Jump to content

Balance Discussion - Aug 2018 - Post Podcast Feedback

Balance

605 replies to this topic

#121 SoulRcannon

    Member

  • PipPipPip
  • The Messenger
  • The Messenger
  • 82 posts

Posted 07 August 2018 - 06:25 PM

So, from what I heard of the podcast the heat scale changes of PTS weren't explicitly covered, or at least it felt like it wasn't confirmed whether or not it's still on the table. I know closely corresponding things were discussed, so I might be wrong. I felt there was an inference that the heat scale trigger changes for cLL would be OK as it would introduce new weaknesses to builds that would capitalise on those changes. Players would trade speed and heat management for more damage, and at range - but said weaknesses would be mitigated by the increased long range damage, and decent positioning to poke then cool off.

So I'll say it again; If you want more targeted changes, don't raise the heat scale trigger for cLL. I sincerely hope it's not still being considered. I've already covered that the collateral damage to lighter clan mechs would go too far. I know they wont be implemented, but the PTS numbers felt to me as though everything that wasn't a cLL of some kind was hit harder than perhaps it should because of the trigger changes. And then there was the cLPL that I'm not sure needs much changing at all? Anyway... Overall not exactly a good idea relative to the stated objectives here.

And when discussing LRMs, you've stated that you look to poorly performing outliers so that a majority of weapon systems within the catalog remain viable, and make changes accordingly, which is why LRMs got buffed. So are LPPC's, Light Gauss, SPL and cSPL on the radar for future revisions? Because if that's the goal there are some weapon systems to buff (or in some cases just tone down past nerfs) that wouldn't necessarily amount to power creep or overall reduced TTK compared to what's being more regularly run now. You've stated your plans to bring overperforming weapons systems down to a specific standard, with that goal in mind improving other underutilised weapon systems as long as it was in line with said standard surely wouldn't hurt, especially if it was in regards to the lighter weapons that more mechs can utilise.

When it comes to LRMs, surely you realise that it's the indirect fire (and the playstyle that encourages) that many object to alongside the maps lacking many of the features that provide players with more counterplay against LRMs. Would it be possible to increase spread/reduce the number of missiles that hit when firing without line of sight, or without mechanics like NARC? Or perhaps just increase spread regardless, but slightly buff the effect that Artemis has on the spread, to emphasise that it's not a waste of tonnage to those who want to fill their chassis with as many missile tubes as they can fit, and further encourage gaining LoS when using LRMs.

Finally, would definitely appreciate any improvements to mobility, particularly heavier chassis but also offenders in scaling/size like the Firestarter. I like piloting mechs in-game, not jars of molasses, and I don't want to feel like a liabilty in slower mechs or those that're bigger for their tonnage than perhaps they should be. I mean, I'm a newer player, and was shocked to learn that the mobility changes had gone in that direction and are only being addressed now. But, better late than never.

#122 Marius Evander

    Member

  • PipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 1,113 posts

Posted 07 August 2018 - 06:26 PM

View PostPaul Inouye, on 07 August 2018 - 01:51 PM, said:

The fact that you 'delete' someone using that build (even though not sustainable) is the problem. This is what we're trying to address. If you listen/read about the podcast, we are looking at ways to address this without using the large numbers seen on PTS. If you read the post I made initially here, we know we're not going to pull the massive Alphas down in any significant manner with the current implementation of duration/damage... but there's a starting point where we can do a minimal adjustment to start closing the gap and we will be looking elsewhere to find a better solution that doesn't penalize on global scale and such high deltas.
That's what we said we're doing in the podcast. Specifically Assault class agility. If there are 'Mechs that will need a bump in agility, they will be investigated and given the quirks they need.
Again, we can buff underperforming weapons as long as those buffs don't widen that gulf between Clan and IS.
As stated in the podcast.. there are more levels to balance other than just the energy issue we are currently focused on. We ARE listening to the community and this is why this is a discussion and not just a post of what's to come.


The 1st huge glaring WTF moment for me was the KDK-3 getting UAC-10's destroyed for any mech that could'nt mount 4, then the small pulse nerf killing any mech that couldnt mount more than 6. This is SEWERAGE WE HAVE BEEN SITTING IN for MORE THAN 24 MONTHS ? whats changed now, why are you suddenly going to address and "fix" this ? Outlying mechs like KDK-3 12 small pulse mechs and other boating problems should be target nerfed/neg quirked in a never ending game of wackamole, if you dont want red numbers, nerf the weapons like you did , BUT ALSO give mechs that cant boat bonus's.

View PostPaul Inouye, on 07 August 2018 - 02:01 PM, said:

I've said this above and in previous posts... our focus right now is the laser vomit issue on Clan side. The community doc points out weapons systems that are low performers and some other outliers (streaks) etc. Those numbers suggested by the community are not being ignored. Like I said before.. there are numbers in the community doc that work correctly and will be implemented in the future.

