362 replies to this topic

[Sable Dove]

Primarily due to mech efficiencies, mechs are simply too agile. This makes it largely impossible for mediums to do their job when heavies and assaults can turn lightning fast, making it impossible for anything that isn't a light to get out of their range of motion.

Assaults are as agile as heavies ought to be, and heavies as agile as mediums. They just have lower top speed.

Mediums, though, aren't as agile as lights, and they're certainly much slower. Mediums are the point where mech efficiencies start having diminishing returns. For an assault or heavy, mech efficiencies are a great boon, allowing you to be more effective in combat, and while this is still true for mediums, the amount of benefit gained is a lot less.

Heavies and assaults are constantly making use of their efficiencies to protect themselves or to hit targets that they wouldn't have been able to without. But for mediums, the base agility is higher, and thus the increased agility of efficiencies is less beneficial. They still take advantage of it, but more often than not, the mech's base agility would have been sufficient.

Once you hit the Cicada, half of the efficiencies are more penalties than benefits, as the mech now twists/pitches faster than the camera, making it more difficult to aim.


Therefore, I propose that mech agility be cut back. Preferably by reworking mech efficiencies so that they are tradeoffs, rather than purely beneficial, but if not, at least reduce base agility and/or the benefit from efficiencies.

In this way, assaults will no longer simply be the best mechs in every situation. They will need support to function best, as they should. If an assault gets isolated by a light or fast medium, it should be an easy kill, because the assault pilot should know that they need support.

[FupDup]

Personally, I find heavies to be overall more versatile than assaults (but assaults are still pretty good).

But yeah, the fatties do seem abnormally maneuverable at times. You have no idea how absurd assault mech agility is until you've tried an XL395-400 Battlemaster with doubled basic proficiencies. Outside of the crappy hill climbing angle, it basically plays like a medium that's on steroids and wearing power armor.

Edited by FupDup, 07 December 2013 - 08:37 PM.

[Nothing Whatsoever]

Agreed.

I'd like to even look into a modular system for the Mech and Pilot Tree.

[Adiuvo]

Agreed. This is a root problem of the game. It's not really weapon systems. Assault and heavy mechs are supposed to hit hard - they should be frightening in a direct engagement or in entrenched positions. But, right now, they can easily respond to any light or medium ambush and combined with their high damage output, make quick work of them.

Siriothrax suggested before that assault efficiencies should be half as effective as they are now. Ie., a basiced assault should be the new 2x version. The 1x version should therefore be half of what it is now.

If this is implemented I'm not sure if lights and mediums will be quick enough comparatively, but further adjustments could be made. Either way it would be a step in the right direction.

[Nothing Whatsoever]

The specific efficiencies need to be looked at and redone.

This thread has a fine analysis related to reducing agility: http://mwomercs.com/...__fromsearch__1

[aniviron]

I'm pretty sure OP knows about this thread, but to anyone who hasn't looked at it, check out http://mwomercs.com/...-class-balance/ . It's an incredible thread with a lot of numbers; the long story short is that an Atlas running between a 300-250 engine can track almost any mech in the game even at under 100m, including max-engine lights. Check out pages 4-5 for more info on turning. The first four pages are mostly an investigation of why the engine speed-weight ratio is not a liner increase, and also makes for excellent reading.

EDIT: Looks like Praetor Shepard beat me to it in the post above. It should tell you something that people are recommending this thread quickly though, you should really go read it if you care about how this game is balanced.

Edited by aniviron, 07 December 2013 - 10:07 PM.

[Hoaggie]

I don’t know what’s more pathetic, that somebody actually thinks this, or that there are players out there that like this. If an assault mech puts on a engine large enough to keep up with mediums it’s going to have about the same armament as a medium. There are major drawback do doing that.

