7 replies to this topic

[DivineEvil]

Since PTS servers went online, I've begun collecting feedback from my unit members and formulating it into a comprehensive shape. Still, very few people aside from me were able to participate in active testing when there were enough players present to effectively perform testing runs, thus the major part of the following thesis is still my own view with an addition of other opinions, mostly based on discussions over raw math and shared screenshots. Unsurprisingly, many TCAF members, just as a vast majority of MWO public, were unable to comprehend the idea behind the first testing phase. To respect PGI intentions, these opinions will be left out. General points will follow:

Sensor/Targeting quirks and changes are major positives. Without a doubt, there were a need for information-related modifiers to incentivise role warfare and increase Light mech efficiency in performing flanking and scouting. With the upcoming nerfs to the ECM, these are going to make a big difference. One thing that bites me, is that base delay to the target spotting is not obvious. It might worth to implement a "signature" parameter to the mechs, together with available Sensor Range, to illustrate the base number of seconds of visual observation required to get a lock on a given mech without considering quirks.

It is unclear why PGI have decided to eliminate IS weapon quirks entirely from this testing procedure. It is a major undermining factor to the whole idea of changes. It is by all means true, that specific weapon quirks have been narrowing down the customization options for particular variants. It might also be true, that the total magnitude of both specific and general weapon quirks when combined are driving some variants out of the picture. But the presence of these issues have never been as severe as to nullify them. Firepower aspect should've been generalized and toned down, and quirks touching different aspects of weapons should've been given to the mechs still. Lack of these draft general weapon quirks made mechs intuitively unrecognizable in live performance from one-another and incomparably worse than their present unchanged relatives.

Russ Bullock have specifically stated, that 'mechs will be evaluated on four designated values - Defense, Firepower, Mobility and Sensors. Taking out one of these parameters out simply made no sense in a frame of these mentioned statements, no matter how early it is. Changing the methodology without even starting to follow it is counter-productive!

Character and/or values of new quirks in present form are inconceivable. The variety of different quirks, that affect a multitude of variant' parameters, makes the balance analysis impossible. At this point is almost futile to try to efficiently detemine where quirks are unbalanced among the mechs. Even superficial understanding of balance between either two different variants or chassie is impossible to reach to without using third-party datasheets or obscene amount of time of trying to recall all the different values. Long story short, first iteration of quirks is inconvenient by relying less on presence/lack of particular quirks and more on actual values of those quirks. It blurs and scatters the apparent differences between variants and mechs. Several quirks affecting the same DFMS aspect should be only applied, if otherwise applying one quirk with the same total magnitude will create a critical case. There's should be some different quirks, rather than many different values of those same quirks!

The amount of quirks makes the whole system unbearably tedious to use. When approaching the strong and weak sides of a variant or a chassis, it is mportant to remain subtle in displaying them. Supplementing the past paragrapth, using quirks to modify every particular aspect of a mech's parameters is obscene. Using one line for each component's structure value, rather than using one percentage value to increase Structure Strength, hurts the visual part of the quirk system and doesnt make sense from the balance standpoint. Arms, Legs and Torsos are all different components and should have different benefits from corresponding quirks. Specific local bonuses are only to be used in separate cases (Centurion's Shield, Dragon's Belly, Cat Ears, Hunch etc.). Same goes with Target Scan Time@Range; Basically it doesn't lights the bulb what are those three range profiles are, and why the disparity between Short Range and Long Range buffs/debuff is so large. It is hoped for them to be simplified to a single, general quirk of Target Info Delay, which would affect all ranges equally.

Quirk Handle and Variety proposals:

Spoiler

Defensive:
Torso Turn Rate +10% (it's a defensive trait, it doesnt affect mech's mobility in any way)
Additional Structure(CT) +8p (points)
Additional Armor(CT) +8p (points)
Armor/Structure Strength +8%
Fall Damage Reduction +10%
Damage Tolerance +2p
(virtual damage treshold, that mech can withstand before taking actual damage, refreshed in N seconds of not taking any damage)
Structure Integrity +10%
(Reduction of internal ammo and weapon explosion damage)

Firepower:
Energy/Missile/Ballistic Heat Control +10%
Energy/Missile/Ballistic Cooldown +10%
Energy/Missile/Ballistic Accuracy +10%
(combines Laser Duration and Projectile Velocity respectively)
Energy/Missile/Ballistic Range +10%


[Unique Quirks for mechs with unhandy missile/ballistic hardpoints]
LB X-10/SRM/LRM Precision +10%
(replaces Spread quirks)
Ultra AC Stability +5%
(replaces UAC/5 jam chance to work with Clan Ultra Autocannons when administered)


Mobility:
Torso Turn Angle(Yaw) +30° (degrees)
(JumpJet Thrust? JumpJet Capacity? JumpJet Recovery?)