You mention bringing agility back and this is exactly what we've been saying in the podcast and the original post in this thread.

Put it to you this way.. help us close that laser vomit gap and we discuss all the other weapon systems on a case by case basis and move forward.


Doing weapons in a cycle ends in never ending rock paper scissors, you have to try to bring them all into a balance simultaneously, otherwise when you "fix " the 3rd or 4th weapon group the 1st group gets thrown out of wack.

View PostPaul Inouye, on 07 August 2018 - 02:22 PM, said:


This is something we mentioned in the podcast and original post.

If the slight nerfs hit a light/medium mech too hard, we'll requirk that mech to allow for better dps/sustainability because of it's hardpoint restrictions and/or weight restrictions etc.


Why the sudden change of view / promise of action after 18 months? since small pulse nerf killed laser lights ?

View PostPaul Inouye, on 07 August 2018 - 02:34 PM, said:


Lasers are being focused because of said gulf between Clan and IS as to what can be brought to the battlefield when boated. Yes, it's a couple of assaults on the extreme edge of things.. but even in lower tonnage classes, the discrepancies are there.

That being said, we're toning down the nerf substantially to make any sort of trickle down less punitive to the lighter weight classes.

Edit: What are your thoughts on negative quirks to bring specific outliers into line?


Hellbringer/direwolf neg laser damage quirks when mounting more than 6 lasers possible solution......

#123 XX Sulla XX

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • 3,094 posts

Posted 07 August 2018 - 06:27 PM

View PostPaul Inouye, on 07 August 2018 - 06:15 PM, said:


denAirwalkerrr asked this earlier. Which brought up my question about thoughts on negative quirks.

Negative quirks and lower base agility etc are fine on the problem mechs like the Deathstrike.

To me it looks like all but 4-6 Clan mechs are fine. So it is more logical to fix those 4-6 mechs than change the weapons and affect every other clan mech and then try and readjust them and then try and readjust IS armor.

EDIT the more I think about it maybe it is more like 3 or 4 mechs that are a problem.

Edited by XX Sulla XX, 07 August 2018 - 06:31 PM.


#124 Gierling

    Member

  • PipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 313 posts

Posted 07 August 2018 - 06:37 PM

View PostChris Lowrey, on 07 August 2018 - 06:12 PM, said:


While I won't get into the reasons as to why its taken until now to look into it, it is something that we will be looking into with the next PTS. So as Paul has said in other posts, post what you feel needs it the most and we'll look into it for testing.


Please keep in mind that it doesn't have to be an across the board thing, it should be more role based then anything else.

Brawlers like the Atlas need torso speed to spread damage, Hill peekers need accell/decel, side peekers need Yaw etc.

There is a lot of options.

View PostChris Lowrey, on 07 August 2018 - 06:23 PM, said:


Same logic for LRM buffs applies to NARC as well. Intent is to reduce the amount of dead weight in the 'Mechlab, but the recent buffs to baseline LRM's have defiantly put their current capabilities under the microscope. Especially their ability to NARC a full team in an incredibly short amount of time without their ability to respond.

That's going to be it for me for the day as well, I'll pick this up tomorrow.


Would you consider giving NARC some benefit beyond just to LRM's, that way you could nerf the LRM Functionality a little and not have everyone stop using it (thus providing a double whammy to LRM's)

#125 Tarl Cabot

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Tai-sho
  • Tai-sho
  • 7,775 posts
  • LocationImperial City, Luthien - Draconis Combine

Posted 07 August 2018 - 06:46 PM

What I do not see is any discussion about engines in regard of bringing the XL/cXL into parity, where the isXL receives the same type of negative, non-lethal benefits the cXL/LFE currently have in a system that is only using a partial of the TT engine crit system. And death to one ST vs death to loss of TWO ST is not even close.

View PostCadoazreal, on 07 August 2018 - 06:26 PM, said:

The 1st huge glaring WTF moment for me was the KDK-3 getting UAC-10's destroyed for any mech that could'nt mount 4, then the small pulse nerf killing any mech that couldnt mount more than 6.


The UAC10 had been a long standing bug but the Direwolf could not effectively showcase the issue. On the other hand the route PGI went with it is something else. As for the cSPL, PGI had furbar it almost from the beginning. It was the only Small energy weapon to see a massive increase in damage. It needed to be rolled back after almost 3 years. Same reason PGI is not looking at increasing Innersphere energy weapon damage nor touching the NO damage zone for weapons for most of the PPC and the isLRMS.

Just for reference.