If I see an atlas flying along at 50kph with full weapons, I realize that it probably has an XL engine, or it has massively shaved its armor on the legs. If you have a fast assault, then it’s not going to have a high damage output. Additionally, because of the heat balance, there is no reason for an assault to carry many weapons if they are just going to overheat at an accelerated rate.

What you are seeing as a speed issue, is the lack of collisions. Light mechs are filling the rolls mediums should be because they just set their throttle to max and run, there is no drawback for them running directly into an enemy.

[Aim64C]

I'm a bit cautious when it comes to this idea.

As I've said in other threads - the key balance issue is that we are playing Team Solaris in a tiny box canyon only 5 kilometers across. Only in some of our largest maps do players actually encounter the scenario where being light and mobile is better than being slow and powerful.

Trying to balance that is an exercise in futility - a fool's errand. You're searching for the I-D-10-T form. Heavier mechs are more efficient for carrying firepower and ammo than lighter mechs by time you work out all the math. A Jenner and a Locust cannot match the weapon tonnage of a ShadowHawk - at least, not without making substantial compromises to speed that are fatal in light mechs that have not the armor.

Fighting is more than statistics on paper - we know this.

Which is why it's a fool's errand to try and balance mechs around the singular goal of killing other mechs in Team Solaris when there's a tonnage disparity of up to 80 tons (or 1:4).

Even balancing teams by tonnage is only acceptable as a stop-gap to current ineptness of the match making system for partially the same reason as I mentioned before.

Light mechs were not really built to be anti-mech platforms that can take on Assault mechs playing on 'even teams.' They are designed for scouting, Quick-Reaction Force (QRF), infantry/armor support, and interdiction. They get there fast, pack enough weapons to make short work of opposing infantry, tanks, etc - and can radio home to report that there are a bunch of bad guys for the Mediums/heavies to intercept and the assaults to lay a trap for.

In the past, hit detection and host-server disparity made hitting light mechs a gamble on a good day. Piloting a light mech was kind of surreal, and made you think everyone was a ****-poor shot (until 6 PPCs finally connected and left two smouldering leg-nubs in their wake). Though we had the streak-ravens to counter Jugger-Jenners with auto-core enabled.

It all boils down to 'role warfare' being reduced to a few square kilometers centering on the objective of breaking the enemy's toys faster and more efficiently. That's not enough room to give Light Mechs much of an advantage (particularly in the balancing of combat versus held territory) - so they are ultimately forced to try and contribute to the overall breaking of enemies - a task they are ill-suited for without completely changing the core concepts of the game.

Many of the weapon balance issues, also, arise because of these faults. An Awesome can operate for extended periods of time along a front (provided it doesn't get chewed on too much - and given its arsenal, it has plenty of ability to keep its distance from enemy teeth) without much concern for supplies. An AC40 jagger, however, cannot. A mech sporting two gauss rifles would be a poor choice for taking on a battalion of hovercraft as compared to a mech sporting two large lasers or an LRM5.

There are other issues with weapon balance that revolve mostly around implementation (the values could use some adjustment, but the key issue is mostly implementation - the best thing MWO did was re-think how lasers work... they just never really seemed to grasp that a similar re-think needed to be applied to almost all weapons in the game - particularly the most troublesome balance quirks through MW history) - but even those iron out considerably when you start to put mechs into a more realistic battlefield with much more ground to cover and varied objectives.

In many ways - doing that removes the need for all weapons to try and be 'balanced' against each other - and they can, instead, be balanced for the roles they fill (with the implementation addressing the potential for abuse to a degree).

