Jump to content

How Would You Do Balance?

Balance Gameplay

37 replies to this topic

#1 Xorkrath

    Member

  • PipPip
  • Bridesmaid
  • Bridesmaid
  • 36 posts
  • LocationPennsylvania, USA

Posted 07 November 2017 - 12:51 PM

Disclaimer: I played extensively from December of 2016 until around July of 2017, and have only played around two matches since then. But I've been keeping up on the forums and this is something I've thought a lot about. If I'm mistaken or speaking upon inaccurate or outdated information, you have my apologies in advance.

-
Many people talk about how they'd balance a game. Any given point may or may not have merit, but I always find it fascinating to get a glimpse of other people's perspectives and experiences in a game. I admit my favorite part of this game, in particular, has been reading the longer threads where hundreds of people contribute, argue, and debate different points around a central theme. I love the discussion, I love seeing other people's perspectives.

I want to start a topic about balance. What your balancing philosophy is, where you think specific problems lie, what you would do to try to limit or eliminate those problems. I'll open the topic with some bits of my own thoughts, which I'm sure will get me flamed to all heck and back. But I'm really more interested to hear what other people would do with this game.

-
On asymmetrical balance - it's a good idea but needs work. Too much asymmetry, and you shoehorn a faction into a specific role where the other can't compete, which in turn stagnates the meta for both parties. I think that's precisely where we are right now. Therefore, I'd bring a lot of stats of both sides closer together, and some I'd push further apart, details forthcoming in appropriate sections.

Battlemech vs. Omnimech - I find that there needs to be sufficient reason to have a desire to use either type of mech. Currently there isn't a lot of incentive to use battlemechs if you have omni as an option (I don't have to go into detail on this reasoning, do I? I think most people on the forums know the drill here). Therefore I would differentiate these two classes of mechs by drastically expanding upon quirks. First, I'd have all battlemechs quirked with bonus XP and cbills. Second, every battlemech would get additional quirks to be determined by the individual chasis. I would do away with quirks that were too specific and replace with more generic ones, i.e. ER-MED duration would be changed to all ER duration. Then we have a dynamic of battlemechs being better specialized towards certain weapons or groups and being more rewarding to play, while omnis are adaptable and easier to boat with.

Lights - a subject frequently brought up on the forums is light mechs and their role in this game. It's a hard thing to approach because lights, especially if they wolfpack and are piloted by experienced players, are incredibly deadly. But they are much harder to perform with when playing solo, or for a newer player who hasn't yet mastered exactly how to play a fast/agile class in an otherwise slow and lumbering game. I'd start by giving the class an XP and cbill bonus at the end of round, doubly-so for any tasks they perform that would be thought of as a "light" role. Such things would include but would not be limited to scouting, target spotting for missiles, and tag/narc assists. Other than that I don't know what to do for the class and I'm certainly interested in hearing other people's ideas on that.

Ghost heat - I think it's necessary for this game to disincentivize certain high-power low-risk builds. It is an imperfect and unpopular solution, but it does much more good than harm. However there are certain circumstances where it isn't enough, some circumstances where it is too much, and a UI problem that only notifies you the problem will come up - nothing about how severe the penalty is, which is a penalty that changes from group to group. I'd update the UI with specifics on how much heat is going to trigger within the existing warning box, plus the same info within the weapon info card.
----But I also said there are circumstances where it isn't enough. That's because a heat spike is only a problem if you can't handle the additional alpha heat, or if you're in a situation where you need to keep shooting rapidly. If you're poking, the main meta of this game, you can just sit behind cover and cool off longer. And if you're in a desperate situation, there is really no reason to worry about them at all. Furthermore, certain ghost heat penalties are so low I haven't even noticed them on builds where I have intentionally triggered them. If we are going to say boating certain weapons is too powerful, then we need to back that up with a stronger stance against them. I'd introduce "misfire", where there's a RNG 20% chance of any given weapon in a triggered ghost heat instance to not fire at all. Yes, you still incur the heat penalty and necessary weapon cooldown, but now you face an actual risk of not even firing your group as effectively, making it an actual risk. (and I'm sure that idea will make me just so very popular around here)
----On that note, a certain weapon group that does not seem to warrant ghost heating is LRMs. I'd experiment with removing LRMs, and only LRMs (both sides) from the ghost heat tables and seeing how that worked out.

Weapons - I'll get into more specific points in the appropriate sections, but in general I think we need to homogenize the crit space, weight, and damage of the same weapons across both factions (using Clan as the standard). I'd give IS weapons a buff to optimal to get somewhat closer to, but never match, current Clan optimals. Having weapon stats too far apart gives too much advantage to one side over the other depending on the map, and I don't like the idea of the map having and outsized influence over who wins or fails.