Quote

cSPL dmg/heat

Lore/TT: 3 dmg / 2 heat

MWO June 2014: initially 3.4 dmg / 3.4 heat
MWO July 2014: increased to 4.4 dmg / 3.4 heat
MWO Nov 2014: Increased to 6.0 dmg / 3.0 heat
MWO June 2017: Decrease to 4.0 dmg / 2.7 heat
Current stats: 4.0 dmg/2.07 heat

MWO Dec 2014: Release of Community Warfare aka Faction Play

Pulse Lasers had a -2 gunnery hit modifier. Heavy Lasers had a +1 gunnery hit modifier.

Edited by Tarl Cabot, 07 August 2018 - 06:55 PM.


#126 Bud Crue

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Rage
  • Rage
  • 9,943 posts
  • LocationOn the farm in central Minnesota

Posted 07 August 2018 - 06:51 PM

I'm just going to put this out there. My take on your balance efforts of late:

- Those efforts are bull**** or based on delusion of how the game is actually played.
- They (as well as phase 3 of CW) are the reason I no longer sponsor contests, nor throw copious (any) amount of cash at you.
- I'm sick of you breaking EVERYTHING to address a few outliers, particularly when those outliers are not much of a problem to begin with.
- Quirks are not the cause of power creep, and nerfing the defensive ones by a third because of your obsession regarding the need to nerf clan energy is a recipe for even greater player disillusionment and the fleeing of players from this game. Quirks are indeed supposed to be about flavor and not about mitigating your nerfs. You have hundreds of variants, trying to make them all the same is a mistake. Give out more "flavor" quirks, not less, so we have a reason to bother with the dross.
- The pros and cons of clan high alpha and IS toughness are a large portion of the fun of this game and you are taking that away. Stop it.

I play this game nearly every night in GQ during NA prime time, as one of your coveted "beer league" level players. I don't pretend to understand much, or that I am remotely good at this game but I do know this: Your 94 point alpha, super scary, it can "delete" an enemy with 1 or two shots is a myth in the game I play. If I've encountered it, I never noticed it. Clan laser vomit and even gauss vomit is fine. Leave it alone. Just leave it alone. I, as a scrub, am no more worried about a Hellbringer than I am a Grasshopper; I worry WAY more about Annihilators, or hell even Piranhas, than I ever have the Deathstrike.

I don't understand your fixation or belief that clan energy mechs are a clear superior choice at all levels of play over those of the IS in the game I play. If I get a hot map I am just as happy to have dropped in my 8MPL Top Dog than I am in my similarly built Ebon. Given what you have done to LRMs I am just as happy in a Jester with its AMS as I am a Hellbringer with its ECM.

In the end, I just want you to leave my stuff alone; to stop breaking many things in the name of the few. And if you must do something, buff the under performers and leave that which is decent or even merely adequate alone. If there is a true OP outlier that players of all levels of skill are complaining about, then perhaps consider nerfs -TO THAT VARIANT/BUILD ALONE- but this nonsense of nerfing whole classes of weapons to address an over performance issue and then considering changes to many mechs quirks to allow the smaller mechs to compensate for those nerfs (as described in the podcast) is just more work for you, less fun for us, and gets nowhere in the end; so don't so it. Please?

#127 IronEleven

    Member

  • PipPipPip
  • Mercenary Rank 3
  • Mercenary Rank 3
  • 84 posts

Posted 07 August 2018 - 06:52 PM

View PostPaul Inouye, on 07 August 2018 - 10:18 AM, said:

reduce this 94 alpha to something like 90 (EXAMPLE NUMBER ONLY) instead of trying to get it to 80 or some other drastic number

You mean the value that the version of that build that doesn't cook you alive is already at?

Edited by IronEleven, 09 August 2018 - 02:36 AM.


#128 SlightlyMobileTurret

    Member

  • PipPipPipPipPipPipPip
  • Lance Corporal
  • 718 posts

Posted 07 August 2018 - 06:54 PM

Please look into buffing mobility of Firestarter, Mauler etc which for some reason have been left with less mobility than their equal tonnage counterparts.

The problem is compounded by WLF getting armour quirks while FS9 gets structure, and Mauler has worse hitboxes and hardpoint locations than the Sleipnir etc, but I think fixing mobility is the first step and might make these mechs being chosen more often by people.

I'm sure that there are more inconsistent base stats (I don't mean low mobility on purpose like MADIIC or NTG, but low stats for even underperforming or outright bad mechs) like this, but these are the ones I remember off the top of my head.

Images for comparison
Posted Image

Posted Image

#129 Bosseen

    Rookie

  • Philanthropist
  • 2 posts

Posted 07 August 2018 - 07:00 PM

Has there been any discussion about how the actual maps are designed and could be influencing different weapon balancing issues?

#130 MisterSomaru

    Member

  • PipPipPipPipPipPip
  • The Death Wish
  • The Death Wish
  • 255 posts

Posted 07 August 2018 - 07:05 PM

View PostBosseen, on 07 August 2018 - 07:00 PM, said:

Has there been any discussion about how the actual maps are designed and could be influencing different weapon balancing issues?