I could ramble on about this for hours - I've re-thought most of the weapons (from pulse lasers to PPCs, SRMs to Streak LRMs) and how to implement them. Some of them have been composited with ideas I've seen from others, while some are more original (or, at least, I like to think I thought of them on my own - someone else has probably come to similar conclusions at some point or another). I've got a different heat system that allows for heat related penalties (by separating "chassis" and "Heat System" temperatures and using a differential equation to vent heat from the heat system and allow bleed-over on the chassis based on temperature differences.... the math is much more complicated than the concept - which is that you have two heat bars - one functioning identically to the current heat bar, except it also 'bleeds' into a second heat bar that is used to derive heat penalties from; short spikes in heat bleed through the insulator more quickly, but heat spikes don't instantly translate into shut-down or penalties. Chain-firing results in heat build-up, but migration across the insulator is slower as the temperature delta is considerably smaller - meaning you can run at no or a mild penalty for a time). I've also thought about how to implement a sighting system that gets rid of ideal convergence but does not require huge amounts of micro-management to use (or introduce more randomness than absolutely necessary) - I'm not completely there, on that one, though.

I've thought up the overall implementation of the game-mode I described, earlier, involving a persistent battlefield, varying objectives, community warfare, etc. I've even done some re-thinking of the skill trees (such as actually making them trees as opposed to linear upgrades), how faction reward points would work, and come up with a way to differentiate Clan and IS tech - with IS weapons and components being able to accept modules that alter performance (representing the variety of IS manufacturers - though balancing this so as to encourage new players rather than discourage them would be crucial).

I'd be more than happy to take up ARMA 3 scripting to show what can be done with this franchise regarding MMO-style games. The main problem is that I don't have the experience, time, or tools to create the art assets necessary - and there are some nuances to ARMA 3 that I will have to learn. But the engine comes out of the box with almost all of the features we need the most (accurate hit detection, ballistics simulation, network conflict resolution, partial destruction/damage of vehicles, absolutely massive terrain possibilities, dynamic and powerful scripting, AI that can be scripted to create a server environment). About the only thing it doesn't have is 64-bit executables (unfortunate, considering the first ARMA did) or have its scripts running in Open CL (which... I'd have to check, you might be able to call OpenCL in the scripting... kind of like how some scripts can be created to call additional OpenGL or DirectX features that were not part of the original game code/design).

[DaZur]

While I can appreciate your premise...

Bare in mind those Assaults and Heavies had to sacrifice something in order to net that speed beit and XL engine, armor and or firepower.

Personally, I run a -3D as a skirmisher... 340XL but only a 30 point damage composite (3x mlas & 2 UAC/5)

Yes, it does haul butt... but it is often outgunned by Mediums and is obviously susceptible to engine crits...

I'd be more inclined to agree with you if there was no down side to the increased speed... but that's not typically the case.

Edited by DaZur, 07 December 2013 - 11:03 PM.

[Carrioncrows]

Agility just need to be un-capped from Engine rating.

All mechs should have an Agility Value that can only be increased by Pilot skills. The faster a mech goes passed stock speed the lower their agility goes while exceeding that value, they have to slow down again to gain that agility back.

That way you can start to define mechs within a weight class.

The commando while smaller might have Excellent Agility but as with the TRO isn't that fast.

So the Jenner might have more weapons and be a lot faster the Commando would be better in city situations where it can turn , stop and go on a dime where the jenner wouldn't be more adept at hit and run or drive by's.

[Nothing Whatsoever]

Hoaggie, on 07 December 2013 - 10:39 PM, said:

I don’t know what’s more pathetic, that somebody actually thinks this, or that there are players out there that like this. If an assault mech puts on a engine large enough to keep up with mediums it’s going to have about the same armament as a medium. There are major drawback do doing that.

If I see an atlas flying along at 50kph with full weapons, I realize that it probably has an XL engine, or it has massively shaved its armor on the legs. If you have a fast assault, then it’s not going to have a high damage output. Additionally, because of the heat balance, there is no reason for an assault to carry many weapons if they are just going to overheat at an accelerated rate.

What you are seeing as a speed issue, is the lack of collisions. Light mechs are filling the rolls mediums should be because they just set their throttle to max and run, there is no drawback for them running directly into an enemy.

That is oversimplified, IMHO. Agility is more than straight line speed.

