I had been meaning to write this in response to the bad parts of the quirkening. While quirks do help certain chassis and variants more than others, there are chassis that are inherently terrible despite that like the Spider-5V, where quirks honestly don't solve the problem that a particular hardpoint adjustment or "inflation" would have addressed the issue better. I don't expect to convince anyone that there is a right or wrong way to go about it, but only to inform people what goes into said decisions AND remind PGI that there's a lot more to consider when the current design is noticeably flawed in some situations...
Anyways, this journey started when it was mentioned that quirks were coming to the IS mechs in an attempt to make them useful vs the Clans. Personally, I think that has more to do with design decisions such as mech scaling, weapon balance... and a host of other things that have allowed the Clans to be prominent. Fortunately, the IS still has options... but in the case of certain mechs, such as the Cicada-3C, it didn't really address actual issues that are more fundamentally more problematic. While I don't believe the other Cicadas are a problem (they are oversized, but it is what it is), the Cicada-3C is one of the poorly thought out hardpoint inflation deals.
So, how do we address this? We have to look at the history and usage... and look at the application before we honestly try to "fix" something that isn't entirely broken.
I can't even try to hide "3" in the discussion, because it always breaks down nicely. 3 sections will cover the overall point...
1) History - Before Mr. Potatohead...
2) Application - War of the IS 55 tonners
3) Solutions - I'm a ForumWarrior... not a Game Developer!
1) History - "Those who cannot learn from history are doomed to repeat it."
Hardpoint inflation is pretty much a necessary evil, born out of mech designs that require a bit of help.
For instance, let's look at the venerable Jenner.
Jenner-F
2E Left Arm
2E Right Arm
Jenner-D
2E Left Arm
2E Right Arm
1M Center Torso
Jenner-K
2E Left Arm
2E Right Arm
1M Center Torso
This is based on the STOCK TRO builds of the mech. They honestly aren't very different... so the only way to make them "different" or differentiate themselves is to inflate their hardpoints and this is what PGI had come up with...
Jenner-F
3E (+1) Left Arm
3E (+1) Right Arm
Jenner-D
2E Left Arm
2E Right Arm
2M (+1) Center Torso
Jenner-K
2E Left Arm
2E Right Arm
1M Center Torso
As you can see, the inflated hardpoints appears to make the Jenner a lot more diverse... although the Jenner-K didn't get any buffs. Remember, this is all down PRIOR to quirks.
While I have lots of bad things to say to PGI about the Jenner-K being screwed time after time (because I love Jenners in MWO), you would at least assume that there should be a benefit to having fewer hardpoints but is unfortunately completely non-existent despite the post-quirk era.
Anyways, from this base idea you should be able to see what could come from this as some chassis and their variants need it... while others are naturally more distinct such as the Hunchback or the Cataphract that didn't need any of this. So, it's up to the stock TRO builds to help out the case...
There are actual rules to this... at least from what I gather based on how PGI has designed mechs in this game.
1) Inflated hardpoints can only be added to sections that already have weapons.
Simply put, we can't add hardpoints that didn't exist previously. If the left arm had energy, we can only add more energy hardpoints to that left arm and nowhere else.
Even if it was added on a different variant, it cannot be added to another variant. For instance, the Cataphract-4X has a head missile hardpoint and 2 CT energy hardpoint. This CANNOT be added to the other Cataphracts if we wanted to inflate their hardpoints. So don't bother asking for it.
I used to ask for something like that back in the day for the Spider-5V (extra energy hardpoint) and Catapult (head energy) but doing so could set a dangerous precedent. While I'm not happy with that (particularly for the Spider-5V), the "legomech" (dynamically visual hardpoint) system that PGI employs "technically" allows for this. It's something to think about.
2) Inflated hardpoints can be increased by 1 and as much as 3 in one section.
The likelihood of a mech getting hardpoint inflation increases based on the number of hardpoints it naturally has (the fewer it has, the greater the chance) and the tonnage bracket it happens to be in (light mechs get fewer hardpoints on average than assault mechs). Of course none of this applies to Clans (it may change someday though) as their omnipod system alleviates most of that. It's pretty straightforward...
Note: Right now, smurfy's is down so I can't easy-reference stuff. I'm aware that the Griffin-3M has a 3 missile hardpoint increase... the Banshee-3E has a 3B increase (it has one ballistic naturally) and King Crab-0000 has a 3E increase (it has one energy hardpoint naturally).