Nope, but we know the answer: Take Polar Highlands to maximize cold lrming.

#131 CommEE

    Member

  • PipPip
  • The Sickle
  • The Sickle
  • 35 posts

Posted 07 August 2018 - 07:14 PM

Quote

• Engine de-sync did what it was designed to do.

• Some Heavies and Assaults on a case by case level will be looked into for mobility increases


As someone who enjoyed playing light mechs significantly more before the desync, these bullet points make me sad. Aside from a few high performing mechs, the ability to take on heavies and assaults has been harmed by de-sync, and there is little benefit for heavier mechs to bring bigger engines anymore as the speed benefits are negligible. Further, desync- harmed the engine locked clan heavies/assaults that benefited from the ability to twist faster.

Edited by CommEE, 07 August 2018 - 07:14 PM.


#132 Mad Dog Morgan

    Member

  • PipPipPipPipPipPip
  • The Raider
  • The Raider
  • 489 posts
  • LocationOutlaw On The Run, Faster than a Stolen Gun

Posted 07 August 2018 - 07:17 PM

Now, I know I'm prolly being a bit wishy-washy here, but I'd really like to see small and large pulse lasers (clan and IS) pushed more towards burst than sustained.

Baseline accel/decel and twist needs to be brought up across the board imo, for lights and assaults especially, maybe even make max speed tweak 10% again.

I'd like to see brawling back in the game. Sans artemis, a lot of SRMs aren't really worthwhile, UAC-20's aren't worth their tonnage either for the prohibitive heat or jam chances, not to mention they significantly spread damage for the clan UAC-20. Direct-fire sustained ballistics could use some love.

And yes, Light Gauss is a huge underperformer.

Tweaking lasvomit I guess has some merit but there are ways to mitigate it without severely impacting their overall performance.

Edited by Mad Dog Morgan, 07 August 2018 - 07:18 PM.


#133 TheSteelRhino

    Member

  • PipPipPipPipPipPipPip
  • Overlord
  • Overlord
  • 600 posts
  • LocationTexas

Posted 07 August 2018 - 07:17 PM

Hey PGI peeps,

I have 2 words...HEAT PENALTIES.

You don't have to nerf the damage, or the wep. But make those high heat levels DANGEROUS.
Exploding ammo, exploding heatsinks. Throw in some damaged capacitor on energy based weps that renders them useless if the heat is to high.

Tabletop always had that risk if you drove that heat to high. Lower the shutdown/override threshold some. I have no problem with a serious high heat, high energy alpha causing a shutdown. And make the damage HURT.

How about damage to the computers? If you ever worked in a datacenter you would learn heat is bad for computers. Posted Image What if you suddenly lost your targeting computer? Or your HUD starting flickering or just died. Effectively makes it much harder to target with that pinpoint damage. Sure you can correct by mark one, mod zero eyeball, but that spreads damage around which helps to negate the advantage of having that high alpha.

That way you don't have to nerf the weapon itself and impact smaller mechs that run fewer energy weps, but you are making the pilot MAKE A CHOICE. Do I alpha and risk some real damage and then back off for 7 seconds while my heat and weps cool down. Is it even worth it anymore? Make ppl make that choice.

Seems extremely logical, and not that hard to code honestly. You already have a heat scale. Just add some code logic that at such and such heat you have X% chance of damage and determine what type ..and I don't just mean internal structure dmg. In short, make high heat alphas RISKY. The systems are essentially in place already I think from the behavior you see in game.

Edited by TheSteelRhino, 07 August 2018 - 07:39 PM.


#134 Bosseen

    Rookie

  • Philanthropist
  • 2 posts

Posted 07 August 2018 - 07:18 PM

View PostMrSomaru, on 07 August 2018 - 07:05 PM, said:

Nope, but we know the answer: Take Polar Highlands to maximize cold lrming.


Well that sort of is my point. A specialized mech can perform much more efficiently on certain maps then others. So when a person tends to queue for quick play they will most likely vote for the map that makes their weapon platform the most effective. That leads to map design. Now it is generally difficult to make a mech that works on all maps equally effective because of how each map is designed. The map being voted for and how it is designed highly influence that particular impact a certain mech with a certain loadout might have. Hotter maps generally are less agreeable with your laser boating mechs while your missile boats enjoy the wide open spaces. Proper map design can uplift or diminish weapon loadouts I believe much more efficiently than simple number changing.

Edited by Bosseen, 07 August 2018 - 07:20 PM.


#135 Alilua

    Member

  • PipPipPipPipPipPip
  • FP Veteran - Beta 1
  • FP Veteran - Beta 1
  • 362 posts

Posted 07 August 2018 - 07:18 PM

Regarding the better communication and using the test server data is really great from the community standpoint and I hope that this level is kept up.