Sensors:
Target Retention time +2s (seconds)
Target Aquisition Delay Target Spot Delay +2s (seconds)
Target Scan Time Long/Medium/Short Range Target Info Delay +30%
Sensor Range Modifier +/- 20%

Negative quirks should be displayed in pastel orange, not in crimson red! Red font is straining the eyes and does not looks good with pastel green, that is used for Positive quirks.

EXTRA: Example methodology for developing role-based quirk variety:

Spoiler

This example represents my personal idea of mech role definitions at the moment when MWO quirks were first implemented, to this day. It is my own outlook, and I'm not a game developer in any way, thus any part of it can be argued and changed if worth any attention to begin with. It is given in desperate hopes for PGI reading and evaluating it, and potentially developing a certain philosophy, that would drive game development and balance process in a reasonable, sophisticated direction.

What first comes to mind when seeking for a variety of all the existing mech designs in Battletech is mech's Role. All machines were made for a specific function, thus the most significant prefix for one chassis or the other is it's role, model of function and individual traits given by lore and techincal specifications. Collecting all the more or less known and developed mech models from Battletech, I've came with the small table, that might be used to determine which quirks are more reasonable for being a baseline to accomodate every variant of said mechs. Here is it:

All the mechs, that are not implemented, are provided to pre-plan for their potential addition in future without dramatically changing the parameters of the present mechs. Some of these might in fact not even be considered for implementation, though.

Using this example might resolve the excessive pressure of developing the divergence within each weight-class. As mentioned earlier, the role guideline might be used to clear out the idea of which quirk a mech should get on every of its variants. In order to further differentiate the specific variants of mechs from one-another, a secondary role is derived from mech' known features. It is used to accomodate the primary role by adding quirks with another aspect to a lesser relative degree. For example, a sturdy, brawler-dedicated Stalker can still have quirks for direct and indirect fire support and even some additional maneuverability, depending on the variant in question, but these quirks will be 30%-60% less significant, compared to the primary ones.

When the draft idea of mech' primary and secondary roles is considered, it is time to look closer to the mech' physical attributes to determine the required quirk magnitude. The same way as PGI have already unveiled, each mech and it's variants is studied for strong and weak sides it has: maximum engine, angular limits, number and location of hardpoints, profile scale, JumpJets, ECM - everything is considered to determine a relative strength factor. This strength factor is repurposed into a working variable, or just a "quirk points". A default value of quirk points is set for every variant, and depending on the summary of strong and weak sides that the mech/variant has, additional points are given or taken from each.

In order to make quirk points to actually mean something, a separate system of equations is made, that would tell, what value a given quirk should get for a price of one quirk point; How much armor, how much turn rate, how much weapon cooldown an each of those points worth, depending on how much these quirks affect any given mech. These values are then set to work as multipliers, that can be then tweaked to better reflect their effect and the game balance in general, without stressfully fixing the values for each given mech. For example, if one quirk point equals 2,5% Energy Range, then 10 points delegated into that quirk on a mech results in a 25% Energy Range bonus, but if the value is then toned down to 2,0%, quirk will give 20% Range instead. This change will affect all mechs, that has Energy Range quirk on them, without manually fixing them for every mech.

When this is done, prepared quirk points are translated into mech variants by prepared role definitions with prepared per-point values. For each of the four aspects, only two to three different quirks are used to make the mech features more obvious to the player and to maintain general mech variance. If resulting values for a quirk are too high, they're divided between the two quirks of the same aspect, or in some cases delegated into a secondary aspect. If they're too low, then negative quirks are also added into role-unrelated aspects, which would give back some extra points to maintain contrast.

In case of Clan mechs, the opposite is done - for each chassiss, the alien aspect is chosen, which would represent something that mech is definitely not affiliated with by it's design. Each chassis is studied for physical features and the same quirk points are given to the flawed and underperforming mechs. Based on them, each Omni-pod is given some negative and positive quirks, that would make them better for one role and worse for the other (with an exception for the chosen "alien aspect"), allowing Clan players to fluidly customize the mech' role and it's depth, applying additional buffs for empty omni-pods and negatives to the omni-pods with special equipment and higher numbers of hardpoints.

