1. 15th of October, 2019: SMD is currently under maintenance, the site may go down while we are working on it. Apologies for the inconvenience.
    2. We've removed some functionality from SMD in preparation for a migration to new forum software.

    Balancing with Mass is fundamentally flawed!

    Discussion in 'General Discussion' started by MrGrey1, Aug 23, 2019.

    ?

    How do you Measure?

    1. I measure with Mass

      83.3%
    2. I measure with Power

      16.7%
    3. I'm now changing to Mass

      0 vote(s)
      0.0%
    4. I'm now changing to Power

      0 vote(s)
      0.0%
    1. MrGrey1

      Joined:
      Feb 10, 2017
      Messages:
      301
      Mass = Thrust + Weapons + Shields + Armour + Power
      OR
      Mass - Power = Thrust + Weapons + Shields + Armour

      Mass Formula (MF) (Non system blocks & decorative blocks are classified as Armour. Power includes stabilizers and chambers.)

      Power = Thrust + Weapons + Shields

      Power Formula (PF) (e/sec) (PF doesn't have to be 100% balanced. For the ease of this discussion I am assuming that the formula is set and that total power usage = total power output. Ie no redlining / overdrawing the reactor.)

      From the PF it can be shown that if you want to increase power then you also have to increase one/some of Thrust, Weapons or Shields. You're wasting power otherwise. It follows that if you want to increase Thrust, Weapons or Shields you have to increase Power or decrease one of the other two factors. This is a balanced constant equation. What effects one variable effects all other variables equally. One goes up the others go down, exactly like the equation for the Pressure/Volume Law.

      From the MF it can be shown that if you want to reduce your Mass the obvious thing to do is to increase your Power. However from the PF we already know that if you increase Power you will also want/have to increase one of either Thrust, Weapons or Shields.
      The ONLY thing that you can do to balance the MF equation after increasing Power(and one of the other variables) is to reduce Armour... The extra Power gained (not mass) has to be put into one of the other variables on the other side of the equation, ie Thrust, Weapons and Shields subsequently also increasing their mass, further requiring that Armour is reduced... If we follow this loop to it's natural conclusion then the Mass calculation formula ends up looking like this.

      Mass - max_Power = Thrust + Weapons +Shields + 0

      Look familiar? Anybody spot a problem?
      The loop quite literally continues until there is no Armour left and we have a maximum powered ship with zero armour, a very powerful, very unbalanced glass cannon... this is why people who base their builds / competition on Mass ALWAYS end up with glass cannons complaining that Armour sucks. The reason armour sucks is because the way you are measuring your ships performance is fundamentally flawed. It grossly distorts the weighting of Armour in the equation that is used to measure performance. To add to that, the reason the balancing keeps failing is because no amount of balancing can ever fix this... re-balancing the same equation is never going to make it give a different answer.
      If you measure with Mass then it is a FACT that if you want to decrease Mass and increase Power you HAVE TO REMOVE ARMOUR. There is no other option!!!

      I've had the feeling something was up with measuring based on mass since I first heard that's how it was done. I'd always based my ship comparisons off Power and thought doing it from Mass was a little odd but never bothered to look at the issue in detail. After sitting down and working it out I am now of the firm belief that a large amount of the grief on these forums has been caused by this base error in measurement. Half the forums basing their performance on Mass (pvp), half basing it on Power (pve), both sides coming up with vastly different results, shit fight ensures because nobody bothers to clarify their starting assumptions and assumes that all their base assumptions are the same as everybody else.

      Like I have stated previously.

      In order to determine balance, you MUST have a way to measure.
      In order to measure you have to have a solid unit of measurement.
      For the above reasons. Mass is a flawed measure. This is mathematical proof that it unfairly weights Armour against ALL other systems.
      The only option for ship comparison must be Power as it is the only constant equation we have.

      If there is an error here I would very much like to know about it so I can correct it.
      Until such time as the flaw in my reasoning is corrected I personally will not accept any judgement of performance based on Mass as it's provably a flawed measurement.
      If the devs continue to 'try' and balance a game based on a provably flawed method then I'm going to be needing a paddle, anybody got a spare?

      For anybody out there who genuinely wants to see this game succeed and who has previously built their ships according to Mass. I beg you, seriously, I am begging, go back and review your ships and assumptions based on a measurement of Power and see if some of the issues with armour and balancing don't evaporate immediately? Run some tests based on Power! See what happens? Pretty please?