ECM buffs might not be the best choice in reigning in the lrms since it has been in a rather good spot with range when it’s skilled, but if it’s unskilled the equipment is really weak; perhaps there is a way to just bring the unskilled and one skilled closer to the effectiveness of fully skilled. I wouldn’t change the maxed out effectiveness too much since it took quite some time to get it to the state that it’s in now.

Damage discrepancies between faster lights/mediums and heavies/assaults is going to be the hardest thing to balance since damage is one of the most important things in a match. I personally liked seeing different weapon quirks for the mechs but I can also see the downsides.

All of the recent ams changes that have been put to live were good starts. How happy are you with the difference in effectiveness between the laser ams and ammo ams? Sometimes it feels like the lack of cooling is nearly as hash as actually gaining heat from using it. Are there plans to see a c-bill related reward tied to using them? I think now it gives score but not really c-bills and that might be a good way to encourage more usage of systems.

Changes to lasers and such are obviously going to eventually happen, but are there any plans to look at autocannons still? There is very few reasons to take the ac5 over an ac2 on the IS as well as clan standard auto cannons being stuck in a cycle of no reason to use since they were introduced. I am kind of more for a huge set of changes being thrown up on the test server than doing balance a few weapon types at a time. I guess I just want to see things not necessarily be left alone for long periods of time when there are longstanding gaps in performance.

There seems to be an idea that to get an idea of balance requires a vast amount of play time but were you happy with the amount of community feedback and testing done with the tests? Did the award for testing work well enough to attract the number of players?

Enhancing current weapons and systems through the use of targeting computers hasn’t really been changed too much since launch of systems on both sides. Would adding additional benefits to targeting computers and command consoles be a possible change or are those benefits difficult to change around. It seems that certain benefits could be added like increasing lock angle and time adjustments to let some of the recently removed aspects softly remain at the cost of equipment slots and tonnage. I wouldn’t go for full scaling with the weight as much as heavier targeting computers give maximum benefits while lighter computers could give just a few benefits that still make a difference. Like a level 1 targeting computer could be a moderate increase in targeting time loading, a level 2 would be a great increase, and a level 3 includes something like a moderate weapons lock angle, level 4 could add some moderate missile lock time level 5 might bump it up to great increase in lock angle.

Despite the rather harsh backlash that sometimes happens I hope these upcoming changes help the game out and satisfy the needs of the community. As for my own personal thoughts just read the first letter of each paragraph.

Edited by Alilua, 07 August 2018 - 07:21 PM.


#136 Y E O N N E

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Nimble
  • The Nimble
  • 16,810 posts

Posted 07 August 2018 - 07:19 PM

Before I address my thoughts on laser vomit, I want to note something that I think is important to the game at a very fundamental level that I sometimes feel PGI, and even members of the community, loses sight of:

Clans and IS in MWO are supposed to be equal but different.

I do not like this idea of lowering IS durability and Clan alphas. I do not want to mutually exclusive sets of equipment that barely play any different from each other. What is the point, then? I would rather PGI go the route of creating philosophical differences between Clan and IS effectiveness, and then emphasizing those differences.

For my part, I like the concept of Clans generally being glass cannons where the IS are the sturdy workhorse. Clan 'Mechs will generally hit harder with more of a focus on evading hits where IS can hit more frequently with a focus on sustaining hits. Clans have raw power, IS rely more on precision. I will revisit this later.

View PostPaul Inouye, on 07 August 2018 - 10:18 AM, said:

.
• Current focus is on the 94 efficient alpha and the discrepancy between Clan and IS laser vomit performance against one another.


View PostPaul Inouye, on 07 August 2018 - 10:18 AM, said:

. The 94 alpha that we are very concerned about cannot be reduced in a significant way without these extreme numbers. That being said, having to go to these extreme values is causing a net change that is not favorable both internally and publicly which means that we will be looking at other avenues to address the situation. We may use numbers from the PTS that reduce this 94 alpha to something like 90 (EXAMPLE NUMBER ONLY) instead of trying to get it to 80 or some other drastic number. After that change is made, we will be looking at other possible means to go from the example 90 to something lower.


Which 'Mechs are running 94-point alphas too efficiently?

Let's examine our two best options:

Dire Wolf:
- 2x cGauss (4 tons ammo)
- 2x cERLL
- 6x cERML
- 23x cDHS
- TC1

88% heat

Pros:
  • 94 point alpha strike
  • Longest duration is 1.35 seconds
  • 100 tonner armor
  • Can lean on cGauss when hot
Cons:
  • Runs at 48.6 kph
  • Low weapon mounts
  • Poor hit-boxes
  • Poor agility
  • One alpha is 88% of the available heat capacity
Is this too efficient? This has as much give built-in as you could reasonably ask to counter-balance a taking large alpha, to the point where it hasn't been a dominant force in any mode of play for years now. No Dire Wolf build has been a source of angst in recent memory, lasers or otherwise. It hardly gets used even with 94 alpha.


