Jump to content

State Of Weapon Balance - 2012-11-25 (Shs Vs Dhs, With Graphs)

v1.0.150

88 replies to this topic

#41 MustrumRidcully

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 10,644 posts

Posted 26 November 2012 - 11:33 PM

View PostZyllos, on 26 November 2012 - 02:05 PM, said:

I am 100% in the boat that convergence is why everything is out of wack.

Lasers are easy to choose locations to hit with. And they make hitting fast/small targets much easier. While I agree that the laser duration was ment to fix this by having to keep the weapon on target, the issue is that this can partly be circumvented with enough lasers firing at the same point.

Projectile weapons are farely hard to hit with, at least against moving targets or yourself moving. Now, if you hit, they deal their entire damage to that location but requires a lot of work to get them to land on the target, much less on a single location. Convergence does not fix this issue due to either having to lead targets to land shots or just completely missing.

Now on top of this, covergence allows small array of weapons to act EXACTLY like larger weapons, dealing all their damage to a single location on each fire.

Another problem that is presented here, the range advantage of weapons. Not so much the damage dealt at optimum range but how much you pay for that extra range. Paying an extra 3 heat per firing of the ER LL is not worth the extra 225m. This is because to conversion from turn base to real time. There are so many intricacies that is generalized or flat out ignored in the CBT turn.

Actually, this has less to do with its turn based nature, and more with its hex-based nature and how they determiend to-hit difficulty. Firing an ER Laser at the range of a Large Laser would mean the ER LL user had a better hit chance - the ER LLs Medium Range (+1 to difficulty to hit) is almost the same as the LLs Long Range (+2 difficulty to hit). The difference between roling 2d6 vs a difficulty of 5 or a difficulty of 6 is about 11 % (for an average pilot). So an ER LL might on average deal 11 % more damage than the LL - and it can fire at a further range as well. That may be worth 3 extra heat.

But this is not how hit-probabilities work in real life, or in MW:O with mouse aiming. It doesn't matter whether you fire an LL or an ER LL at 500m - both is equally dificult. Range advantages are overvalued in MW:O and the table top values don't fit the (simulated) realities of the game.

#42 Kmieciu

    Member

  • PipPipPipPipPipPipPipPipPip
  • Urban Commando
  • Urban Commando
  • 3,437 posts
  • LocationPoland

Posted 27 November 2012 - 12:27 AM

I think the flawed weapon convergence is THE reason for boating smaller weapons.
A Jenner should not be able to hit the Commando with all 6 lasers - it lacks lower arm actuators.
A Commando should be able to hit a Jenner with all of its arm mounted weapons.
A Jenner with a large laser would not be a joke anymore, if it were the only one able to do pinpoint damage.

Think about it.

In chain fire mode aiming the aiming reticle would show the correct point of impact while for group fire we would see a circe depending on the range to the locked target. It would even discourage constant alpha striking with medium lasers from 300 meters away.

Edited by Kmieciu, 27 November 2012 - 12:28 AM.


#43 shabowie

    Member

  • PipPipPipPipPipPipPip
  • 877 posts

Posted 27 November 2012 - 02:02 AM

Errr, I couldn't understand some of your graphs because it says axis title as a placeholder.

Edited by shabowie, 27 November 2012 - 02:07 AM.


#44 MustrumRidcully

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 10,644 posts

Posted 27 November 2012 - 02:20 AM

View Postshabowie, on 27 November 2012 - 02:02 AM, said:

Errr, I couldn't understand some of your graphs because it says axis title as a placeholder.

If only there was a way to teach Excel to keep those damn titles around.

The vertical axis lists the efficiency (damage dealt in the targeted time divided by tonnage to run it), the horiziontal axis obviously lists the weapons (within the categories of ballistic, energy and missiles, ordered by range):

Edited by MustrumRidcully, 27 November 2012 - 02:23 AM.


#45 Tuhalu

    Member

  • PipPipPipPipPipPip
  • 250 posts

Posted 27 November 2012 - 02:36 AM

View PostMustrumRidcully, on 26 November 2012 - 11:33 PM, said:

Actually, this has less to do with its turn based nature, and more with its hex-based nature and how they determiend to-hit difficulty. Firing an ER Laser at the range of a Large Laser would mean the ER LL user had a better hit chance - the ER LLs Medium Range (+1 to difficulty to hit) is almost the same as the LLs Long Range (+2 difficulty to hit). The difference between roling 2d6 vs a difficulty of 5 or a difficulty of 6 is about 11 % (for an average pilot). So an ER LL might on average deal 11 % more damage than the LL - and it can fire at a further range as well. That may be worth 3 extra heat.

