Jump to content

So Here's An Idea To Curb "boating" And Ttk Being Too Low From Rampant "optimization"


38 replies to this topic

#21 WarHippy

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 3,836 posts

Posted 15 July 2016 - 08:12 AM

I know you mean well, but this is a big no from me. Poorly thought out silliness that brings nothing of value to the table.

#22 Lily from animove

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • The Devoted
  • The Devoted
  • 13,891 posts
  • LocationOn a dropship to Terra

Posted 15 July 2016 - 08:22 AM

View Postmeteorol, on 15 July 2016 - 03:02 AM, said:

It would do nothing but kill diversity and mech building. The idea is so outstandingly terrible i have problems understanding how anyone could even come up with it, ever.

Sorry.


That is when people come up with an idea fitting for the current situation and it's problem yet not trying to figure out which new situation their change will cause and which possible problems this new situation will have.
The reason many of PGI's adjustements didn't work were similar, they fixed the situation wihtout thinking how the playerbase reacts.
They shoudl share a lot more of their thoughts with us, so we cna rip them apart, because only an idea not able to be ripped apart is a good one.

#23 Dogstar

    Member

  • PipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 1,725 posts
  • Twitch: Link
  • LocationLondon

Posted 15 July 2016 - 11:57 AM

This idea would really screw over mechs like the Vindicator where you really don't want to fill the useless ballistic slots as 3xMG+1/2 ton ammo on a PPC build is a serious waste of vital tonnage (maybe if MGs actually did damage like they're supposed to it would work).

#24 oldradagast

    Member

  • PipPipPipPipPipPipPipPipPip
  • Philanthropist
  • 4,833 posts

Posted 15 July 2016 - 12:19 PM

Um, nope.

Bad idea - mechs have lots of hardpoints that were meant to allow different builds on the same chassis, not one crazy build that is a pile of random weapons. This would basically make nearly every mech in the game into useless, Frankenstein messes, with a handful of mechs that can boat properly with the hardpoints they have the new meta.

#25 Telemachus -Salt Wife Salt Life-

    Member

  • PipPipPipPipPipPip
  • The Deadly
  • The Deadly
  • 364 posts
  • LocationTerra

Posted 15 July 2016 - 12:25 PM

View PostEx Atlas Overlord, on 14 July 2016 - 10:55 PM, said:


Pros:

*No more complaints of LRMs OP



No one says LRMs OP

#26 Ex Atlas Overlord

    Member

  • PipPipPipPipPipPipPipPip
  • Survivor
  • 1,018 posts

Posted 15 July 2016 - 01:21 PM

View PostCarl Vickers, on 15 July 2016 - 01:40 AM, said:

It would make Clan lasor barf king in the timby.


Actually it would do the exact opposite. Using this system, it would not be possible to build a laser boat *only* timber.

And anyone that tired would have to make sacrifices compared to the current system in order to do it.

Aka, boating is reduced / less effective and variety is increased.... and most importantly *choices would actually matter*.

Success.

View PostKarl Streiger, on 15 July 2016 - 01:54 AM, said:

Love the idea - no change for me - although we need to get rid of hardpoint inflation - CN9-AH; King Crab - even if i want i can't place something different but a AC 20..... and the AC 20 is everything i ever need


So PGI could go in and adjust the filler slots that mechs have that they don't need....

Remove the lower arm act from the Cent9-A and put it on the 9-AH, and everyone get's their AC20 cent (It's literally what they did to the 9-AH....they just removed the actuator that every other cent has)

Easy fix, for a "problem" so minor it's hardly worth mentioning.

View PostLily from animove, on 15 July 2016 - 02:27 AM, said:

because that is what would happen, poeple use the mech that does what they want it to do best.


I think what you're trying to say is "People would find a mech with the hardpoint layout that fits the role they want to play best"

And if that IS what you're trying to say, then that's exactly what I want to happen.

the chassis would semi-define their own roles to an extent rather than having it forced through constant ineffective quirk passes

View PostLily from animove, on 15 July 2016 - 02:27 AM, said:

. Which means they won't use mechs inferior because they have to use pointless additional wepaons, and this is why such a change will reduce diversity. as you see all chassis disappear form beign used having pointless additional hardpoints one would have to fill with useless wepaons just to fit your new ruls but not fitting the optimised playstyle. It will simply not make them carry more loadouts as you told.