Running test sessions trough that kind of a mathematical system can result in a interconnected and easily manageable and customizable structure. Changes can be done to the particular variants, mechs and the underlying general values and affecting the balance to the required extent and scale. The End.

EXTRA: Potential ways to address the general MWO issues:

Spoiler

There's three main widely known issues present in the game. Unfortunately, vast majority of players are unable to grasp their scope; to this day, all of these issuaes are suggested for being adressed with very narrow tweaks to specific value. I personally do not share that opinion - general problems has to be addressed with general solutions. These issues are:

Determined high-alpha based meta-game. Changing values for particular weapons or weapon-related value will not fix it! The easiest way to change the situation is to reduce the convenience and efficiency of high alpha-strikes. This can be done by reducing or eliminating the base heat capacity each mech has, while increasing the heat dissipation. As you might know, each mech has 30 points of heat capacity built-in, without considering engine and external heatsinks, that are added on top of that value. Placing a 250+ rated engine with Double Heatsinks bloats this value up to 50 units. Heat containment skill adds another 5 units, and unlocking 2X basic skills doubles that value to 10, which results in a 60 heat capacity.

This resulting value is the root of a problem - it creates a window for effective functioning of so-called "laser-vomit" mechs. It does not affect any other weapon types anywhere near, as they are harder to use, larger to fit, heavier to boat and/or limited by Ghost-Heat mechanics. At the same time, there's no such thing as "base heat dissipation", as cool-off rates are directly dependant on the amount of heatsinks alone. This makes brawling and any prolonged combat doomed to failure. Each particular heatsink installed or removed do not have as much effect on combat performance as a ton of ammunition or an additional Medium Laser placed in its place. Additional heatsinks are often installed just to use the free tonnage in place of a weapon, that would make the build too hot for high-alpha playstyle; The actual effect of these heatsinks on heat management is negligible.

Considering that, reducing the heat capacity would encourage MWO players to avoid conventional high-apha builds and instead to build mechs with different combinations of weapons, that would allow them to be relevant in different situations. Only one or two laser-vomit strikes would be usable before overheating any given mech, thus Missile and Ballistic weapon will recieve more attention. Laser boat will still be useful in framework of chain-fire attrition and hit-and-run tactics. This will be provided by the increase of heatsink dissipation rates, that would allow energy-heavy mechs, such as Awesomes, to use volley fire for continuous barrages of their weapons, such as PPCs. Depending on the effects of this change, Ghost-Heat mechanics can become redundant, as it were meant to address the same problem.

Low Time-to-Kill rates. The origin of that issue lies with PGI's original decision to address the transition of Tabletop value to the FPS environment. Armor values have been doubled across the board, which have icreased the apparent durability of mechs. Unfortunately, structure values have been left out, which made the actual effect to be around 66% of increased durability (150% total), rather than 100% increase(200% total). Not only that, but leaving internal structure out of equation have made the effective result even less impactful when taking Critical Strikes into account. Weapons with critical strike multipliers are still considered inferior, since relatively brittle internal structure is just as easy to destroy with any other weapon. Their ability to cause equipment failures and ammo explosions is irrevelant, as a whole component readily crumbles under a mere fraction of firepower, that were used to reach it, anyway.

On top of that, weapon cooldown values have been also increased, and by the large margin the increase went over 250% or more (4 seconds reloads for heavy weapons compared to 10 second round length in TT for all weapons). This change has completely nullified all the benefits, that the armor growth were trying to achieve. Even more, the FPS nature of MWO obviously neglected the random-chance nature of TT combat almost entirely, which cut the TTK even shorter, especially in competetive scene of the game.

Obvious and simple solution of that problem is to double the internal structure as well. This will increase mech' life length and will increase the relevance of critical damage mechanics, and weapons/equipment, that are tied to these mechanics. It will also remove the need for "additional structure for everyone!!!" quirks, that we were seeing in the first PTS re-balance launch; Only the mechs with oversized components and/or Defensive roles would require any quirks adressing armor and structure values, especially if prior Heat Management changes will also be incorporated as well.

Finally, Clan VS Inner Sphere balance bias. That problem, ironically, can be adressed simply by tweaking either of two previously mentioned operations in favor of Inner Sphere Mechs. If only Inner Sphere mechs would get double internal structure, or if only Clans would lose their virtual Heat Capacity, this would create a significant rift between the concepts of two Factions. Whether Clans would be more compact and mobile, yet more brittle, or if they would get more powerful weapons, that are harder to use and manage with less heat capacity, the balance can shift in a place that most people are anticipating. Even more, if both changes are administered together in favor of IS mechs, the changes can be so dramatic, that there will be no need to bother with any weapon balance, and both IS and Clan mechs can be quirked on equal terms with similar amount of positive and negative values.