But this is not how hit-probabilities work in real life, or in MW:O with mouse aiming. It doesn't matter whether you fire an LL or an ER LL at 500m - both is equally dificult. Range advantages are overvalued in MW:O and the table top values don't fit the (simulated) realities of the game.

Actually, the hit modifiers for range are +2 at medium range and +4 at long range. So the difference is even bigger that you calculated. The general idea is correct though :P

#46 MustrumRidcully

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 10,644 posts

Posted 27 November 2012 - 02:42 AM

View PostTuhalu, on 27 November 2012 - 02:36 AM, said:

Actually, the hit modifiers for range are +2 at medium range and +4 at long range. So the difference is even bigger that you calculated. The general idea is correct though :P

Ah, thanks for the correction.

Heck, I couldn't even remember if I needed to roll under or roll over anymore.

I am not really a BT player. 1-2 rounds of Battletech on the board, and maybe two dozen Megamek rounds. Battletech rules seem pretty time-consuming to play (roll to hit and roll hit location for every weapon or every cluster)... Megamek avoids the time-consuming issue. But the nature of combat seemed too random to me "Oh, look, a lucky head shot..."

#47 Tuhalu

    Member

  • PipPipPipPipPipPip
  • 250 posts

Posted 27 November 2012 - 02:59 AM

View PostMustrumRidcully, on 27 November 2012 - 02:42 AM, said:

Battletech rules seem pretty time-consuming to play (roll to hit and roll hit location for every weapon or every cluster)... Megamek avoids the time-consuming issue. But the nature of combat seemed too random to me "Oh, look, a lucky head shot..."

I hardly touched Megamek myself because a lot of the charm of randomness is lost when you can't feel the dice and agonize over every roll. MWO replaces most of the randomness with twitch-reflex requirements. Both have their own unique adrenline rush.

Now for a comment about the chart....

Pulse Lasers are not underpowered, you are simply under-rating their efficiency. A Pulse Laser puts out its damage over a period 0.25 seconds shorter than a weapon of comparable size. This reduces laser spread against moving targets considerably and helps ensure that all of the damage hits the target. This creates a situation that your chart cannot account for. Pulse Lasers are efficient compared to regular lasers against fast/dodgy targets (or while moving fast/dodgily!) and unefficient compared to regular lasers against very slow or immobile targets.

Most weapons have a continuum of efficiency/accuracy depending on the relative speeds of the mechs involved, but pulse lasers (and SSRMs) fall off in efficiency less than other weapons.

Edited by Tuhalu, 27 November 2012 - 03:31 AM.


#48 197mmCannon

    Member

  • PipPipPipPipPipPip
  • Go-cho
  • Go-cho
  • 265 posts
  • LocationCincinnati, OH

Posted 27 November 2012 - 05:49 AM

This chart is very well constructed but is missing two pieces of crucial piece of information.

1. Humans.

How often do you have a situation where two Mechs are standing still blasting at each other's CT? When your ducking in and out of cover or doing the stupid circle strafe dance that's so common now your not hitting at max rof. In some of these situations the slower firing weapon is more efficient.

2. Hardpoints and tonnage.

If your running a LaserBack with 9 laser hardpoints (or something like that) then yeah you want smalls or mediums. But what if you only have 4 ? Well if your an atlas with extra tonnage and crit space then you probably want larges.

Edited by Daemian, 27 November 2012 - 05:52 AM.


#49 EmperorMyrf

    Member

  • PipPipPipPipPipPipPip
  • The Howl
  • The Howl
  • 740 posts
  • LocationMinnesota, USA

Posted 27 November 2012 - 06:07 AM

So I had a thread about convergence in a previous patch that didn't get too much attention. I'll post the good bits here, but basically what I think should happen is that torso mounted weapons should be limited by an angle of 0.25 degrees.

The following was calculated using the Hunch-4P as both the firer and the target.

Posted Image

"Key". Forgot to label that the box represents the target armor location:
Posted Image
Purple is the reticle, blue is the centroid of the laser impacts, and green is the lasers.

I don't think that convergence should be removed from the game, which apparently a lot of people are trying to push for. I just feel that it needs to be limited at short range so that smaller weapons can't have their naturally high efficiencies exploited.

Edited by EmperorMyrf, 27 November 2012 - 06:18 AM.


#50 MustrumRidcully

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 10,644 posts

Posted 27 November 2012 - 06:18 AM

View PostDaemian, on 27 November 2012 - 05:49 AM, said:

This chart is very well constructed but is missing two pieces of crucial piece of information.

1. Humans.

How often do you have a situation where two Mechs are standing still blasting at each other's CT? When your ducking in and out of cover or doing the stupid circle strafe dance that's so common now your not hitting at max rof. In some of these situations the slower firing weapon is more efficient.