1) A huge portion of the mechs in the game are *already* completely ignored b/c they're not "optimized". The issue that you're pointing out is with the MWO playerbase.

2) Yes, it actually would force them to bring more varied loadouts in the vast vast majority of cases. The few outliers could be addressed by PGI individually.

View PostGreyNovember, on 15 July 2016 - 02:55 AM, said:

You want to force MGs on ravens? And the Locusts? and the Urbies?

the HBK 4G can't mount a single AC20, because it has leftover hardpoints?

If the answer is yes, then, well. Okay. I guess those mechs can go suck it.


1) If those mechs are incapable of filling the hardpoints with anything else, then yes.

2) Not to worry, the 4H can fill your AC20 hunchback need. But I have a feeling it has more to do with your fear of not being to "optimize" as effectively. Which is one of the points, as per the title.

View Postmeteorol, on 15 July 2016 - 03:02 AM, said:

It would just lead to clan mechs cherry picking their pods while trying to minimize "trash" hardpoints. It would instantly kill some mechs and reduce the meta to a very few mechs that reign supreme thanks to less "trash" hardpoints.


1) I'm not going to pretend I have every single omipod memorized for every single much. But in every case I've checked, it is not possible to super boat the way we do now without sacrificing something...which is the point.

2) Which would serve to easily highlight the few mechs that need attention as compared to trying to balance 200+ at once. As mentioned earlier.

View Postmeteorol, on 15 July 2016 - 03:02 AM, said:

People will just fill up empty hardpoints with MGS and SL.


That is possible.

But they would then be down a few tons, and slots... and in most cases that's all it takes.

View Postmeteorol, on 15 July 2016 - 03:02 AM, said:

Just look at the KDK 3. Even if you had to fill every single hardpoint, people can just slap 4 sl, 1 ams and a ton of ammo for 3.5 tons on it.


That's two double heatsinks, or over 700 damage worth of ammo gone, or a few entire steps down in engine.

It's a good change in the right direction for the mech.

View Postmeteorol, on 15 July 2016 - 03:02 AM, said:

while it would instantly send countless other mechs to the trashpile.


It would send *a few* mechs to the trashpile, but only the ones that people currently force into builds and roles they should have never been used for.

Which I consider a good thing.

And those same builds and roles could be used on other mechs, so nothing would actually be removed it would just be shifted to a more appropriate and varied suite of mechs.

View PostMrJeffers, on 15 July 2016 - 08:05 AM, said:

If nothing else it invalidates about 60% of the stock IS builds, any that have hard point inflation.


Suddenly we care about stock builds?

View PostDogstar, on 15 July 2016 - 11:57 AM, said:

This idea would really screw over mechs like the Vindicator where you really don't want to fill the useless ballistic slots as 3xMG+1/2 ton ammo on a PPC build is a serious waste of vital tonnage (maybe if MGs actually did damage like they're supposed to it would work).


It's funny that you mention the vindi, as I was playing with one last night testing out this very theory of mine.

I went from having yet another cookie cutter pop-tart PPC sniper, to a mobile medium laser brawler with both team supporting narc, and team supporting AMS.

It was an amazing transformation.

Edited by Ex Atlas Overlord, 15 July 2016 - 01:28 PM.


#27 MrJeffers

    Member

  • PipPipPipPipPipPipPip
  • 796 posts
  • LocationThe Rock

Posted 15 July 2016 - 01:40 PM

View PostEx Atlas Overlord, on 15 July 2016 - 01:21 PM, said:



Suddenly we care about stock builds?



Given the poorly thought out idea and the amount of rework it would create for about 60% of the mechs in the game to make then valid, yes. And some would even complain that they could not even run stock builds anymore, there are still people that want to play stock mode.

But as I said in my original post, that's just a starting point. Others have made plenty of other reasons why it's a bad idea. It's tons of work to implement, and doesn't fix anything but narrowing down the number of valid mechs to the ones that have the right hard points for the new system. A smaller number of mechs than are seen now.

#28 1453 R

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Bridesmaid
  • Bridesmaid
  • 5,569 posts

Posted 15 July 2016 - 02:02 PM

View PostEx Atlas Overlord, on 15 July 2016 - 01:11 AM, said:

Edit: BTW how can you possibly interpret "require players to carry more varied loadouts" as a "decrease in diversity"?