Mad Cat Mk. II Deathstrike:
- 2x cGauss (3.5 tons ammo)
- 2x cHLL
- 4x cERML
- 18x cDHS
- 1x JJ

Pros:
  • Runs at 58.5 kph for the same alpha as the DWF
  • Can lean on cGauss when hot
  • High weapon mounts
  • Moderate agility
Cons:
  • Longest duration is 1.55 seconds
  • One shot is 95% of the heat cap
Is running 10 kph faster worth only being able to alpha once every 18.5 seconds or so? There is a reason why the meta build for this 'Mech is using 2x cERLL instead, and it's because the 94-point version simply isn't sustainable unless you knock the 'Mech down to almost DWF speeds anyway. And 1.55 seconds with that big frontal profile and no armor quirks certainly doesn't throw anything in its favor.

I can't fathom that the DWF is causing PGI any consternation right now, the Blood Asp and Kodiak do it even hotter with even worse geometry, so I can only reasonably conclude that it's the Deathstrike causing the heartburn. If this is the case, then there is a much simpler and more focused solution in simply removing a single energy hardpoint.


View PostPaul Inouye, on 07 August 2018 - 10:18 AM, said:

• Lasers aren't being targeted because they are the considered the highest performers so much that there is a clear rift in viability of energy centric load-outs between the IS and the Clan side.


There are two reasons for this:
  • IS lasers have the same damage-to-heat ratios too similar to their Clan counterparts at the mid-range and long-range brackets but with much less cooling capacity
  • IS lasers have lower maximum DPS - often significantly lower - for any given range-bracket
What both of these things mean is that, even though PGI has sort of tried to give IS more of a pressure-play feel by letting their weapons cycle faster instead of dealing more up-front damage, they cannot actually use that. They don't have the heat efficiency and they don't cycle quite fast enough. An IS laser team trying to pressure in the open like that gets deep-fried by volleys of high alpha Clan lasers. Even in their strongest bracket, between 200 and 350 meters, IS MedLas+Large Pulse boats do not compare favorably to Clan Medium Pulse+Large Pulse boats...it's just that Clans seldom bring such builds because they have an even larger baseline advantage from further away.

Note, this is not necessarily about looking at lasers 1-for-1, you have to look at the entire build. A Clan laser might appear inferior to an IS one in heat efficiency, but it is never enough so that it overcomes the ability Clans have for stacking 25+ DHS even on Mediums. A Clan laser might appear inferior to an IS laser in terms of how much damage it does per second of burn time, but you consider the agility of the target and the IS duration advantage doesn't count for much.

You also can't necessarily compare two lasers of the same size and class; a Clan Large Pulse is in a different bracket than the IS Large Pulse, being more comparable to a standard IS Large Laser in every way: range, max range, duration, cool-down. As another example, massed Clan Medium Pulse Lasers are more akin to bundles of IS LPL and IS MedLas than they are to bundles of IS MPL. It would behoove PGI to take advantage of these more natural, emergent similarities than to try and fit a square peg into a round hole and make two guns compete against each other just because they share a name and a class.

There are two caveats to the above:
  • IS Assaults will always be at an disadvantage when it comes to laser vomit because they are no more potent than IS Heavies but are much bigger targets; Clan Assaults can afford to cool significantly larger volleys
  • All of this gets flipped on its head when you talk about Lights

IS Assaults cannot accelerate, decelerate, or twist fast enough to spread a competently played Clan laser build. They will almost always lose this trade. They lose trades even against IS Heavies, which are much faster and often have better hardpoint placement to exacerbate things.

With Lights, because IS Lights take XLs same as their Clan counterparts (where Mediums and up typically do not, too risky for not enough reward), and because Lights are tonnage-limited more often than they are slot-limited, the advantage in cDHS count no longer exists. So IS can make comparably potent volleys from similar ranges and they can do it faster and with less heat. It's almost the complete opposite of what happens when you compare heavier 'Mechs from the two factions. Strict ghost-heat limits needed to prevent the bigger 'Mechs from getting out of hand do nothing to help the situation for Clan Lights. That said, IS also do not have any MG-boating Lights which can compete head-on with their Clan counterparts (FLE-19 lacks DPS, FS9 Ember lacks torso pitch) , so make of that snake-vs-mongoose situation what you will.

View PostPaul Inouye, on 07 August 2018 - 10:18 AM, said:

• It has never been the intention of MWO online to have Time to Kill in a state where you can destroy a mech in only a handful of shots.