Are you aware of how Engine size increases torso and arm yaw and pitch speeds and turn speed for example?

If you click on the hyperlinks for mech variants such as "Engine" and "Torso/Arm" where you can see for yourself some of the values taken from the game here: http://mwo.smurfy-net.de/

Those aspects of agility then gain a further bonus from the current Mech Tree Efficiencies. So that kind of info can be calculated as to how a Medium mech relates to an Assault.

[Green Mamba]

Aren't Mechs supposed to be more maneuverable than Tanks though,thats the idea

[Diego Angelus]

Aim64C, on 07 December 2013 - 10:52 PM, said:

Epic wall of text

somebody has to start slow clap

[Nothing Whatsoever]

Green Mamba, on 07 December 2013 - 11:07 PM, said:

Aren't Mechs supposed to be more maneuverable than Tanks though,thats the idea

Well, modern tanks need a great suspension system or they can get caught on rougher terrain.

There are several real projects such as the Boston Dynamics' Atlas and Big Dog that look to handle more extreme terrain as they keep working on their prototypes.





And what we have in the BT universe should be more rugged than those prototypes.

Edited by Praetor Shepard, 07 December 2013 - 11:18 PM.

[aniviron]

Not really sure where all these "Assaults and heavies need to make huge sacrifices to run bigger engines" arguments are coming from. Here's a pretty standard AS7-D: you will notice that its 350 engine is standard, and more than enough to help it keep up with heavy mechs, along with giving it the ability to track even the fastest targets in the game at 30-50m. Additionally, to do this it didn't have to sacrifice speed, armor, weapons, or cooling efficiency. Here's the standard AS7-D-DC brawler build, which uses the same engine, and also carries terrifying firepower (or would, if they fixed SRMs) and armor that is within 10 points of being maxed.
On lighter assaults like the Victor, the side torsi are fairly well-protected, and it's not uncommon for them to run engines in the 350-380 range. Again, with an engine like that, they can still run near maxed armor, an AC20, and a pair of SRM6s and medium lasers, plus jumpets, which only increase mobility. With an engine that size, there is no mech in the game that can run fast enough to outturn it.

It's great that you feel invincible when you bring your assault out to play, but I'm not really sure how it's fair that assaults get to kill everything and lights have to run when combat starts.

Aim64C, on 07 December 2013 - 10:52 PM, said:

[snip- lots of stuff about CW and role warfare]

While those are good points, and absolutely things to keep in mind while the game is being balanced, I want you to take a step back for a second. We're months away from the first module of CW, which doesn't even involve combat at all. We're almost certainly years away from having meaningful role warfare, if it ever happens. That's more than long enough of a time that this problem needs to be addressed now, and if need be, can be adjusted back down later.

I'm not as optimistic as you that the first coming of CW will herald a new golden age that will save us all and restore balance to the weight classes. Please keep in mind that one of the three (mandatory) modes that will determine if a planet is conquered is deathmatch, and another is Conquest, as it exists now. I very very much doubt that having a galaxy map will change the fact that deathmatch and Conquest (without some big changes) are not gametypes where being a light mech is terribly relevant. Owing to this fact, I am also not holding out much hope that attack and defend will wind up being the place where lights shine, seeing as it is being designed by the same people.

[Smitti]

I really strongly want to disagree with this, but there is a very valid point here. Big bipedal robots should not be able to traverse even half as quickly as they do. Hell, larger excavators/re-claimers/cranes etc need slew brakes just to slow down & stop while traversing, and the only reason they can do so is a) the majority of their weight is in the tracks and not the turret and they have such a large footprint and heavy load that provides them with adequate friction so that they don't spin around on the spot while doing so. A bipedal robot with a much smaller footprint, non rigid structure and the majority of their weight in the rotating torso should simply spin, lose balance and collapse thanks to the inertial forces of doing so.

I will most definitely shed a few quiet tears if/when my XL-400 ballerina battlemaster has it's agility nerfed into the ground, but it looks like it has to happen.