Ballistics - I think there's a great dynamic at play with the asymmetrical weapons here, where IS weapons have shorter range but fire a single slug. I think this is an instance of asymmetry working well. What I would do is give IS a shorter optimal than clan, but a much longer falloff and higher velocity. The clan would have longer optimal, but multiple slugs to hit to target and a stunted falloff. That means in a vacuum, the IS pilot could judge for himself to waste ammo/heat hitting an approaching clan mech before the clan mech can fire back...but the clan mech would rapidly get an edge in damage once he got within his optimal range (but still outside IS optimal), until the range closed within IS optimal so everything was equal at that point.
----AC/UAC balance - We ought to treat these two as the ballistic equivalent of ER and pulse lasers. To keep it simple, I'd give AC's better range and velocity (ballpark: 20%) over their UAC cousins, giving each a fighting chance of having a use-case. If need be, I'd look into the idea of reducing one crit space for some of the AC group.
----LBX's - because of the nature of this game's damage mechanics, these weapons just under-perform compared to their slug brethren. I'd give all LBX weapons an ammo buff, probably in the ballpark of 20-40% per ton. Also since I stated earlier that I'd homogenize weight and size stats, that would mean the IS LBX20 would be much more usable.
----Gauss - I'd eliminate charge-up from the light gauss only. Other than that...a high-powered no-heat weapon can keep the charge-up mechanic. Maybe introduce some recoil for the standard gauss, and heavy recoil for the heavy gauss.

Lasers - same basic premise here, I'd give clans better optimal, with IS having a lot more falloff, and retain the current status quo in regards to burn duration between the two factions. With the gutting of Clan falloff, that would mean an overall nerf for the C-ER laser line, but I think that's fully justified to keep the theme and balance of these changes.
----Standard/ER - I'd restructure the burn and cooldown times within each faction to do away with nonsense like ER-LAR firing faster than ER-MED. Stuff like that is just unacceptable and should never have made it to the PTS, nonetheless the live server. I might also go for a small blanket nerf to all regular (non-pulse) laser cooldown, to further differentiate them from pulses because I do feel that it is very important to keep a stark valley between the respective hills of alpha and DPS. As as specific example, I'd drop the ER-MED to 6.5 damage and reduce it's heat to around 5.5.
----Pulse - Again, I'd restructure to keep a logical and consistent curve from small to large, respective of the damage, duration, and heat values to their ER cousins from both factions.

Missiles - I don't see the need for much to change here, other than the already-stated weight/crit homogenization and removal of ghost heat from LRMs. It's a personal preference, but I might give the NARC missile a buff to hitpoints and speed to make it more usable in this AMS-saturated environment.

Engines - I'd go with a 20/40/60 system for the engines. When using a non-standard engine and your ST gets blown out, it's a 20% penalty (speed and heat) for LTE, 40% for clan XL, and 60% for IS XL. IS is still getting the shorter end of the stick, but they do have more options for engine size and class plus they can survive an ST blowout with these changes.

Heat sinks - I would toy with the idea of halving their heat capacity bonus, while doubling the dissipation rate. If I were in charge, I'd at least PTS it for a few weeks. Also, homogenize the sinks using the clans as the template (meaning 2-slot DHS for IS).

Mechs - Especially for the IS side, these changes have the potential to accidentally make certain mechs vastly over-perform because their prior performance was based on equipment weight and space that is no longer the same. On a chasis-by-chasis basis, I'd evaluate the needs of the mechs to determine if they suddenly need a reduction in crit space, though a part of me doubts it'd be necessary.

-
Now, I won't claim this and this alone is perfect or takes everything into account. But this is where I'd start. This is where I'd begin, then monitor and make appropriate specific changes thereafter. Keep in mind there will never be perfect balance and certain specific things might always emerge as too powerful, but that's what continuous monitoring and balancing changes are for.

#2 FupDup

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 26,888 posts
  • LocationThe Keeper of Memes

Posted 07 November 2017 - 01:11 PM

In general (wall of text inbound!):

Try to emphasize the differences between weapon types, such as pulse vs. ER/standard lasers (make pulse into DPS guns), and also within each weapon type (e.g. all IS PPCs vs. each other, all missile salvo sizes).

Another pass looking at missile weapons to un-normalize their cooldown and spread values, especially the new missiles. This is a very unpopular idea, but I think that missiles with the best alpha strike shouldn't have the same spread and cooldown as missiles with only a tiny alpha strike. Less alpha should mean better cooldown and spread, period. It doesn't have to be a super huge difference, but there should be at least some difference. However, I will throw a bone to the larger launchers with heat, that way they're much more efficient for big mechs. Ghost Heat may also be used to help pacify fears of small numbers being used too much.

Try to create weapon "feel" difference between each faction where possible, rather than the "heavier but superior" verses "lighter but inferior" that people keep asking for (currently seen with SRMs). E.g. buff IS MG direct damage but reduce crit multiplier, so they're better against armor than Clan MGs but not as effective at destroying equipment. Clan SRMs should probably be streamfire with tighter spread and maybe longer range than IS SRMS, at the cost of being less heat efficient (see below) and having some facetime. Etc.

Reduce the heat on IS guns across the board, with the amount varying by specific gun. This helps solve the faction heatsink issue because less heat means you don't need as many heatsinks in the first place. I prefer to approach it from this angle because then it actually benefits all IS mechs like lights and mediums instead of only helping the big guys that can afford to boat a crapton of DHS.

Reduce Clan ER/Heavy laser damage across the board, exact amount varying by class (e.g. -2 for HLL, -1 for ERML/ERLL, -0.5 for ERSL). Micros excluded of course. Some compensation will be given in each case as needed (e.g. 6 damage and about 5.5 heat for CERML).

Clan Heavy Lasers need to get Ghost Heat linked more universally. The whole point of them was supposed to be helping mechs with low hardpoint counts and/or low tonnage pack a respectable payload, but their current function is to just universally buff Clan laser vomit on the strongest chassis. That's stupid.

IS PPC, ERPPC, and SNPPC ghost heat cap increased to 3.