      Cheers,
      MrGrey1

      Edited speeling n stuff., ;p
       
      #1 MrGrey1, Aug 23, 2019
      Last edited: Aug 23, 2019
    2. M.Waites

      Joined:
      Apr 21, 2018
      Messages:
      7
      Hm, isn't the only thing affected by mass basically just the turn rate of the ship (plus the dimensions?)?

      I mean speed, shields, weapons etc mostly depend on the power avaiable to...well power them :D

      That said I wish there'd be basically massless "deco armor" so interior wouldn't add much to the mass of the ship.

      And the benefit of bigger ships would be to have the potential bigger energy to play with to power more shields, bigger weapons, equivalent thrust etc.

      Hm, I wonder but who would follow into the "Why have much armor if you could just have more shields?" Isn't, assuming they work as intended, it better to get zero persistent damage than just less of it? Given that shield HP does regenerate... unless there'd be like a Borg chamber effect or so which would regenerate armor blocks lost to damage, during combat too. (so not the current support stuff)

      So I wonder if the term glass cannon would qualify too. Wasn't the deal with the new armor that you basically have layers now and the attacker has to peel it off depending on the damage/penetration values? So a lot of smaller hits which are just enough to peel of one layer could quickly nullify the armor benefit too, right? While in the shield and no/less armor example it would be ineffective until it breaches the shields.

      While big hits would penetrate deeper into blocks. If the average pvp weapon would penetrate the armor blocks of common armor focused builds regardless then there'd be no benefit to using such a focus I guess.

      Question should also be, what's the supposed balance?

      Should the goal be to make the choice of shields vs armor be a solid one, so they're having similar benefits? If that's the case then one could possibly have other effects in place, like a "nebula" or sectors/zones where shields are entirely ineffective (Like sometimes in Star Trek etc). So for a universally usable ship it would not only rely on their shields too.

      That way at least for regional battles I doubt everyone would bring armorless ships into battles taking place in a system where shields don't work. Or the armor heavy one might have the upper hand there.

      But yeah basically Mass would basically only have an impact on the turn rates of the ship (and potential stuff crammed into it, but that could be anything, hence it's rather unspecific)

      While potential Power of the ship would also be rather unspecific as it could go into different areas too, from thrust to shields to weapons etc.
      So a whole 1 value to determine how "good" (quite subjective) a ship/station is is unreliable.

      And when we check the offensive capabilities, from which angle do the weapons have a firing arc, and how much lesser is it from a different angle? For builds with armored sections, how deep does it go from which angle too, what about other areas etc.

      I feels it's overall quite complex to narrow it doen to a single value. Both are just potential and it's uncleas how much of it is fullfilled or rather how heavy does it lean to one or the other offensive to defensive to maneuverability etc.
       
    3. Energywelder

      Energywelder Director-General, EDY

      Joined:
      Oct 5, 2013
      Messages:
      81
      What about secret option #c, "I measure in blocks and then assume the worst"?
       
      • Like Like x 1
    4. Az14el

      Az14el Definitely not a skywanderers dev

      Joined:
      Apr 25, 2015
      Messages:
      848
      mass i find most useful because it's the given metric on nav :u
      if Reactor size was given in the same way I'd probably find that more useful, but its not, and imo that's something you should have to scan for at the least.

      no time to try and figure out their regen or block count or whatnot when getting shot at etc for those cases, practical standard
       
      • Like Like x 2
    5. Scypio

      Joined:
      Sep 18, 2014
      Messages:
      570
      You take a given mass and it is how good you use said mass to accomplish whatever you want. It was more relevant in the previous power system where there was no reactor level and reactors shape could give more or less power depending on them.

      I'll still use mass until i have access to other's reactor level easily.
       
    6. Crashmaster

      Crashmaster I got N64 problems but a bitch ain't one

      Joined:
      Oct 18, 2013
      Messages:
      406
      What if you have two different sized reactors?
       
    7. SchnellBier

      Joined:
      Sep 10, 2014
      Messages:
      207
      I would say the the proper way to compare ships are resource cost. But that is a bit complicated. Might get more common once survival is more than just clicking asteroids.
       