So, PGI has an average TTK they are targeting, but PGI needs to ask how they want to obtain that TTK. Does PGI want to reach that TTK passively through increased durability and/or increased hard limits on damage output, or does PGI want to reach that TTK by increasing player agency through higher agility and increased threat level?

It is my assessment that PGI has been pursuing the passive option more so than the agency option. That works for new players, who don't yet have the skills to really take advantage of the options they are given, but it's an incredible snooze-fest for veteran players because it just becomes a game of zerg-rushing the enemy with high-armor, high-sustained-DPS builds. That's what QP is right now.

I would prefer to see PGI pivot more towards the agency option with increased agility and increased lethality of certain range brackets.

View PostPaul Inouye, on 07 August 2018 - 10:18 AM, said:

• Will not be blindly balancing upwards to meet Clan Laser vomit without proper give/take and taking into consideration all other areas of balance.


One issue that some of us have noted is that, almost every time PGI makes an adjustment to something, an adjustment is also made to its complementary component.

This does not make sense.

For example, when PGI nerfed the cool-down on Medium Lasers, an event which was also instigated by the power of the Deathstrike as well as the Supernova, they nerfed both the Clan and the IS options. There was a gulf there that needed to be shrunk (and still does), and instead it was maintained.

In this case, if TTK is too low on account of high laser volleys even with armor quirks, does it make sense to lower the values of the latter while lowering the values of the former? You'll lower Clan volleys to be more similar to IS volleys, lower IS armor to be more similar to Clan armor, and then we're left with Clan heat efficiency running the rest of the show...more or less like how it is now when it comes to laser vomit per my earlier statement. How long are you willing to make those Clan laser durations to offset this? There are absolute limits you can't really go past without things feeling clumsy and awkward and not fun to play: see also reasons why the Heavy Medium, Heavy Small, and Micro lasers are not popular.

Quote

• While upwards trajectories always sound good, there are consequences that would massively disrupt other states of play within the game. Such as massively buffing armor to compete against higher alphas out of heavy and assault 'Mechs, but then light on light mech gameplay would take forever to whittle through all of the armor with low weapon counts and high armor values.


I'm less concerned with Light-on-Light, since that is easily rectified if you make such a durability buff through the skill tree and give increasing percentages to increasingly heavier 'Mechs. More concerning would be Light-on-Heavy/Assault. If the armor gets too high, your popguns don't have any credible effect within a timely fashion. That one is harder to rectify, you'd have to straight power-creep the weapons on them via quirks.

View PostPaul Inouye, on 07 August 2018 - 10:18 AM, said:

• Engine de-sync was never designed to replicate the old 'Mech mastery system, but replace the baseline attributes prior to skill investment. (So a 'Mech with zero experience under the older system.)


See, this is not what the community had in mind when we suggested engine desync over the years. What we had in mind was using it to give every 'Mech the kind of agility that you got when you equipped the ideal engine size on the ideal 'Mechs, that way you weren't forced into taking a big XL to get the kind of damage-spreading capabilities you needed to be competitive. That's what we had in mind. Instead, what PGI did was use it as an opportunity to bring down baseline agility; you cannot get the kind of agility we had before at all even by maxing out the skill tree (and that's why so few people do so).

This loss in agility, as well as the reduced duration of lasers with the introduction of Laser Duration nodes, has directly contributed to many of the problems with laser vomit being faced today. Prior to Skill Tree, and even prior to Kodiak, nobody batted an eye at a 108-point alpha Dire Wolf because it took longer to burn, because agility was high enough that one could twist that damage effectively, and because IS durability was not restricted to a special handful of 'Mechs (IS laser vomit alphas were also larger and colder than today, so that helped).

View PostPaul Inouye, on 07 August 2018 - 10:18 AM, said:

• Prior to the velocity buffs the LRMs where a sub par weapon system at most levels of play. And even with the most recent buffs to the weapon system, they are still a fairly sub-par weapon system over the general lineup.


The issue is less that it's effective, more that the stagnant play it encourages is not fun. It's not unlike when laser vomit was the one-true-meta back in late 2014 through mid 2015.

View PostPaul Inouye, on 07 August 2018 - 10:18 AM, said:

o NARC seeing a 100% increase in cooldown.


The NARC duration is slightly absurd, especially when fired from a RVN-3L; I feel that's where you should be looking more so than cool-down. Getting hit by a NARC on a map like Polar or Caustic pretty much takes you out of the fight if the enemy has competent LRM players, either because you are dead or because you have to sit out of range or under cover for prolonged amounts of time. It's not fun.

At the very least taking advantage of short cool-down requires the NARCer to be exposed and engaged.

Edited by Yeonne Greene, 07 August 2018 - 07:20 PM.