A revisit to quirks across the board, with a new goal of trying to differentiate mech roles rather than just giving every mech an even spread of every kind of quirk. For example, maybe give the Adder pure armor quirks and little else, Cougar gets even more firepower, Kit Fox is an all-rounder with a bit of everything, etc. Bad robots in general need more consideration for quarks. Specifically I'd like to start seeing ammo quirks for mechs with super low pod space so that they aren't always shoehorned into laser boating.

Speaking of which, most ammo-based weapons should have their damage per ton of ammo increased to 200. ATMs can get 100 missiles I guess, they're weird.

Buff SHS to have 2.0 cooling in the engine but only 1.0 or so externally. Now they're actually good for something.

Reduce IS LFE penalty to 10-15% compared to the Clan 20%.

I'd ask for a STD engine buff but I'm not sure if MWO's existing coding can do it.

IS TCs should get their projectile speed and target info speed increased to be higher than Clan TCs of equal rating. Clan TCs retain laser range and critical hit advantages.


Those are the big ones for now.

Edited by FupDup, 07 November 2017 - 03:57 PM.


#3 mogs01gt

    Member

  • PipPipPipPipPipPipPipPipPip
  • Shredder
  • 4,292 posts
  • LocationOhio

Posted 07 November 2017 - 01:22 PM

I'd remove all differences from IS vs Clan. ERLL is an ERLL, done. This would show which mechs actually need quirks but at what point do you stop quirking mechs?

#4 MechaBattler

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Survivor
  • Survivor
  • 5,122 posts

Posted 07 November 2017 - 01:34 PM

I agree with bringing the weapons closer in line with each other. It means you'll have to take less drastic measures to balance between them. Also agree with reworking the heat system. Bye bye ridiculous high alphas. Yeah, I said it, I don't care.

I would redo quirks for each mech. Based on number of hard points, hard point location, number of same type hard points, hitbox sizes/shape relative to other mechs in it's weight range, engine cap and extra equipment options. That would determine the percentage sizes. Then I'd lay out the different variants. Try to create roles for each one using the allotted quirk percentages.

#5 Tarogato

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Civil Servant
  • Civil Servant
  • 6,558 posts
  • LocationUSA

Posted 07 November 2017 - 02:30 PM

View PostXorkrath, on 07 November 2017 - 12:51 PM, said:

Disclaimer: I played extensively from December of 2016 until around July of 2017, and have only played around two matches since then. But I've been keeping up on the forums and this is something I've thought a lot about. If I'm mistaken or speaking upon inaccurate or outdated information, you have my apologies in advance.

-
Many people talk about how they'd balance a game. Any given point may or may not have merit, but I always find it fascinating to get a glimpse of other people's perspectives and experiences in a game. I admit my favorite part of this game, in particular, has been reading the longer threads where hundreds of people contribute, argue, and debate different points around a central theme. I love the discussion, I love seeing other people's perspectives.

I want to start a topic about balance. What your balancing philosophy is, where you think specific problems lie, what you would do to try to limit or eliminate those problems. I'll open the topic with some bits of my own thoughts, which I'm sure will get me flamed to all heck and back. But I'm really more interested to hear what other people would do with this game.


Yeah, I can't really say that there are any other games that drum up as much private discussion as this one. There's a lot of politics in MWO...
- lore conservative vs gameplay liberal
- asymmetric balance vs symmetrical solutions
- casual players vs serious players
- Clan-indoctrinated players vs IS-indoctrinated players
- people who sympathise with assaults vs people who sympathise with lights
- people who prefer FP vs people who prefer QP
- LRMurs vs anti-LRMurs
- simulation crowd vs arcade crowd
- the never-ending topic of time-to-kill
- ghost heat proponents vs advocates of alternative (or none) solutions
- quirks vs anti-quirks
- the plethora of discussion on the topic of aesthetics (mech modeling, etc)
- PGI supporters vs PGI cynics
- DOES IT EVER END
- there's plenty more


Quote

What your balancing philosophy is, where you think specific problems lie

I don't think that MWO has specific balance problems per se. I simply think that balance has been poorly informed over the entire history of the game. Those in charge of adjusting balance have never seemed like they fully understand what is going on, it's like they're not aware what the good weapons are vs the bad weapons, they aren't aware what the good mechs are vs the bad mechs. For every good change they've made, they've also made a completely backward change. Pretty much without fail.




Quote

Battlemech vs. Omnimech - I find that there needs to be sufficient reason to have a desire to use either type of mech. Currently there isn't a lot of incentive to use battlemechs if you have omni as an option (I don't have to go into detail on this reasoning, do I? I think most people on the forums know the drill here)

Yes, I think you very much have to explain yourself here. Because you have it completely backwards. Battlemechs in MWO are vastly superior to omnimechs. Omnimechs cannot change their engines or toggle their structure/armour, plus they have locked slots that prevent them from using the hardpoints that they already have (try fitting an LB20 on a Cougar, or putting gauss into the Night Gyr's torsos). Omnimechs depend very much on the equipment they are stuck with and cannot change. Compare the Dire Wolf against the Kodiak. Compare the Huntsman against the Hunchback IIC. Compare the Marauder IIC or Madcat Mk.II against Warhawk or Executioner.