Simple.

'Varied' does not mean 'good'.

What you're talking about is forcing 'Mechs to waste tonnage on useless FrankenMech idiocy. That will not make them better 'Mechs. That will not make them better 'team players'. A Grasshopper forced to take a one-critslot low-tube missile weapon in its head hardpoint (no two-slot NARC in the noggin, folks!) will not suddenly be better because it has a low-damage, low-impact missile popgun which will not even so much as annoy any of its enemies. The only person such weapons will annoy is the one forced to use them.

Ergo, the 'Mechs that do not have to waste tonnage on useless FrankenMech idiocy, and can thus use that tonnage on things that actually help them fight, become objectively better than 'Mechs which waste tonnage on useless FrankenMech idiocy. Those 'Mechs easily defeat 'Mechs saddled with FrankenMech idiocy, given equivalently skilled pilots.

Diversity: decreased.

Plus, your argument invalidates the AWS-9M. 2E/2M in its center torso - the AWS-9M chassis becomes impossible to launch in any valid configuration. Your proposal would make it impossible to launch that 'Mech.

MechLab changes that make it impossible to launch in certain 'Mechs are bad MechLab changes.

Edited by 1453 R, 15 July 2016 - 02:03 PM.


#29 Ex Atlas Overlord

    Member

  • PipPipPipPipPipPipPipPip
  • Survivor
  • 1,018 posts

Posted 15 July 2016 - 02:18 PM

View Post1453 R, on 15 July 2016 - 02:02 PM, said:

Plus, your argument invalidates the AWS-9M. 2E/2M in its center torso - the AWS-9M chassis becomes impossible to launch in any valid configuration. Your proposal would make it impossible to launch that 'Mech.


Edit:
Derp, just glanced at the stock build without remembering that it shows the numbers for ones that aren't filled...

Doesn't really matter, the only option for this one mech would have to be hardpoint adjustment. Aka, no system exists without an exception.

Edit2: I'm under no illusion that this idea is a magical "fix-all", however I think it is an easily implemented step in the right direction that doesn't require the constant tweaking a weapon value, followed by attempted to balance 200+ mechs with quirks, while trying to make each chassis unique, AND salvage some sort of unique role for each.....for every single weapon.

Edited by Ex Atlas Overlord, 15 July 2016 - 03:55 PM.


#30 Ex Atlas Overlord

    Member

  • PipPipPipPipPipPipPipPip
  • Survivor
  • 1,018 posts

Posted 15 July 2016 - 02:30 PM

View Post1453 R, on 15 July 2016 - 02:02 PM, said:

Ergo, the 'Mechs that do not have to waste tonnage


Like which ones?

Even if there were 30 of them (which I highly doubt, but I don't have every single mech memorized), that brings the number of mechs that need serious attention down to that number....rather than trying to balance 200+ simultaneously.

Edited by Ex Atlas Overlord, 15 July 2016 - 02:31 PM.


#31 cazidin

    Member

  • PipPipPipPipPipPipPipPipPip
  • FP Veteran - Beta 2
  • FP Veteran - Beta 2
  • 4,259 posts

Posted 15 July 2016 - 02:33 PM

This idea is mediocre™. Posted Image

#32 1453 R

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Bridesmaid
  • Bridesmaid
  • 5,569 posts

Posted 15 July 2016 - 02:50 PM

The fact that EAO honestly believes the AWS-9M and the AWS-9M[C] have different hardpoints is...really all I should need to say for this idea. If for some reason its patent absurdity doesn't shoot itself down just peachy-keen fine for anyone reading it.

Mediocre™ indeed. Let's just see how Ghost Heat 2.0 falls out before we try to do ridiculous, nonsensical things like 'emtpy hardpoint lockout' mechanics, eh?

#33 Lykaon

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,815 posts

Posted 15 July 2016 - 03:10 PM

View PostEx Atlas Overlord, on 15 July 2016 - 01:11 AM, said:


Which would balancing that much easier by highlighting the few outliers to be targeted specifically rather trying to "quirk" 200+ mechs simultaneously while also trying to make each specific chassis unique via said quirks b/c allowing players to over customize has diluted the game into.... as you put it.... "cookie cutter" loadouts.