Obviously, you can only add hardpoints to sections that you can inflate.... you can't inflate the head hardpoint at all (I mean, unless there are special rules and stuff that changes the situation) and the CT is only limited to an increase of 1...
3) The number of hardpoints total must be "equal" across all variants of a chassis, with some exceptions...
There are really some lame exceptions to this rule... like considering actually having JJs as a "hardpoint" or having the max capacity of JJs increase by 2 and calling that a substitute. That's not even the equivalent. A mech like the Victor-9K is affected by this where a second ballistic slot would have made it more useful than 2 more JJs (in order to sell the Dragon Slayer... but I digress). The Shadowhawk-5M is in a similar situation... but inevitably that is what PGI went with.
The Awesome-9M is a peculiar case, where the "mass" of energy and missile hardpoints in the CT still "equal" the same thing when you really think about it as they are "shared".
However... this particular rule brings to the forefront of a more pressing balance related issue. Not all hardpoints are "the same" when a mech is evaluated.
Some of it is a personal belief, but I think labrats will agree that there's a particular distinction as this is the primary reason I wrote this thread (it won't be as useful in my grand example)... but it's really simple.
Here are my 3 rules...
1) For the IS, when you have a head energy hardpoint, that is 1/2 a "large energy" hardpoint. 2 CT energy hardpoints equals 1 "large energy" hardpoint. On any other section of the mech, each energy hardpoint is 1 "large energy" hardpoint.
For the Clans, any energy hardpoint is a "large energy" hardpoint.
The reason for this is very simple. It is based on the Large Laser or the Clan ER Large Laser.
It seems arbitrary, but it's significant in how you build a mech. It is woefully unfair to compare the Large Laser to the CERLL, but the one crit difference allows the Clan version to be fit virtually anywhere... whereas the IS LL is limited to the CT size at minimum.
2) For the IS or Clans, when you have a ballistic hardpoint on a mech (that is generally 40 tons or less), the use of the Machine Gun means that ballistic hardpoint is 1/2 a "large ballistic" hardpoint.
While the MG in MWO is nothing like the MGs in previous MW games (as in reasonably useful), they are the last option for most light mechs that have to boat them in great number. What this implies is that they are genuinely "half as effective" as a legitimate ballistic like the AC2 (which is just bad, but whatever). This is what fundamentally for me makes the Spider-5K and even moreso the Cicada-3C are "underpowered" by comparison. It "translates" the 1E, 4B mech into "1E, 2B" which is just "3 hardpoints"... far below the standard of 5 hardpoints on the Cicada-3M or Cicada-2B.
This also makes the Spider-5K effectively be a "3 hardpoint mech"... and it's kinda depressing to be in such a niche. It is what it is though.
3) For the IS, a NARC (missile) hardpoint requires a minimum of the CT size. However, it can't run ASRM6s.
For the Clans, a NARC hardpoint can be in the head (if/when such a mech exists, it will exist to troll) and have no issues running ACSRM6s in the CT.
That was just a minor nitpick, but it is worth pointing out.
Now having given the history and background... let's look at the mechs that I'd like to discuss... the 55-tonners and how hardpoint inflation has affected them.
2) Application - What would happen had we gotten the Hoplite or Dervish over the Kintaro...?
So, let's look at the 55-tonners in question, and see the hardpoints added (at a glance). Note that this DOES NOT include hero mechs since those are not part of the discussion...
Commentary follows after each chassis...
Griffin
GRF-1N
3E (+2) Right Arm
3M (+2) Right Torso
GRF-1S
4E (+1) Right Arm
2M (+1) Left Torso
GRF-3M
1E Right Arm
1E Left Torso
4M (+3) Right Torso
This is one of this rare instances that I can see the thought process here.
The Griffin-1S defines where things have to begin due to having the most hardpoints. At minimum, the missile Left Torso has to get hardpoint inflation because we know missile hardpoint usability starts at two. At this point, you could go "either way" on making it 3E and 3M or 4E and 2M. I would say that the usability of the mech greatly increases had it been 3E and 3M, but that's just me.
The Griffin-1N naturally has all its weapons to one side, so whatever you give to it would probably be unbalanced. The question you'd have to ask yourself is that what you'd want. It seems balanced, but something "naturally unbalanced" could work, like a 2E and 4M or even 4E and 2M to one side. It is a necessary discussion because of the Griffin-3M.