The only good thing about omnimechs is that because their engine is locked, if they have a bad sized engine you can quirk the mech out the wazoo without worrying about players putting in a better engine and getting extra quirks they don't deserve. I suppose you could mention that omnimechs have the "advantage" of choosing their hardpoints, but really they are subject to the same hardpoints they get in lore just the same as battlemechs are. This is why the Kitfox is mostly bad except for the completely made up hero omnipods. This is why the Adder needs loads of quirks and still isn't a good robot. It's why an omnimech like the Mongrel is unlikely to make it into MWO - because out of eight variants, there's only eight unique omnipods, spread across only three variants. It's why the Shadow Hawk IIC is more likely to be included than the Mongrel - because the Shadow Hawk IIC can be given hardpoint inflation (because it's battlemech, not an omnimceh) and it can choose its engine freely. Look at the Black Python - a clan 75-ton omnimech where all four configurations have pretty much identical omnipods. It's more limited in hardpoint selection than any battlemech chassis.




Quote

Lights - a subject frequently brought up on the forums is light mechs and their role in this game. It's a hard thing to approach because lights, especially if they wolfpack and are piloted by experienced players, are incredibly deadly. But they are much harder to perform with when playing solo, or for a newer player who hasn't yet mastered exactly how to play a fast/agile class in an otherwise slow and lumbering game. I'd start by giving the class an XP and cbill bonus at the end of round, doubly-so for any tasks they perform that would be thought of as a "light" role. Such things would include but would not be limited to scouting, target spotting for missiles, and tag/narc assists. Other than that I don't know what to do for the class and I'm certainly interested in hearing other people's ideas on that.

If you're going to reward lights for doing non-combat roles (such as scouting, spotting, narc/tag) then you have to make sure that those actions contribute positively to the match outcome. Ie., does it help you win?

Hint: it doesn't. Lights are better off doing zero scouting, doing zero spotting. Their job is dealing damage to the enemy, just like assault mechs. That's just the reality of the situation because that is how MWO is designed. If you aren't killing enemies with your light mech, you are dead weight the vast majority of the time. There's no way to fix this without fundamentally changing the whole of how MWO works. We would need completely different map design, completely different gamemode design, possibly a full respawns mode... you know something kind of like what Mechwarrior Living Legends has.

My personal solution for light mechs in MWO is to rescale them back to where they were before the Great Upscale of 2016. Mechs like Firestarters and Wolfhounds were scaled appropriately for gameplay... then PGI made them so much larger that they now need massive amounts of durability quirks just to be viable. The Adder is still a huge piñata. The Jenners (both of them) are now deathtraps on legs.






Quote

Weapons - I'll get into more specific points in the appropriate sections, but in general I think we need to homogenize the crit space, weight, and damage of the same weapons across both factions (using Clan as the standard). I'd give IS weapons a buff to optimal to get somewhat closer to, but never match, current Clan optimals. Having weapon stats too far apart gives too much advantage to one side over the other depending on the map, and I don't like the idea of the map having and outsized influence over who wins or fails.

No. Posted Image


One of the things that makes this game fun, imo, is that Clan and IS are completely different. You can't fit similar builds, you can't build them the same way. Right now it's to the disadvantage of the IS side because Clan can simply boat more weapons and boat more heatsinks and boat faster engines. So why aren't IS heatsinks, being larger in crit size... just better? Why aren't a lot of IS equipment, being larger in crit size and tonnage... just better?

I fall along the lines of the fairly conservative in terms of equipment. I don't think that tonnage, or crits, or damage should be different from Battletech. But there are ways to balance all of this equipment without stepping on the toes of the lore. For instance Clan ER lasers compared to IS are supposed to be smaller, lighter weight, deal more damage, and have more range. Better in every way. But nobody said anything about duration. In MWO, clan lasers are just too easy to use. Too effective. They need a longer duration. They can have better stats than IS, keeping true to the lore, but they wouldn't be more effective than IS, because Clan lasers would take so much longer, so much more face time to deal their damage. I might start by just increasing the burn durations by 10% on each of the cERML, the cHLL, and cLPL, and seeing how things settle before moving forward. There will be people who will complain that "clan lightsabres are unusable". Tough shjt, bros! Right now they're more than usable... they're the most universally effective weapons in the game! They need adjusted somehow. But don't think I'm not aware that their competitors (dakka, and missiles) are also in need of some buffs. The past couple years of MWO has been nothing but nerf nerf nerf nerf nerf nerf. We need to start balancing in the other direction too. UAC5 and UAC10 are too hot. UAC20 is too unreliable. AC20 are too slow. SRMs have too much spread. IS PPCs just largely aren't effective. AC2s don't deal enough DPS. Small lasers (almost all of them, IS and Clan) don't deal enough DPS. LB5s just have too much spread. All the LBs need better crit DPS (except for maybe the IS LB10), possibly better DPS in general. LRMs need to not be garbage at direct fire. There's a lot of small buffs that need to happen to restore balance and bring everything back to par with laservomit. Oh, and the IS laser cooldowns do need to be buffed back to where they were... they don't have damage, they don't have range, what they need is the DPS in order to compete against clan tech.

What's funny, is I don't think balance was ever better then when the Kodiak was added to the game. Sure, the Kodiak was overpowered, but if you took the Kodiak out of the game, then the game was much better balanced at that point in time. Ever since then I feel like we've been regressing more and more. And it's really depressing.

Edited by Tarogato, 07 November 2017 - 02:39 PM.


#6 Angel of Annihilation

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The Infernal
  • The Infernal
  • 8,881 posts

Posted 07 November 2017 - 03:02 PM

Well I know how not to balance and that is by tossing in tons and tons of nerfs each and every patch making each and every patch notes an exercise in frustration at just what mech or build got broken this time around.