One of the benefit in doing it this way is that the roles and variety are already build into the mech, and don't need to be forced with constant quirks.



Perhaps quirks could be completely removed from clan mechs b/c their ability to swap out hardpoints would become the massive advantage it was always supposed to be?



Suddenly we care about stock builds, and lore humping?



Firepower would undoubtedly be reduced across the board (more in some cases than others) but overall you're wrong here (especially with the GH2 coming soon).



All of these have already been mentioned, please read the thread before repeating someone else's point

Edit: BTW how can you possibly interpret "require players to carry more varied loadouts" as a "decrease in diversity"?



Sigh...

Clearly you are not a forward thinker.

If your plan does indeed cause the limiting of optimized choices this is not a good thing. Your idea creates a situation where some mechs are superior choices and most are crap.(more so than it is now)

So the highlighted outliers as you put it would become evident (like they arn't now?) then what? your idea has created the situation where under the rules set by your idea has created the situation where the few outliers are potent. It was your idea that created the situation. If your plan didn't exist then these mechs would not be in need of adjusting to fit the plan that to be blunt is poorly thought out. If your idea creates the situation that then causes imballance and then requires adjustment isn't it evident that it's a bad idea?

I fail to see how roles will be defined by forcing builds according to your proposal. The core issue isn't mech build options it's game play mechanics. The game only really rewards one role and only requires that one role. Combat is all there is to it.

Do we need recon when the maps are small and predictable? nope we don't.

Do we need mobile fire power to delay or temperarily hold territory on tiny maps that allow even the slowest mechs to reach the point of engagment in 3 minutes or less? What role does a fast skirmisher medium have really? why not just use another heavy mech?

Role warfare is supported with our current mechlab and options open to use. Role warfare is NOT supported by the map designs and game modes and certainly not adiquatley supported by the rewards system in place.

Quirking or unquirking who really cares if the best choice would be a handfull of clan mechs we will all be using the same half dozen chassis. Would it matter if it was a game of sameness?

Light mech = Arctic Cheetah
Medium Mech =Stormcrow
Heavy Mech = Timberwolf
Assault mech = Direwolf

We will not need any other mechs in solo quick play and because of this we can just nail the coffin lid down of faction warfare.


"suddenly we care about stock builds and lore humping" Wow you missed the boat on this point.

Do you expect PGI to redesign EVERY mech they put out so far and to come to be compliant with your idea?

Any mech with hardpoint inflation (more hardpoints than in use on the stock loadout) would need to either have it's hardpoints reduced to stock compliance or the stock loadout needs to be altered to meet the new criteria you suggest.

Otherwise most of the mechs you buy in the game store are invalid and illegal builds.So now you need to save c-bills to aford a mech that you can't even pilot until you spend more c-bills to make it compliant with your terrible plan.

Also, There are players who do enjoy stock mech matches and PGI will be implimenting support to do so in private matches.


Firepower will not be effected one bit by your plan because the mech choices will be reduced to those that are uneffected.

Like Black Knights.The previously mentioned Omnimechs Crabs Locusts A1 catapults and at least a half dozen others that already have all hardpoints in use in their popular comp builds.


Your plan solves nothing but causes new problems to develope.So it's a bad idea.

#34 TheArisen

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 6,040 posts
  • LocationCalifornia

Posted 15 July 2016 - 03:18 PM

View PostLykaon, on 15 July 2016 - 12:41 AM, said:



The natural progression would be a mass migration to the few chassis that have just the right hardpoints to meet the criteria of building for a massive alpha strike.

So this would make most of the chassis obsolete due to being painfully inadiquate for the optimized meta.

Also, Omnimechs would be vastly superior to battlemechs because you can use optimized pods to prevent excess hardpoints from becoming an issue. This idea heavily favors Omnimechs.

Native stock builds would need to be adjusted by reducing inflated hardpoints or else become invalid under this rule. For example a Hunchback 4G would need to lose 2 ballistic hardpoints because it's native stock build only uses one ballistic and three energy hardpoints.

Also this idea would not significantly impact TTK. The only change would be we would all be piloting similar mechs.The game would likely become very cookie cutter.


P.S. the only people who think LRMs are over powered are morons.


This is my response to this idea as well.

#35 Lykaon

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,815 posts

Posted 15 July 2016 - 03:50 PM

How about an idea to actually impact TTK.