Nevertheless, no particular changes to weapon or mech stats will fix these issues. Both methods and alternatives are, in my opinion, worthy of considering and brought to the PTS trials.

Thank you for reading!

Edited by DivineEvil, 14 September 2015 - 02:49 AM.

[Nexano]

Hope this topic won't get lost, and PGI see it. Whining in other topics not very informative.

[Krivvan]

DivineEvil, on 14 September 2015 - 02:18 AM, said:

Finally, Clan VS Inner Sphere balance bias. That problem, ironically, can be adressed simply by tweaking either of two previously mentioned operations in favor of Inner Sphere Mechs. If only Inner Sphere mechs would get double internal structure, or if only Clans would lose their virtual Heat Capacity, this would create a significant rift between the concepts of two Factions. Whether Clans would be more compact and mobile, yet more brittle, or if they would get more powerful weapons, that are harder to use and manage with less heat capacity, the balance can shift in a place that most people are anticipating. Even more, if both changes are administered together in favor of IS mechs, the changes can be so dramatic, that there will be no need to bother with any weapon balance, and both IS and Clan mechs can be quirked on equal terms with similar amount of positive and negative values.

There are parts that I disagree with and parts that I agree with, but that section is what I agree with the most. Reducing heat capacity, possibly only for Clan mechs, is what a lot of us "meta tryhard" players have been asking for. Big offenders like the Dire Wolf will be less able to throw out 86 damage alphas over and over and over again to the point that it overshadows any other assault.

We all know that bracketed builds are terrible. If Clan mechs were a forced to delve into that a little bit, it really would be a significant Clan nerf that has the added advantage of balancing the upper tiers without making life terrible for the lower tiers.

Edited by Krivvan, 14 September 2015 - 12:15 PM.

[Nothing Whatsoever]

I'd like to see the Heat System addressed at some point. Here's where I'm starting to lean towards:

Reduce the value of 30 that you mention to 14
Disable Cool Run and Heat Containment
Change Heat Scale Penalty buckets to better account for damage above say 30 to 40 Damage, including more weapon combos
Raise Dissipation and Adjust Capacity with Cool Run and Heat Containment being disabled and replaced.

SHS
0.25 Dissipation
1.00 Capacity

IS DHS
0.40 Dissipation
2.00 Capacity

Clan DHS
0.30 Dissipation
1.60 Capacity

This way 28 SHS would provide 7.00 Dissipation a second and 42 Capacity to the First Shutdown we can Override. For comparison firing three PPCs with current AWS-8Q PPC Heat Gen Quirks and Heat Scale Penalties factored is ~31.95 Heat, HPS is in the range of 7.98 to ~9.98 depending on Fast Fire, cooldown quirks and Weapon Modules. Then two PPCs would be 15 Heat, so the player would need to alternate firing to avoid shutting down, along the lines of 3/2/3 and so on.

20 IS DHS would provide 8.00 Dissipation and a Capacity of 54 Capacity to the First Shutdown we can Override. If we consider the AWS-9M with its ERPPC quirks, an alpha of all three is ~47.93, with an HPS of ~11.98 to ~13.69. and so would need to be careful.

28 Clan DHS would provide 8.40 Dissipation and a Capacity of 58.8 to the First Shutdown we can Override. If we consider the Warhawk Prime. Firing three would spike heat really close to shutdown at ~58.79 in most maps and if moving. Firing two is a spike of 27.6 with an HPS of ~6.90 to 7.26.

An Adder Prime with 15 Clan DHS has 4.50 Dissipation and a Capacity of 38 to the First Shutdown we can Override. Firing two ERPPCs is a spike of 25.5 with an HPS of ~6.38 to 6.71. A single ERPPC is 12.75 spike with an HPS of ~3.18 to ~3.35. So much more sustainable than what we currently see, even with a reduced Capacity.

A 6xMPL Timber with 26 DHS, would have a Dissipation of 7.8 and a Capacity of 55.6 to the First Shutdown we can Override.
At their current heat of 6 each, the mech would spike 36 Heat with an HPS between ~9.35 to ~12.00. So there could be a Heat Scale Penalty that can be applied since the damage sits at 48. And I would have no problem seeing C-MPLs return to 7 Damage each while retaining their current 6 Heat, since base Dissipation would be increased anyway.