As to actual balancing mechanic, I actually liked the Power Draw system because it forced you to group your weapons in such a way as to avoid hitting the PD maximums and then learn how to cycle your weapons though these groups to maximize DPS and effectiveness. It also would have reduced TTK due to removing most of the crazy 16 weapons at a time alphas that tend to strip a section of the targeted mech in a single volley. It also upped the skill requirement as well again because it eliminated the one click attack sequences we have now. Additionally we wouldn't have to have had all the medium lasers nerfed substantially in some absurdly lame attempt to reduce the TTK because PD would have done that in and of itself. Additionally the vast majority of mechs that took major nerfs wouldn't have had to have been neutered to balance them because their vast firepower would have had to be handled in those smaller groupings as I earlier mentioned. Finally we might have been able to get back the old school JJs that actually let you, you know...Jump...back since some of the crazy poptart builds wouldn't have been viable.

But the community killed that. They would rather complain about everything than give up there 70 point Alpha Strikes. Go Community, you rock.

#7 Tarogato

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Civil Servant
  • Civil Servant
  • 6,558 posts
  • LocationUSA

Posted 07 November 2017 - 03:23 PM

View PostViktor Drake, on 07 November 2017 - 03:02 PM, said:

I actually liked the Power Draw system because it forced you to group your weapons in such a way as to avoid hitting the PD maximums and then learn how to cycle your weapons though these groups to maximize DPS and effectiveness.
[...]
But the community killed that. They would rather complain about everything than give up there 70 point Alpha Strikes. Go Community, you rock.

Don't forget *why* the community killed it.

It was a system that encouraged boating and discouraged variety and diversification. With GhostHeat you're encouraged to give up some SRMs and spend tonnage on a ballistic instead. With PowerDraw you're punished for firing SRMs and ballistics at the same time, so the more effective tactics is to forego the heavy ballistic and just take the more efficient SRMs because you can't get around the heat scaling system anyways. Same example is true for gaussvomit. Under GhostHeat, you might be encouraged to spend some tonnage on a gauss rifle to boost your alpha. But with PowerDraw you can't fire gauss and lasers together, so they're no reason to branch out, it's better to just stick with laservomit alone, it's more efficient because you can't get around the system anyways. Even with laservomit itself... right now GhostHeat forces us to mix medium and large lasers - we have to use different weapons with slightly different ranges and different durations and different cooldowns. Those large lasers cost a lot of tonnage... you sacrifice a lot of heatsinks to boost your alpha damage. But under PowerDraw, achieving that alpha is impossible no matter what you try, so you're encouraged to give up your large lasers and just stack the more efficient mediums and boat more heatsinks. It's a lot more boring. Not to mention it favours mechs with MORE hardpoints - it introduces a new chassis imbalance that we never had before. That is why the PowerDraw system was broken, that's why people hated on it and made absolutely sure that it didn't ruin the diversity in this game.

#8 Athom83

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Death Wish
  • The Death Wish
  • 2,529 posts
  • LocationTFS Aurora, 1000km up.

Posted 07 November 2017 - 03:24 PM

BattleMech vs OmniMech; How they are doing now is mostly fine, but seriously needs some adjustments. BattleMechs should have some locked equipment (not engine) and Omnis should have a locked engine (not equipment).

GH; Similar to Energy Draw, yet modified so it isn't based on damage of the weapon. Pushing your reactor to make more energy than normal should cause it to heat up quite significantly. Also kind of makes me thing the total energy supply should be based somewhat on the engine rating, with other equipment that could boost it like power amplifiers.

Weapons; General decrease in cooldowns (slower firing). General increase of component health.

Ballistics; General max range buffs all around (optimal ranges are about perfect).
LB; Made to a flak like shell, where the shell remains a single slug until it gets to within a certain range of the target where it shotguns out from the shell.
cAC; Drop the shells per volley and time between shells within the volley down. 2 and 5 should be 1 shell, 10 should be 2, 20 should be 3 at most (possibly even 2).
RACs; General DPS increases, as well as increasing spread.
IS UAC; Reduce the time between shell in a single volley.
Gauss; Remove charge time.

Energy; Fix their values so their damage/heat more align with BT instead of having them deal more damage for less heat than they are supposed to (looking at you clan lasers).
Lasers; Decrease in duration.
ER Lasers; Further decrease in cooldown (slower firing) and have around 50% greater duration when compared to standard equivalents;
Heavy lasers; Pretty fine as is in all honesty.
Pulse lasers; Slight buff in damage or cooldown on a case by case basis.
PPCs; Remove min range, but give them reduced damage in those ranges (not applicable to ERPPCs or SnPPCs).
Flamers; Increase range and heat buildup on the target.

Missiles; Eh... mixed bag.
SRMs; Increase velocity.
LRMs; Fix "velocity" (I've mentioned this many times in other threads), speed up lock time when in LoS, slow lock time and increase spread outside of LoS, and remove the min range for IS ones (make them like cLRMs with a reduced damage).
Streaks; Extend range, fix the targeting.
ATMs; Decrease time between missiles in a volley (condense the stream). Remove min range (again, make is scale like cLRMs up to "minimum" range).
MRMs; Pretty nice. Would like it if they tracked your cursor like a TOW missile (or maybe not and leave that to MMLs or the TBolt).

Engines; Honestly fine. Maybe give IS XL some bonus engine heatsinks.