I am not even supporting this idea but here is a means of doing.

Currently our damage dealing mechanics have armor as the first line of defense and structure os the las bits of health to keep that segment of a mech functional. Armor is generally the higher value and is depleted on a 1 armor point for 1 damage point ratio.

If you can alpha strike 30 damage into one mech section it will deplete 30 armor. It's this simple.

What we have now is an ablative armor value that protects the functioning bits and peices of a battlemech. Once the armor is depleted damage is applied to the mech's structure and components.


Now onto my idea.

Improve the critical hit mechanics to function more like the source material (table top) What we have now is a basic outline of this mechanic but I propose a more robust and complete critical hit mechanic.

When a mech takes damage onto a location that has depleted armor the internal components will take the damage.

My sugestion is to have ALL components as valid to take damage and that all components have negative effects when destroyed.

The following effects are provided as examples.

Head components

Lifesupport: first hit will cause the pilot's view to become blurred slightly when heat threshold exceeds XX% of heat capacity.This represents the pilot strugging under the high heat to remain concious.As the heat is displaced the vision clears to normal.

Life support second hit: Pilot falls unconcious when heat threshold exceeds XX% of heat capacity.When unconcious the mech can not act (just like a shut down) As the heat level is displaced the pilot returns to conciousness with blurred vision eventually returning to normal after sufficent time has passed and heat levels remain normal.

Cockpit hit: mech destroyed.

Sensor hit one: HUD elements flicker and scramble at random intervals.

Sensor hit two: HUD elements and minimap are OFF!

Torso Criticals:

Engine hits each hit on an engine builds XX% additional heat and reduces speed/accel/decel and turning by XX% all hit are cumulative and accumulating thre engine hit destroys the mech.

Gyro hits: First hit cause the targeting reticule to shake when the mech is moving the rate of the shake increases with the higher the speed value the mech is moving at set at a fixed % of the mech's top speed. (so all mechs are effected the same )

Second Gyro hit destroys the mech.

Arm Actuator damage: Each actuator destroyed reduces the range of motion available to move the arm aiming reticule.

Leg Actuator damage: Each leg actuator destroyed reduces the mech's speed and turning speed by XX% All leg actuators destroyed destroys the mech.

Weapons and equipment cease to function when they take a critical hit.


Now here is how we exstend TTK using this idea.

We increase the structure health of mechs mainly focused on torsos and increase componet health as well.

The proceedure is as follows.

Damage is taken to a location deplete that location's armor value by the damage received.

If damage exceeds armor generate a target located in the location hit to be the recipient of the damage.Apply damage to that target. Viable targets are any item on the critical table for that body segment.

Weight the target selection "roll" to favor structure hitpoints over component hit points something like 1-50% structure 51-100 random component housed in the targeted location.

So what has changed?

Our current combat model is very basic.

Apply X damage to Y locations health value if X exceeds Y Y is destroyed. This leads us to build mechs that are designed to apply high volume of damage to singular target locations in order to overwhelm the health of the armor and structure in that location to destroy the target. The objective is dish out loads of damage to one spot as fast as possible to kill a target.

My proposal shifts the damage dealing away from destroying a mech's body outright (because it will take to long) to focusing on breaching armor to disable the mech's ability to function.

Currently we see a Jagermech with twin Gauss Rifles we would aim for a side torso because it has fairly low health and is highly likely to contain an XL engine. So let's say 50 armor and 30 structure and we have a dead Jager. With 40 point alpha strikes a norm that's 2 or 3 alphas or about 8-12 seconds if everything goes well.

The new way of dealing with this mech would be to either shoot to kill the engine (requiring more shots because the structure and engine health are higher) or try to remove the ability to fire the gauss rile(s) by targeting the weaker arms.

The focus is shifted from killing whole mechs to destroying components on mechs to remove the capacity to fight.



Now for the residual impact on game play.

Mechs will live a lot longer and soak more damage.

Ammo values may need to be altered

Game modes focused exclusivley on destoying mechs would take longer

Further encourages "deathballs" and focus fire.

#36 Ex Atlas Overlord

    Member

  • PipPipPipPipPipPipPipPip
  • Survivor
  • 1,018 posts

Posted 15 July 2016 - 04:08 PM

View Post1453 R, on 15 July 2016 - 02:50 PM, said:

The fact that EAO honestly believes the AWS-9M and the AWS-9M[C] have different hardpoints is...