That's why I generally say that slow-firing weapons have an advantage - if you can't hit the trigger every time the weapon is ready, you lose less damage that way. Of course, otherws will bring up that missing means you lose more damage. Is it a wash, does it balance out in the end? I don't know exactly (my gut feelings says the advantage is still on the low ROF weapon). Maybe server statistics could reveal this.

Quote

2. Hardpoints and tonnage.

If your running a LaserBack with 9 laser hardpoints (or something like that) then yeah you want smalls or mediums. But what if you only have 4 ? Well if your an atlas with extra tonnage and crit space then you probably want larges.

That is true - but if the weapons are still only half as efficient, you have to ask yourself - why even bother with the larger mechs if I don't get more damage out of it?

#51 RragnarR40k

    Member

  • PipPipPip
  • 71 posts

Posted 27 November 2012 - 06:42 AM

Nice graphs,

Except you forget a vital factor to being able to calculate actual efficiency of a weapon.

The aiming factor and how easy/difficult the aiming is on various weapons.

Some hit almost automatically, some needs to stay on target for a while to do full damage and some are slow moving projectiles where you actually have to guess or predict where the shot will hit.

On top of that there is the fact that a fast firing weapon like an AC2 needs to actually hit with every shot to be up there. While slower firerate weapons can be easier to hit with since you can wait with pulling the tricker till your almost certain it will hit. Thus the hit propability of those weapons is higher than the AC2 f.i.

All of this taken into consideration your charts are kinda worthless for showing the actual balances of weapons...

The much more interesting data, would be sheets of various mechs with different weapons performances overall in a large number of battles and then analyzed. Or even broken down into say, how many shots that the average gamer actually hits with on each weapon broken down into numbers and statistics.

So please stop trying to prove stuff with faulty facts and datasheets.

Now balance would be nice and easy in comparison if the devs just made all weapons auto-hit. Wouldn't that be a fun game????

I think not...

#52 MustrumRidcully

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 10,644 posts

Posted 27 November 2012 - 06:48 AM

I encourage you to work with me and find out reasonable values for these factors. I am well aware of them.

But: THe hit difficulty is not different between all weapons. Hitting with a ML is as difficult as hitting with a LL. PPC and Gauss Rifle have the same projectile characteristics.

So I maintain that the values can be quite comparable.

And do you see any sign at all in these numbers that the devs are compensating for hit difficulty anywhere? Especially if you consider how these values evolved over time - they started with straight table top values, with some variant fire rates. They didn't assign beam durations to make lasers not that easy to hit with, but they gave LLs a longer duration, making it more, not less difficult, to hit with them, despite the origin system they got the values from in the first place did balance the weapons around the exact opposite concept.

So say what you will, even ignoring the difficulty of aiming or ignoring the finer points of rate of fire, my tables already provide considerably amount of information that can be used to improve the game's balancing.

#53 Antarius

    Member

  • PipPipPip
  • Bridesmaid
  • Bridesmaid
  • 97 posts
  • LocationGermany

Posted 27 November 2012 - 07:00 AM

View PostEmperorMyrf, on 27 November 2012 - 06:07 AM, said:

Posted Image


sorry i dont get your statistic, could you explain it bit more, lead me through step by step?
A legend, what every term mean would be nice.
Convergence angle, between what?
minimum intersectionrange?
etc.

#54 EmperorMyrf

    Member

  • PipPipPipPipPipPipPip
  • The Howl
  • The Howl
  • 740 posts
  • LocationMinnesota, USA

Posted 27 November 2012 - 07:10 AM

View PostAntarius, on 27 November 2012 - 07:00 AM, said:


sorry i dont get your statistic, could you explain it bit more, lead me through step by step?
A legend, what every term mean would be nice.
Convergence angle, between what?
minimum intersectionrange?
etc.


Yea I tried to keep it short, so sure. Again, all math is done with the 4P Hunchback.

Convergence Angle: Angle between pointing straight forward and maximum rotation.

Minimum Intersection Range: For the 4P, the shortest distance that all 6 shoulder lasers will intersect.

Range: As you can guess, distance between target and firer.

Convergence Difference: The horizontal spread, in meters, between the two columns of lasers. Essentially how wide all of your impacts are given a set distance.

Angular Room: This data was collected assuming the target was another 4P's Center Torso. The angular room is the amount of room for error in your aiming you have, measured in degrees. Essentially a measure in how difficult it is to put all 6 lasers on target, smaller values being more difficult.

Aim Error: The distance between where you're aiming and the centroid of all 6 lasers on the target. Basically, a measure in how accurate your shot will be, lower values being more accurate.