SHS; Increase heat capacity, reduce heat dissipation.
DHS; Reduce heat capacity, increase heat dissipation.

IS Fero-fiber armor; Buff to the armor amount as a percentage like current skills. (Clan version does not need it)
IS Endo-Steel; Buff internal structure HP. (Clan version does not need it)

#9 Shifty McSwift

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,889 posts

Posted 07 November 2017 - 03:55 PM

Just quickly and in general I would probably move away from asymmetrical balance and focus more on standardisation with more of a focus on the clan/is quirks in weapons and mechs to sell that sense of difference, as well as having clan/is unique weapons and mechs (IS racs and clan ATMs etc).

#10 BrunoSSace

    Member

  • PipPipPipPipPipPipPipPip
  • The Guide
  • The Guide
  • 1,032 posts
  • Twitch: Link
  • LocationAustralia

Posted 07 November 2017 - 04:36 PM

If I was going to change one thing. Id make the IS xl like the Clans. If it lost a side torso it would stay up untill it lost both torsos. Leave this change for a month and see how balanced the game was after.

#11 Sjorpha

    Member

  • PipPipPipPipPipPipPipPipPip
  • Philanthropist
  • Philanthropist
  • 4,480 posts
  • LocationSweden

Posted 07 November 2017 - 05:04 PM

View PostTarogato, on 07 November 2017 - 02:30 PM, said:

No. Posted Image


One of the things that makes this game fun, imo, is that Clan and IS are completely different. You can't fit similar builds, you can't build them the same way. Right now it's to the disadvantage of the IS side because Clan can simply boat more weapons and boat more heatsinks and boat faster engines. So why aren't IS heatsinks, being larger in crit size... just better? Why aren't a lot of IS equipment, being larger in crit size and tonnage... just better?

I fall along the lines of the fairly conservative in terms of equipment. I don't think that tonnage, or crits, or damage should be different from Battletech. But there are ways to balance all of this equipment without stepping on the toes of the lore. For instance Clan ER lasers compared to IS are supposed to be smaller, lighter weight, deal more damage, and have more range. Better in every way. But nobody said anything about duration. In MWO, clan lasers are just too easy to use. Too effective. They need a longer duration. They can have better stats than IS, keeping true to the lore, but they wouldn't be more effective than IS, because Clan lasers would take so much longer, so much more face time to deal their damage. I might start by just increasing the burn durations by 10% on each of the cERML, the cHLL, and cLPL, and seeing how things settle before moving forward. There will be people who will complain that "clan lightsabres are unusable". Tough shjt, bros! Right now they're more than usable... they're the most universally effective weapons in the game! They need adjusted somehow. But don't think I'm not aware that their competitors (dakka, and missiles) are also in need of some buffs. The past couple years of MWO has been nothing but nerf nerf nerf nerf nerf nerf. We need to start balancing in the other direction too. UAC5 and UAC10 are too hot. UAC20 is too unreliable. AC20 are too slow. SRMs have too much spread. IS PPCs just largely aren't effective. AC2s don't deal enough DPS. Small lasers (almost all of them, IS and Clan) don't deal enough DPS. LB5s just have too much spread. All the LBs need better crit DPS (except for maybe the IS LB10), possibly better DPS in general. LRMs need to not be garbage at direct fire. There's a lot of small buffs that need to happen to restore balance and bring everything back to par with laservomit. Oh, and the IS laser cooldowns do need to be buffed back to where they were... they don't have damage, they don't have range, what they need is the DPS in order to compete against clan tech.

What's funny, is I don't think balance was ever better then when the Kodiak was added to the game. Sure, the Kodiak was overpowered, but if you took the Kodiak out of the game, then the game was much better balanced at that point in time. Ever since then I feel like we've been regressing more and more. And it's really depressing.


So much this!

IMO engines are also a very important thing to balance, and in their case I think you just have to kinda give it up and allow ISXL to survive a ST loss like the clan XL with slightly lower penalty to compensate for the 2 extra crits, and no ST loss penalties for the LFE at all. Standard could give a structure bonus or something.

I'd like single heatsinks to be viable too, and the best suggestion I've heard is to treat all heatsinks in the engine as DHS. That way the single heatsink advantage of being smaller would really matter because the SHS vs DHS choice would only affect the heatsinks that are actually different in size.

#12 R Valentine

    Member

  • PipPipPipPipPipPipPipPip
  • Heavy Lifter
  • Heavy Lifter
  • 1,744 posts

Posted 07 November 2017 - 05:28 PM

1) Convergence: Torso weapons = zero convergence. Arm weapons = good convergence. Good bye laser vomit Ebon Jags and Hellbringers. They'll work good at range, but if brawlers close the distance their torso only, or torso heavy, guns will spread out considerably. Now you profit much, much less from dumping all that arm armor. You want to converge? Use those tissue paper plated limbs.

2) Terrain layout. Less hills. Less low and partial cover. More complete, high cover and flatter land. Less hill peeking and more side peeking. That'll take a large amount of mechs out of the garbage heap. You don't need high mounts, because concealing your mech behind a hill won't be the only way to peek and not fully expose. Maps with city layouts, but lacking the stupid valley that Frozen has or the unnecessary no man's land that River City has. Seriously, who builds a city without leveling the ground first? The Aztecs? Think we've progressed past that. Very few, or no structures that can conceal an entire team. In fact no structures that can conceal more than 2 mechs. Good bye Nascar. 2 mechs might be able to circle jerk around a building big enough for 1, but an entire team won't when nearly everyone will be able to be shot regardless of your position around it. More skirmishes and less death balls.

