 04:44 PM, said:

Why are we using Solaris 7 dueling rules?


Actually we are not! I just used the Solaris VII recycle times. If I had used all the rules I would have needed to quadruple heat and range (at least the hex values). And I would think that it makes a lot of sense since I (and I bet most) don't want to pilot some giant robot with all weapons on a 10 second recharge timer.


Edit : But you have a Point .... I think I miscalculated the heat part of my CPs .... have to mull that over a bit.

Edited by Nighthound, 13 April 2012 - 06:26 PM.


#30 GrimFist

    Member

  • PipPipPipPipPip
  • 167 posts
  • Google+: Link
  • LocationAlter-Ego - Death Watch Warship - Retribution - Ageis class (M)

Posted 13 April 2012 - 06:11 PM

OMG.. the numbers... the numbers.. ahhh...

If I did not have a PMP I'd be resorting to violence right now.


Grim indeed!

#31 Kyle Hawkins

    Member

  • PipPipPipPipPip
  • 122 posts
  • LocationSearching for Gensokyo

Posted 13 April 2012 - 06:48 PM

When it comes to taking these stats into a first person sim, you do have to remember of course the innumerable little variances that will come up. Concepts such as the "window of opportunity" all come into play. Compare the pulse lasers and the large lasers in MW: Living Legends for example. The pulse lasers will overall give you more damage per second, but at the expense of forcing you to keep on accurately on target the entire time you wish to use them. Any aim rocking, attempts at evasive maneuvers, or interrupting terrian, will all reduce the effectiveness of such a weapon. You are also forced to expose yourself to the enemy the entire time you wish to deal damage. For a slow fire weapon, whether LL, PPC, or AC20, you can pick your moment of shot, timing it in-between being rocked by return fire, deal the damage, and then duck back behind cover or make any other maneuver you need whilst staying comparably safe from retaliation. If one of those high DPS knife fighters gets you in a bad spot however, you are likely out gunned and in trouble.

All the same, I'll definitely be taking an interest in these kinds of charts once the actual game is released. It can be surprising the small insights they can reveal.

Battlefield3 charts for example. A 9mm pistols 32 damage compared to the .45 pistols 34 damage and halved ammo capacity might seem a worthless difference, but that extra 2 damage is the difference between a kill in 3 shots or 4 shots.

#32 Nighthound

    Member

  • PipPipPipPipPip
  • Stone Cold
  • Stone Cold
  • 146 posts
  • LocationGermany - Düsseldorf

Posted 13 April 2012 - 07:05 PM

Ok, thanks to Yeach, I spoted a rather big mistake .... I calculated the heat part of my CPs with normal CBT rules while calculating DpS with the Solaris VII recharge times and it threw my whole calculation a little bit off.

So now the revised calculations (Links only for Condensed and Short Range Sheets)
Posted Image

Aegis Kleais Diagram:
Posted Image

Sorry for that .. as you can see some has changed (the medium laser and the UACs are more in line now) but not that much so my initial overviews should still be valid.

#33 BarHaid

    Member

  • PipPipPipPipPipPipPipPip
  • Urban Commando
  • Urban Commando
  • 1,072 posts
  • LocationMid-Cascadia

Posted 13 April 2012 - 10:18 PM

I have nothing constructive to add here; I am just really loving the graphs and math!

#34 EDMW CSN

    Member

  • PipPipPipPipPipPipPipPip
  • Knight Errant
  • Knight Errant
  • 1,073 posts
  • LocationOutreach

Posted 13 April 2012 - 11:19 PM

AC20 should be in the range of at least 5 dps while UACs can bring it up to 10 DPS :o

#35 Logan Winters

    Member

  • PipPip
  • 45 posts
  • LocationColorado

Posted 13 April 2012 - 11:42 PM

Holy Crap! This is the coolest thread I have ever seen!

And with the most accommodating poster as well!


I wanna see damage per second..

Boom! (chart)

I wanna see heat vs damage per second..

Boom! (chart)

I wanna see damage vs heat vs c-bill cost per second..

Boom! (big chart)

Can we get all that in color?

Boom!

Reading this pretty much made my day haha.

and wow.. Informative..

Just another reason I love battletech fans. :o

#36 Nighthound

    Member

  • PipPipPipPipPip
  • Stone Cold
  • Stone Cold
  • 146 posts
  • LocationGermany - Düsseldorf

Posted 14 April 2012 - 07:03 AM

Thanks ! :o

#37 Tuhalu

    Member

  • PipPipPipPipPipPip
  • 250 posts

Posted 14 April 2012 - 07:20 AM

I disagree with your use of range in the formula.

Try adjusting the formula to (1+Range/3)*DPS/CP

This results in range modifiers anywhere from 2 for a small laser/flamer/MG to 10 for an LB-X/2. Rather than anywhere between 3 and 27. This should reduce the value of extreme range weapons somewhat, while improving the value of closer range weapons. This should help reflect that most extreme range weapons have minimum range problems and the diminishing value of range. Because you don't need super long range on your weapons, just more range than the other guy!

Additionally, I'd like to see the numbers assuming Rules of Warfare recycle times (ie. 10 seconds).

Edited by Tuhalu, 14 April 2012 - 07:22 AM.


#38 Mr MEAN

    Member

  • PipPip
  • 29 posts
  • LocationBehind You

Posted 14 April 2012 - 07:52 AM

You Analyze The Numbers
And Make Informed Decisions
I Turn the Corner
And Blast Your Leg Off
'Cause That's MEAN

#39 Pyotr

    Rookie

  • 8 posts

Posted 14 April 2012 - 09:51 AM

View PostNighthound, on 13 April 2012 - 03:50 PM, said:



Something like this?
Head 3%
Arm 14%
Leg 11%
Side Torso 14%
Center Torso 17%
TAC 3%

Yes, but recall that, for firing 4 times, the probability of hitting the CT at least once is 1-(1-P)^N, where P is the probability of hitting the CT and N is the number of times you fire, so the probability of doing ANY damage to the CT in a 10 second set of four rounds with, say, an AC/2 would be 1-(1-P)^4 = 47.5%, while, with an AC/20, the probability would be 1-(1-P) = 17%. However, the probability of the AC/2 doing all of its damage to the CT would be given instead by: P^4 = 0.08%, which is simply not a very good chance.

Keep in mind that this assumes I remember my statistics course right, which may not be the case.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users