I think the Griffin-3M could have actually been better than it is... but it is "unique" in a sense. I would have pushed for a 2E Right Arm and 2E Left Torso to go with a 2M Right Torso. The reasoning for it is having clusters of energy and making it more viable for a Standard engine setup. It would still be the best mech, but for a different reason.
Of course, this is just an opinion and this would obviously affect the other 55-tonner balance while not knowing the other details. So, what I'm suggesting could be a bad idea balancewise as this becomes more difficult. It's easier to worry within the same chassis than it is on a more broader scale....
Kintaro
KTO-18
1E Right Arm
2M (+1) Right Torso
1E, 2M Left Arm
1M Center Torso
KTO-19
2E (+1) Right Arm
1M Right Torso
1E, 2M Left Arm
1M Center Torso
KTO-20
2E Right Arm
2M Right Torso
1M Left Arm
2E (+1) Center Torso
Not much hardpoint inflation was necessary here, but I think "unofficially" the magic # for hardpoints on a 55-tonner seemed to be 6 (with JJs) or 7 (w/o JJs). It should seem obvious this is the only 55-ton IS chassis w/o JJs... so it is what it is.
The Kintaro-18 is an interesting case as it was decided to expand its missile hardpoint. The choice was between the Center Torso and the Right Torso... and in some ways the Golden Boy (as it was released first) was the defining reason. If the missile hardpoint is in the CT, it wouldn't really differentiate itself very well due to it being pegged as a bigger LRM carrier... where the Golden Boy is limited in that form but gets the extra energy hardpoint....
The Kintaro-19 and Kintaro-20 really differ in the location of the extra energy hardpoint. The KTO-20 gains it in the CT, and while it has the most energy hardpoints of the Kintaros, it isn't that much more "effective" than the KTO-19 in energy damage potential. Had it been moved into the Right Arm instead on the KTO-20, it would have the potential for like 4LL builds... while not overpowering, it is a balancing thought. The KTO-19 properly has the energy hardpoint on the outside as it needs this...
Wolverine
WVR-6K
1E Head
4E (+1) Right Arm
1M Left Torso
WVR-6R
1E Head
2B (+1) Right Arm
3M (+2) Left Torso
WVR-7K
1E Head
2E Right Arm
1M Right Torso
2M (+1) Left Torso
While I love the 6K as is, I cannot help think why it wasn't given an extra hardpoint since it has no JJs. The only thing that was given to it is a greater engine cap... and at that point is a wasted effort in futile buffing. What could have been done is to give a 5E right arm (which this game doesn't have as an option, outside of the 6E in a Hunchback-4P or a Banshee-5S) or even a 2M Left Torso would have worked given the situation with LRMs. I mean... seriously... why bother?
The 6R is atrocious and is the worst of the bunch IMO, but the inflation there was appropriate into making it somewhat useful. The 1/2 energy hardpoint in the head is what it is for all the Wolverines anyways...
The 7K could have actually got in a different direction. Inflating the Left Torso's missile hardpoint allows it to be useful for a Standard engine design leaving one side not being that much more important than the other. What could have also have happened is leave it "unbalanced" and added an additional energy hardpoint on the arm. Either change wouldn't have broken the mech and probably would have made it the best variant anyways.
Shadowhawk
SHD-2D2
1M Head
1E Right Arm
2M (+1) Right Torso
1B Left Torso
1E Left Arm
1M Center Torso
SHD-2H
1M Head
1E Right Arm
2M (+1) Right Torso
3B (+2) Left Torso
SHD-5M
1M Head
2E (+1) Right Arm
1M Right Torso
2B (+1) Right Torso
SHD-2D
1M Head
1E Right Arm
1M Right Torso
1B Left Torso
2E (+1) Left Arm
1M Center Torso
SHD-2K
3E (+2) Left Torso
3M (+2) Right Torso
The Shadowhawk is the standard that all mediums are build around.... which says a lot. However, it has its own fair share of terrible foibles.
The 2D2 is perfect... in that the CT hardpoint wasn't inflated (and the RT got it)... although it could have been for the purposes of balance.
The 2H from day 1 IMO was an abomination due to use of 1 energy hardpoint. While it isn't like the Wolverine's head laser, I'd always like my options open energywise, so I would rather inflate that hardpoint instead of the missile hardpoint. I'm not a fan of the AC2 dakka, but the option is there.