      • Like Like x 1
    8. Ithirahad

      Ithirahad Arana'Aethi

      Joined:
      Nov 14, 2013
      Messages:
      4,133
      I'm not huge on the idea of balancing via mass either, tbh, but I don't really have the expertise to weigh in on that subject too much, so instead I'll just point out that armor sucking is an issue that goes beyond trying to minmax for mass.

      If you want to measure with power output, then the power it indirectly costs through thrust is ridiculous above fairly small ship sizes. As you add armor, you can easily cripple your mobility and give the enemy an advantage at the same reactor class/power output long before the accumulated armor starts actually mattering for survivability. If you want to measure with volume or dimensions the way that creative builders and less competitive players tend to, then the sheer volume of armor needed to be dedicated for meaningful protection with vanilla setting is, again, ridiculous.
       
      #8 Ithirahad, Aug 24, 2019
      Last edited: Aug 24, 2019
    9. MrGrey1

      Joined:
      Feb 10, 2017
      Messages:
      301
      So I've been doing some testing in regards to the meta pvp ship here and my above Mass and Power formulas.

      I only had one vertical ship at a <20k reactor, (the above pvp ship also having very close to <20k reactor.)
      My ship was covered in crystal and multiple layers of armour and significantly heavier. So I stripped it to the systems as shown.
      starmade-screenshot-0638.png
      Now this is my default starting ratio for a new ship, I do X# reactor blocks and then do 2.5X# of caps and chargers.
      The mass of this ship in it's stripped down form with thrust at 2 is about 32k, 8k less then the pvp example above. The pvp ship can not get through this ships shields at all. So I've now got about 8k mass and 100k e/sec of weapons to put on it to be able to beat the pvp example... the pvp exampls has <5900 mass of weapons total for comparison.

      So my base pve ratio reactor/shields/thrust ship, stripped of all its non essential blocks is unbeatable by that pvp ship and would easily have enough weapon power and mass available to destroy it in return.... without using any exploits. Just a big stack of systems.

      If you are comparing ships by mass then just don't bother with armour at all. Your ship will be better then anything that uses any sort of armour what so ever if you build it correctly, ie put enough weapons on it to just overcome it's own shields... that is literally the way the mass formula works and that includes doing all the fancy entity stacking you like. Judging the way everyone's geared toward beams I'd say dropping the shield caps by 50% would make it even lighter and more powerful...

      Measuring by mass is broken as amour is weighted against all other systems.
      Measuring by power puts armour into the thrust mechanic where it should be so that armour is not weighted against powered systems, armoured ships are heavier and slower, non armoured ships are lighter and faster. A balanced ship is somewhere in between, whatever that mass may be. The correct mass for any given reactor size will be determined organically on the battlefield.
       
    10. Energywelder

      Energywelder Director-General, EDY

      Joined:
      Oct 5, 2013
      Messages:
      81
      Absolutely MATHEMATICAL but then why not just compare mass vs block count? Then you get an idea of ship capabilities without making armor cubes seem more scary than they are?
       
    11. Nauvran

      Nauvran Cake Build Server Official Button Presser

      Joined:
      Jun 30, 2013
      Messages:
      2,192
      main reason why using mass is bullshit is because of how unbalanced everything is.
      Always said it's better to use block count but nooooo, no one ever listened and now we got this stupid thread
       
    12. Az14el

      Az14el Definitely not a skywanderers dev

      Joined:
      Apr 25, 2015
      Messages:
      848
      You don't get these values unless from estimation or word of mouth, except in the case of a shared/stolen design.
      There's other metrics that could be used but there simply not useful if there completely obscured & up to guess work, compared to the easy guesstimate from nav. Armor is the most difficult thing to approximate as a weight proportion since it's used very differently, which is why i always assume the most effective armoring tactic i know of, whatever that happens to be at the time.

      Ultimately it still comes down to the actual accessibility of the metric, the most accessible tends to be the most used by simple psychology. More important question seems to be, should nav/other on screen data give more data on targetted/in range entities, like reactor size, armor weight, etc. Probably good ones to brainstorm considering the coming scanning system rework.
       
      • Like Like x 1
    13. Energywelder

      Energywelder Director-General, EDY

      Joined:
      Oct 5, 2013
      Messages:
      81
      I vote lvl 1 recon chamber gives block-count, who's with me?
       
    Loading...