#137 Ridingwolf1

    Member

  • PipPip
  • The Wolf
  • The Wolf
  • 27 posts

Posted 07 August 2018 - 07:26 PM

View PostBosseen, on 07 August 2018 - 07:00 PM, said:

Has there been any discussion about how the actual maps are designed and could be influencing different weapon balancing issues?


Yes. I have tried to discuss this in a previous post along with the thought that the issue is pilot skill based and not weapon damage based, but was told this was not appropriate to the conversation.

#138 Andi Nagasia

    Volunteer Moderator

  • PipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 5,982 posts

Posted 07 August 2018 - 07:47 PM

Ideas to help LRMs (dont like the Proposed Nerf to Artimus as it removes its use)
Spoiler


#139 Marius Evander

    Member

  • PipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 1,113 posts

Posted 07 August 2018 - 08:00 PM

View PostBud Crue, on 07 August 2018 - 06:51 PM, said:

Spoiler



Well said.

Another problem is.. perfect balance (which is impossible unless everyone pilots the same mech with the same build and loadout) would get very stale very fast. A "Balanced" rock paper scissors is the best we should hope for.

Heres an idea, make ams fixed equipment on all mechs / components that can carry it and buff your beloved LRMS some more. I hate it, but its an idea.

On a more serious note, the increase is LRM/AMS usage, ATM's need a health buff to stay viable when ams is meta.

Edited by Cadoazreal, 07 August 2018 - 10:21 PM.


#140 Elizander

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 7,540 posts
  • LocationPhilippines

Posted 07 August 2018 - 08:04 PM

Suggestion for Laser Duration (or other weapon stat) changes affecting balance between light/medium and heavy/assault
  • You already have a solution for this instead of a full quirk rework. Just make the laser duration and heat nodes in the Firepower tree scale with weight or weight class. If you make Clan ER Medium Laser do 1.60 seconds base duration, then maintain the -10% laser duration nodes for heavy+ and increase it for smaller mechs to adjust as needed. The range could be -25% laser duration for the smallest mechs and maintain -10% for larger mechs. Scaling can start at 55 tons and lower while anything at 60 tons up is locked in at 10% duration reduction
  • This can be done for other stats as well like heat and cool down. You already do it for the survival tree so I don't see why it can't be applied for other trees like Firepower.
I feel that the argument about deleting someone is lacking different perspectives and is only a case for a 50 ton mech being destroyed by a 90 ton mech from fresh via CT kill when it is not twisting.
  • 4x SRM6 can instantly delete a lot of mechs from behind if it is close enough. I've done this a lot on my Griffin 2N.
  • Headshots with PPFLD can delete mechs who stand still or shut down. I've had other players intentionally try to get headshots on me if I was standing still too much.
  • Having one single heavy gauss rifle can delete a fresh Flea.
  • Going strictly with math, a 96 point laser alpha strike that is not twisted within the 1+ second burn is not enough to kill a 55 ton mech without crits. A 55 ton mech with no quirks to armor or structure will have 104 total HP to blow out the CT assuming you go extreme with like 4 back armor. Even with 10 back armor it's still 98 hp to 96 damage. We are talking about a 90 ton mech doing this to a non-moving 55 ton mech. Anything bigger will have even more health to survive without twisting.
PGI can use "Shield Gate" or "Health Gate" (or call it Armor Gate) mechanics to increase TTK and negate one-shot wonders and future proof anything from being one-shot. This can maybe be applied to front armor only to still allow sneaky mechs to take targets out quickly as a reward for getting behind them.
  • Mass Effect uses shield gating mechanics where damage that blows past shields that are still active are reduced by a certain amount.
  • Warframe has a shield gating mechanic that temporarily negates damage done to health when shields are destroyed.
  • Borderlands has a health gating mechanic where you cannot be one shot as long as you are above 50% health.
  • MWO can have Armor Gating mechanics where as long as you have a single point of armor in the front then any excess damage to structure is greatly reduced (50%-90%) and no critical hits are allowed for a brief period of time (0.5-1.5 seconds) to reduce the impact of high alpha while still allowing high alpha. This can easily be reasoned out with ablative armor used in Battletech. You can start with something like 50% damage reduction and have it go up higher via skill tree or quirks. This will also prevent lights from being one shot by Heavy Gauss rifles if the numbers are right.
I feel like PGI is chasing its own tail with these string of nerfs. Nerfing clan lasers to nerf IS armor and then buff LRM to nerf LRM subsystems. It feels like it's just a merry-go-round or see-saw to me. It might be that the overall message on what is desired of these changes is not clear to me and others. It personally feels like change is just being made for the sake of change and the practical end goal of PGI for these changes is not being properly conveyed. I feel that the simpler solutions are being overlooked in favor of more complicated chain-nerfs that drag on for months.

Edited by Elizander, 07 August 2018 - 08:07 PM.






14 user(s) are reading this topic

0 members, 14 guests, 0 anonymous users