Yell at me if I'm still not clear, I have a hard time explaining things sometimes :(

#55 Indoorsman

    Member

  • PipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 792 posts
  • LocationTexas

Posted 27 November 2012 - 08:11 AM

View PostEmperorMyrf, on 27 November 2012 - 06:07 AM, said:

Posted Image
Purple is the reticle, blue is the centroid of the laser impacts, and green is the lasers.


The only problem w/this is that people will just adjust for a change like this. Unless you are saying it would be random, in which case this change wouldn't affect all weapons equally. Lasers would be affected the least because people could see where their lasers are actually hitting and just move em over mid-beam.

#56 MustrumRidcully

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 10,644 posts

Posted 27 November 2012 - 08:21 AM

You cannot always compensate for it, indoorsman. The further away your convergence point is from the target, the wider will the individual laser beams be apart on the target mech. That eventually means that it is impossible to hit with all lasers together at the same target, since your shots are spread over a larger area than the actual section you wanted to hit. And even if they can all fit in the target section, also keeping your beams centered stable so that they do will be harder than if you only have a single point to focus.

Edited by MustrumRidcully, 27 November 2012 - 08:22 AM.


#57 Asatruer

    Member

  • PipPipPipPipPipPip
  • Knight Errant
  • 235 posts
  • LocationSeattle

Posted 27 November 2012 - 08:48 AM

View PostIndoorsman, on 27 November 2012 - 08:11 AM, said:

The only problem w/this is that people will just adjust for a change like this. Unless you are saying it would be random, in which case this change wouldn't affect all weapons equally. Lasers would be affected the least because people could see where their lasers are actually hitting and just move em over mid-beam.

I think the following is the key

View PostEmperorMyrf, on 27 November 2012 - 06:07 AM, said:

I don't think that convergence should be removed from the game, which apparently a lot of people are trying to push for. I just feel that it needs to be limited at short range so that smaller weapons can't have their naturally high efficiencies exploited.

As I understand it, the lasers would not be able to achieve full pin-point convergence past short range, what is meant by "short range" is a bit ambiguous (it could be short as in BattleTech's Short range, which would be 3hexes or 90m, or it could just be a vague short range) but my interpretation is that it at least means less than the optimal range of the weapon (270m in the case of the MLs). If we go with the possibly incorrect assumption that it is BattleTech's Short range of 90m, then the 6MLs in the RT of the -4P would only be able to get pinpoint accuracy at 90m, and the farther out from there the wider the gap between beam strike points.


edit: Ninja'ed by MustrumRidcully... how on earth did it take me 27 minutes to write the above?

Edited by Asatruer, 27 November 2012 - 08:51 AM.


#58 MustrumRidcully

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 10,644 posts

Posted 27 November 2012 - 08:53 AM

You must be getting old, Asatruer.

#59 Indoorsman

    Member

  • PipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 792 posts
  • LocationTexas

Posted 27 November 2012 - 09:55 AM

View PostMustrumRidcully, on 27 November 2012 - 08:21 AM, said:

You cannot always compensate for it, indoorsman. The further away your convergence point is from the target, the wider will the individual laser beams be apart on the target mech. That eventually means that it is impossible to hit with all lasers together at the same target, since your shots are spread over a larger area than the actual section you wanted to hit. And even if they can all fit in the target section, also keeping your beams centered stable so that they do will be harder than if you only have a single point to focus.


Well, the image depicted for the hunchback shoulder grouping is a parallelogram. If you fire them all at once then yes you will have that problem. But say you break down the top 3 and the bottom 3 into 2 triangles, you're now more accurate. But yes that would still "spread" the damage, over time not space though :-p

IF an aiming nerf HAD to happen it would have to be random not anything predictable. IMO

Edit, and once again that wouldn't have the same effect across all weapons. Lasers would fare better than the others due to beam duration for making aiming adjustments.

Edited by Indoorsman, 27 November 2012 - 09:57 AM.


#60 MustrumRidcully

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 10,644 posts

Posted 27 November 2012 - 11:36 AM

One of the problems ofthis methodology is unfortunately that it ignores that plenty of the weapon combination actually waste some of their heat dissipation. Most of the low-heat weapons basically have spare heat dissipation - just adding one ton for a medium laser could sginificantly increase their damage output at almost no additional cost. This is difficult to account for...


Oh, and Vapor Trail - if I see things correctly, if your min/max assumptions on the UAC jamming duration are correct, double shotting makes the average ROF lower than the base ROF!

Edited by MustrumRidcully, 27 November 2012 - 11:39 AM.






19 user(s) are reading this topic

0 members, 19 guests, 0 anonymous users