3) Poptarts. The reticle shake now remains through the entire jump cycle, including the fall. Good bye jump snipers. Now you can make jump jets good again. I don't mind mechs being able to maneuver to high ground, utilize escape routes, and jump for positioning. What I hate is them using it to take a cheap shots with little to no chance of return fire. Can you use it to dodge fire? Sure, but you don't get a dodge and shoot at the same time.

4) XL engines. IS XL, 2 side torsos to die. Clan still MUST be better? Fine. Clan XL's lose less speed and heat dissipation than IS. LFE's lose no heat dissipation and speed with 1 shoulder gone. 2 shoulders gone kills all mechs except standard engine mechs. Now all engines have a place and an advantage and clan XL is not super duper special.

5) Crit splitting. How is this balance related? See LB20-X autocannons. One can fit on your arm. One can't. Guess which faction got screwed? Crit splitting too hard? Reduce the number of criticals on any weapon too big to fit in the shoulder without standard engine. Make it fit in the shoulder without standard engine. Any weapon that can fit in the arm in TT is reduced to fit in the arm on MWO. Herpa derpa, why was that so hard? Why the burning need to stick to table top rules when you can't even code table top correctly? Less dumb decisions based on dumb reasons. 90% less stupid.

6) Lower arm and hand actuators. All mechs can remove them. All mechs can add them. Again, herpa derpa why was that so hard? Retain current restrictions on PPCs and gauss.

7) Consumables. There are none. You want a coolant pod? Fit it. It takes weight and critical space. You want artillery strikes? Fit it. It takes weight and critical space. You want air strikes? UAVs...? You get the idea. Hell, even add long tom artillery as a module. It exists in other Mechwarrior games and in TT. But it takes weight and critical space... and a lot of it.

#13 Asym

    Member

  • PipPipPipPipPipPipPipPipPip
  • Nova Captain
  • 2,186 posts

Posted 07 November 2017 - 05:38 PM

First off, can someone provide a definition of what "balance is?" Something we can all agree on....???

Because, everytime anyone who isn't a founder says anything, we're castignated and maligned....

What is "balance?"

#14 R Valentine

    Member

  • PipPipPipPipPipPipPipPip
  • Heavy Lifter
  • Heavy Lifter
  • 1,744 posts

Posted 07 November 2017 - 05:44 PM

View PostAsym, on 07 November 2017 - 05:38 PM, said:

First off, can someone provide a definition of what "balance is?" Something we can all agree on....???

Because, everytime anyone who isn't a founder says anything, we're castignated and maligned....

What is "balance?"


So?

View PostAsym, on 07 November 2017 - 05:38 PM, said:

we're castignated and maligned....


So? I don't care what they say. They're giving an opinion on an opinion, so why that should carry any weight is just beyond me.

#15 sycocys

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Moderate Giver
  • Moderate Giver
  • 7,700 posts

Posted 07 November 2017 - 06:37 PM

Mechs -
Sized hardpoints, pre-built pods for omnimechs.
Heat scale/energy scale more similar to TT.

Weapons - manufacturer buffs/debuffs (zero sum, basically take the few options and make them a dozen each) that also extend to your faction/mech compatibility
Hard limit on alpha shots, possibly having them also cause damage/vastly increase cool down/decrease heat threshold

Tech - ECM would be ECM, information/target sharing would require tiered levels of/possibly multiple equipment. Targeting computers required for X-full damage ratings.
Also much more options here, even if you have to create something new that stays in the MW theme.

--
"Skill" tree - 100% zero sum. You buff something you lose something equivalent.
Faster CD = More heat generation, Longer range = lower damage, More structure = less launch speed/crit slots, More radar range = slower lock break

And so on, point is you should be able to specialize without gaining any net advantage over any other player regardless if they are new or vet. Zero sum specialization.

--
QP Matches - First of all they need to not be static drop points and mission locations, that would help an insane amount.

After that the MM needs to be dynamic and make match sizes based on the population in the queue (4v4,8v8,12v12) rather than opening the bucket up as a first measure.

Additionally there needs to be a lot more, and a lot more wide open large maps rather than the hyper detailed maps that are over-designed to give a particular type of play priority. QP should be the primary testing ground to add a lot of maps to and filter the best ones into more custom modes.

FP/CW - Multi-match campaigns. Solo drops get short campaigns, small groups get a slightly longer, medium groups a little longer, and full groups get the longest - as well as a choice for some sort of ranked contests with additional rewards.

Add Solaris. Kind of a no-brainer, should be essentially the QP version of tournament/comp mode. Both could really fall under the Solaris hat.

--
Maps
First off they all need to be much larger, even if that means taking down the detail level of them. The current map sizes other than Polaris are barely big enough for quality 8v8 play. A couple others come close but are hindered greatly by the next point.

As I stated before drop points and mission points need to move around these being static + the pointed map design that is designed to filter matches to particular points makes every single match on a map nearly the same which destroys replayability and on some maps gives one team a significant advantage.

Similar to the larger map point - maps also need to be more open an organic. The closest things to this are a couple of the attacker sides of a few of the FP maps (the defender sides are really not a good design on any of them in my opinion, just roads to a target on them all), and a couple of the QP maps have decent designs, but only in the areas the map is functionally designed to not ever use.

Modes - for really all but Solaris if that ever happens.