Yeah, derped and forgot it only lists hardpoint numbers for empty hardpoints

View Post1453 R, on 15 July 2016 - 02:50 PM, said:

Let's just see how Ghost Heat 2.0 falls out


We all know exactly what's going to happen...

Lights and most mediums for the most part will either be not affected at all....or affected to such a slight degree that the game will not change at all in that weight ranges

Heavies and assaults will have their return fire nerfed across the board, with the only thing being left that anyone will use will be builds that can fire two 25-30 alphas on two different buttons quickly.

Aka, splat cat will still exist, it will simply exist in two bursts separated by 1 second, rather than one burst all at once...

Aka -> won't fix much, a few assaults will be completely gimped.

Which means we need something else to help with the issue.

View PostLykaon, on 15 July 2016 - 03:10 PM, said:

So the highlighted outliers as you put it would become evident (like they arn't now?) then what?


Already addressed, please actually read through the thread.

View PostLykaon, on 15 July 2016 - 03:10 PM, said:

If your idea creates the situation that then causes imballance and then requires adjustment isn't it evident that it's a bad idea?


If you believe there is a magical fix all that will in one fell swoop fix every possible thing wrong with the game, then there is nothing anyone could ever suggest that you would be satisfied with.

Progress happens in stages

View PostLykaon, on 15 July 2016 - 03:10 PM, said:

Do you expect PGI to redesign EVERY mech they put out so far and to come to be compliant with your idea?


Of course not, but no one suggested that so I'm not sure why you're bringing it up other than to be vaguely argumentative without making a clear substantiated point just for the sake of it.

View PostLykaon, on 15 July 2016 - 03:10 PM, said:

Also, There are players who do enjoy stock mech matches and PGI will be implimenting support to do so in private matches.


There are players that enjoy a full sim experience, where is support and lobbying for that?
There are players that want OP clans in a 10v12 experience, where is support and lobbying for that?
There are players that want only IS vs IS forever, where is the support and lobbying for that?

The fact of MWO is that PGI doesn't care about stock, or lore for the vast majority of the cases... so holding off from improving the game based on stock or lore is nothing but a slight minority keeping the game from improving at the expense of the majority (including PGI) that clearly isn't that concerned with stock or lore.

Edited by Ex Atlas Overlord, 15 July 2016 - 04:10 PM.


#37 Bud Crue

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Rage
  • Rage
  • 9,942 posts
  • LocationOn the farm in central Minnesota

Posted 15 July 2016 - 04:27 PM

Another way this reduces diversity is one that, admittedly probably doesn't affect many people, but would sure as hell affect me:
I have a certain affection for Quickdraws (and a few other mechs), to the point where I have several versions of the same variant. I build each differently. Several include builds that do precisely what you plan calls for (use all the hard points) several do not. Your plan eliminates all of those versions that do not use all the hard points...probably 15% of my mechs would be instantly invalidated. So for me, and my particular affliction, this plan is most certainly a reduction in diversity.

That said, I think your heart is in the right place as far as trying to create a more immersive BT game. Alas, with PGI's clear indication that immersion is not something they care about, nor BT, I can't imagine that you would get much support for the idea. Keep thinking. I for one appreciate the effort.

#38 Pjwned

    Member

  • PipPipPipPipPipPipPipPipPip
  • Little Helper
  • 4,731 posts
  • LocationDancing on the grave of Energy Draw LOL

Posted 15 July 2016 - 08:04 PM

This is the only point I'm going to address here.

View PostEx Atlas Overlord, on 15 July 2016 - 01:21 PM, said:

It would send *a few* mechs to the trashpile, but only the ones that people currently force into builds and roles they should have never been used for.

Which I consider a good thing.


You are just flat out wrong here, you obviously have no idea how many mech variants would instantly go in the trash pile with this ridiculous idea, meanwhile it would do nothing to mechs that already have optimized hardpoints.

Frankenmech garbage doesn't belong in MWO.

#39 LordNothing

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 17,120 posts

Posted 16 July 2016 - 01:19 AM

with the way pgi works this will be the one 'community' 'idea' they actually pick up on.

Edited by LordNothing, 16 July 2016 - 01:19 AM.






1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users