[DivineEvil]

I woudn't be so specific with the values. It requires some preliminary testing to see clearly what the values should be. Simply starting with all DHS on their base 2,0/0,2 values and SHS on 1,4/0,14 would be a good start. Then we can see if further adjustments are needed.

It boils down to that we need to avoid addressing the specific nerfs to Clan weapons and equipment, and instead look at their core values, because core values is where their advantage lies. We do not need to address the advantages of the Clan mechs, but we need to address some values, that Clan and IS are equal on and use them to round the balance up.

Edited by DivineEvil, 14 September 2015 - 09:05 PM.

[Kdogg788]

I can get behind the system implied by your chart. Could be a more rational means of "quirking" mech chassis then arbitrarily giving mechs of the same chassis exceedingly different quirks that make no sense especially when compared to one another.

-k

[Lily from animove]

DivineEvil, on 14 September 2015 - 02:18 AM, said:

Since PTS servers went online, I've begun collecting feedback from my unit members and formulating it into a comprehensive shape. Still, very few people aside from me were able to participate in active testing when there were enough players present to effectively perform testing runs, thus the major part of the following thesis is still my own view with an addition of other opinions, mostly based on discussions over raw math and shared screenshots. Unsurprisingly, many TCAF members, just as a vast majority of MWO public, were unable to comprehend the idea behind the first testing phase. To respect PGI intentions, these opinions will be left out. General points will follow:

Sensor/Targeting quirks and changes are major positives. Without a doubt, there were a need for information-related modifiers to incentivise role warfare and increase Light mech efficiency in performing flanking and scouting. With the upcoming nerfs to the ECM, these are going to make a big difference. One thing that bites me, is that base delay to the target spotting is not obvious. It might worth to implement a "signature" parameter to the mechs, together with available Sensor Range, to illustrate the base number of seconds of visual observation required to get a lock on a given mech without considering quirks.

This is the point where I would need to ask you if you or your members really play the MWO I play. What incentives? There was a need? what need? for what? The current MWO has not need for these information warfare, because nothign in MWO supports this except lrms (dead in higher tiers) and srms (not good due to spread lock).
Even with ecm nerfs, this is NOT going to mae a difference, Good pilots shoot on sight, not on scan or targetting. And information wafare will not change this. all those negtive "information warfare" quirks will not have any influence on mechbuilding for most good players. it does not even increase the scots efficiencies if they target ot not, or if they scan or not.

[DivineEvil]

Lily from animove, on 15 September 2015 - 07:15 AM, said:

This is the point where I would need to ask you if you or your members really play the MWO I play. What incentives? There was a need? what need? for what? The current MWO has not need for these information warfare, because nothign in MWO supports this except lrms (dead in higher tiers) and srms (not good due to spread lock).
Even with ecm nerfs, this is NOT going to mae a difference, Good pilots shoot on sight, not on scan or targetting. And information wafare will not change this. all those negtive "information warfare" quirks will not have any influence on mechbuilding for most good players. it does not even increase the scots efficiencies if they target ot not, or if they scan or not.

- Light mechs suffer from uniform sensor range. Since an enemy only needs to get you in his periphery to spot, you cannot effectively spot for your team. Only ECM mechs can do that at the moment.
- Light mechs can't perform well as strikers in most occasions, for the same reasons. Even if you'd manage to flank an enemy group undetected, as soon as you've landed a single shot on the target you've chosen, you can be spotted instantly, and the entire group will put their attention on you. If you carry an ECM, you're probably screwed even before making a move.
- Heavy LRM boats would now require an assistance from lighter mechs, and these lighter mech would be able to offer a nice support simply by providing extended target retention.
- On a competetive scene, information makes a lot of difference. One UAV placed on the right spot at the right time can decide whether your team wins or loses. With sensor quirks, a choice between a harasser, brawler or scout Light mech will make a significant impact.
- On a competetive scene, having a knowledge of enemy damaged components also determines between victors and losers. Simply being unable to focus your fire on a designated target affects the outcome of a fight.

It might seem meaningless from a standpoint of a regular Random Queue player, but from my point of view, being a two-year veteran player from a unit, which participates or participated in RHOD, MRBC, ISW, MLMW, NBT and some smaller leagues, its going to have some considerable effect.

Having said that, I'm afraid, we have vastly different definitions for a "good pilot", not to insult you or anything.

Edited by DivineEvil, 15 September 2015 - 08:28 AM.