Edited by Smittiferous, 08 December 2013 - 01:50 AM.

[Mudhutwarrior]

Nerf Warriors

[OneEyed Jack]

aniviron, on 07 December 2013 - 11:27 PM, said:

Here's a pretty standard AS7-D: Here's the standard AS7-D-DC brawler build,

If you think those are standard builds, or that they aren't giving anything up, you know little about Atlases (and probably Assault mechs and mech-building in general).

Ore maybe you're referring to the "standard" build of those that don't know better..... hmmmmm....

Smittiferous, on 08 December 2013 - 01:39 AM, said:

blah, blah, physics, prattle, blah....

Seriously? It's called "suspension of disbelief," and if you can't manage that, you're playing the wrong game.

In the BT universe, mechs are the dominant war machines primarily because of mobility.

BTW

[aniviron]

OneEyed Jack, on 08 December 2013 - 05:24 AM, said:

If you think those are standard builds, or that they aren't giving anything up, you know little about Atlases (and probably Assault mechs and mech-building in general).

Ore maybe you're referring to the "standard" build of those that don't know better..... hmmmmm....

In the BT universe, mechs are the dominant war machines primarily because of mobility.

Really? I know the ct llas on the AS7-D is slightly aberrant, but the D-DC is pretty much exactly the same brawler D-DC that people have been running since the day ECM was patched in. Doesn't see much use these days because SRMs are awful, but then, you don't see that many atlases at all because SRMs are awful, and any brawler atlas relies on them to some degree.

[FupDup]

aniviron, on 07 December 2013 - 11:27 PM, said:

Not really sure where all these "Assaults and heavies need to make huge sacrifices to run bigger engines" arguments are coming from. Here's a pretty standard AS7-D: you will notice that its 350 engine is standard, and more than enough to help it keep up with heavy mechs, along with giving it the ability to track even the fastest targets in the game at 30-50m. Additionally, to do this it didn't have to sacrifice speed, armor, weapons, or cooling efficiency. Here's the standard AS7-D-DC brawler build, which uses the same engine, and also carries terrifying firepower (or would, if they fixed SRMs) and armor that is within 10 points of being maxed.
On lighter assaults like the Victor, the side torsi are fairly well-protected, and it's not uncommon for them to run engines in the 350-380 range. Again, with an engine like that, they can still run near maxed armor, an AC20, and a pair of SRM6s and medium lasers, plus jumpets, which only increase mobility. With an engine that size, there is no mech in the game that can run fast enough to outturn it.

It's great that you feel invincible when you bring your assault out to play, but I'm not really sure how it's fair that assaults get to kill everything and lights have to run when combat starts.




While those are good points, and absolutely things to keep in mind while the game is being balanced, I want you to take a step back for a second. We're months away from the first module of CW, which doesn't even involve combat at all. We're almost certainly years away from having meaningful role warfare, if it ever happens. That's more than long enough of a time that this problem needs to be addressed now, and if need be, can be adjusted back down later.

I'm not as optimistic as you that the first coming of CW will herald a new golden age that will save us all and restore balance to the weight classes. Please keep in mind that one of the three (mandatory) modes that will determine if a planet is conquered is deathmatch, and another is Conquest, as it exists now. I very very much doubt that having a galaxy map will change the fact that deathmatch and Conquest (without some big changes) are not gametypes where being a light mech is terribly relevant. Owing to this fact, I am also not holding out much hope that attack and defend will wind up being the place where lights shine, seeing as it is being designed by the same people.

To build off of this, here is my "Crackmaster" Battlemaster that I referenced earlier:
Crackmaster

Not the most heavily armed assault by any means, but it still packs a good punch and can avoid sticky situations better than slower mechs. Even with slower Battlemasters like XL365, they can still get around pretty quickly.

Edited by FupDup, 08 December 2013 - 08:36 AM.