Objectives need to take precedence over eliminating the other team. Less so in QP, but they should still be the primary goal of the match and rewarded appropriately and/or cause a loss overall if not tended to in lieu of simply killing the other team.

There should also either be secondary/teritary objectives or the primary split into lances then to an overall as the match progresses.

In FP/CW objectives should be by far the primary focus - causing buffs/debuffs for current and future matches as well as influencing what the next match setup in the campaign will be. Everything tied ONLY to the two teams competing with each other.
^This is the only way you are going to create a compelling mode to fill out the FP mode that I can see, and having 3-4 tiers of play sorted by group size will naturally buffer the difficulty curve of unit play.

FP specific - I would also personally eliminate tech restrictions and instead go with faction reward bonuses for abiding by your houses preferred tech + meeting performance goals (rather than just accumulating x amount of LP). Could even go as far as losing reputation with your chosen house for not adhering to their code of conduct closely enough (which would be a good way to add some layers of balance on the back end of the mode).

This greatly reduces the number of buckets to put players in, allows mixed faction groups so whoever is online can play with whoever else is online, and would still be able to retain the silly meta map concept by tallying faction w/l by individual player instead of team.

It also should, if faction bonuses are done well enough, add a lot of variety and flavor to what people are putting on the battlefield depending on what house they choose to align with. For lone wolf/pure merc, I would suggest instead of mech limits they have more restriction based on cost of mech and armament.

Edited by sycocys, 07 November 2017 - 06:40 PM.


#16 Khobai

    Member

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

Posted 07 November 2017 - 06:44 PM

Quote

Sized hardpoints, pre-built pods for omnimechs.


do you even know what an omnimech is?

the whole point of omnimechs is that they can swap out weapons freely and dont have sized hardpoints or pre-built pods.

#17 sycocys

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Moderate Giver
  • Moderate Giver
  • 7,700 posts

Posted 07 November 2017 - 06:48 PM

View PostKiran Yagami, on 07 November 2017 - 05:28 PM, said:

1) Convergence: Torso weapons = zero convergence. Arm weapons = good convergence.

2) Terrain layout. Less hills. Less low and partial cover. More complete, high cover and flatter land.

3) Poptarts. The reticle shake now remains through the entire jump cycle, including the fall.

4) XL engines. IS XL, 2 side torsos to die. Clan still MUST be better? Fine. Clan XL's lose less speed and heat dissipation than IS. LFE's lose no heat dissipation and speed with 1 shoulder gone. 2 shoulders gone kills all mechs except standard engine mechs. Now all engines have a place and an advantage and clan XL is not super duper special.


7) Consumables. There are none. You want a coolant pod? Fit it. It takes weight and critical space. You want artillery strikes? Fit it. It takes weight and critical space. You want air strikes? UAVs...? You get the idea. Hell, even add long tom artillery as a module. It exists in other Mechwarrior games and in TT. But it takes weight and critical space... and a lot of it.


Can totally agree on these, although I think rolling hills aren't an issue - its the random mountain sized nonsense that seems to be in every map that's the issue. #1 and #7 especially though, and with convergence its completely weird that they don't have it since there are 2 aiming points for arms and torso anyhow, just need to lock the hidden reticule and open up its size so each weapon would essentially fire to the far edge of the map instead of going to the center point.

#18 sycocys

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Moderate Giver
  • Moderate Giver
  • 7,700 posts

Posted 07 November 2017 - 06:53 PM

View PostKhobai, on 07 November 2017 - 06:44 PM, said:


do you even know what an omnimech is?

the whole point of omnimechs is that they can swap out weapons freely and dont have sized hardpoints or pre-built pods.

The point of omnimechs is that they can swap out pods efficiently across their battalion, not so each mech has completely customized pods for every conceivable situation they run into.

So in the scope of balance locking the gearing of the pods on omnimechs would be one of the most viable solutions to balancing IS and Clan out. Sorry that it would make your omnimech an omnimech and not a battlemech with swappable hardpoints, but it would be actually one of the best tech balance methods they could do.

Edited by sycocys, 07 November 2017 - 06:53 PM.


#19 Deathlike

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • Littlest Helper
  • Littlest Helper
  • 29,240 posts
  • Location#NOToTaterBalance #BadBalanceOverlordIsBad

Posted 07 November 2017 - 06:54 PM

I would not use a dartboard to balance the game. Full stop.

I would consider comp play first and foremost to see where the overperforming/underperforming/OK(MLs before the nerf) weapons are, including mechs.

Feedback will be taken, but I will ignore all poor arguments (like Small Lasers being OP) outright.

Still, I would just ignore NGNG and teh dartboard with regards to any/all of this.

#20 Mcgral18

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • CS 2019 Top 8 Qualifier
  • CS 2019 Top 8 Qualifier
  • 17,987 posts
  • LocationSnow

Posted 07 November 2017 - 07:37 PM

View PostAsym, on 07 November 2017 - 05:38 PM, said:

First off, can someone provide a definition of what "balance is?" Something we can all agree on....??? Because, everytime anyone who isn't a founder says anything, we're castignated and maligned.... What is "balance?"


Where Spheroid mechs don't have a sub 15% pick rate at the "highest skill tournament in Esports" level of competition?
I'd say that's a decent place to call balance, where Sphere mechs might actually be chosen VS a Clam mech.


Asymmetrical balance is fine, but what we have now is BAD balance
That isn't fine.

I don't feel like making a paragraph
I've done that too many times already





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users