The 2M would look eerily like the 2H modification I had described... but I would inflate the missile hardpoint instead of giving it more JJs. It would work and be unique on its own merits.
The 2D is not what I would called balanced in the sense that the ballistic torso side also has the most energy. My personal preference here would be moving the energy inflation to the other side as losing the left torso would be more catastrophic despite having a "shield side". It's just more harmonious in terms of mech building, but it is what it is.
The 2K needed the mass inflation, so it having "even" levels of inflation is alright. It's perfect as is despite having one less hardpoint like the 5M as shield arms are really valuable here anyways (JJ # increase doesn't interest me whatsoever here).
3) Solutions? - I just wanted to get this post done.
I don't have any big ideas really... I'm just hoping that this is taken into consideration into the dynamic hardpoint revamp.
Simply just double the energy hardpoint on the Cicada-3C and that should help this variant out greatly. Failure to do so will keep it to being "slightly better" than the Spider-5K that can carry a ERPPC that can't use JJs. While you could run AC2s on that due to the tonnage availability.. it is the least practical solution (primarily due in part with the state of the AC2.
So... please fix the Cicada-3C. Don't let it keep sucking like the Vindicator-1X (or others I can't really think off the top of my head because I gotta get other stuff done). It's not unsalvageable... it simply needs some tweaking.


The Cicada-3C: A History Of Hardpoint Inflation - Not All Hardpoints Are Created Equal
Started by Deathlike, Feb 12 2015 03:56 PM
BattleMechs Balance Gameplay
5 replies to this topic
#1
Posted 12 February 2015 - 03:56 PM
#2
Posted 12 February 2015 - 04:11 PM
Honestly for mechs under 50 tons ballistic hardpoints past the 1st shouldn't count against the hardpoint total. Or if they ever boost MGs back up, for every stock ballistic give the mech 4 ballistic hardpoints in that section if it is under 50 tons.
#3
Posted 12 February 2015 - 04:49 PM
Two things come to mind:
PGI's philosophy seems to be that big stock weapons = many hardpoints. E.g. 1 stock AC20 = 3 ballistic hardpoints
Unfortunately, this doesn't apply to the CDA-3C, But it should. It should also apply to the Adder, Warhawk and Summoner, if you ask me. But Clan hardpoints are a different story, for whatever reason.
Secondly, PGI has a policy never to modify the number of hardpoints. Ever. Even the worst mechs in the game, that would be easily fixed by adjusting the number of hardpoints, are locked. Quirks are the only balancing tool being used right now.
It's weird. But I don't think these things are ever going to change.
PGI's philosophy seems to be that big stock weapons = many hardpoints. E.g. 1 stock AC20 = 3 ballistic hardpoints
Unfortunately, this doesn't apply to the CDA-3C, But it should. It should also apply to the Adder, Warhawk and Summoner, if you ask me. But Clan hardpoints are a different story, for whatever reason.
Secondly, PGI has a policy never to modify the number of hardpoints. Ever. Even the worst mechs in the game, that would be easily fixed by adjusting the number of hardpoints, are locked. Quirks are the only balancing tool being used right now.
It's weird. But I don't think these things are ever going to change.
#4
Posted 12 February 2015 - 05:19 PM
Alistair Winter, on 12 February 2015 - 04:49 PM, said:
Two things come to mind:
PGI's philosophy seems to be that big stock weapons = many hardpoints. E.g. 1 stock AC20 = 3 ballistic hardpoints
Unfortunately, this doesn't apply to the CDA-3C, But it should. It should also apply to the Adder, Warhawk and Summoner, if you ask me. But Clan hardpoints are a different story, for whatever reason.
PGI's philosophy seems to be that big stock weapons = many hardpoints. E.g. 1 stock AC20 = 3 ballistic hardpoints
Unfortunately, this doesn't apply to the CDA-3C, But it should. It should also apply to the Adder, Warhawk and Summoner, if you ask me. But Clan hardpoints are a different story, for whatever reason.
For Clan Mechs, I don't believe it is necessary due to the bountiful hardpoints on most of the Clan Mechs... usually. Most of the time the Clan superior weapons (outside of ballistics/Gauss) do most of the work.
There are still some exceptions (Summoner being at the top of the list, the Warhawk is screwed for different reasons though). I don't think the Adder needs more honestly, unless there's some specialized build that needs to be constructed... (if it involves MGs, please
I still think to some degree that the hardpoint inflation counts are rather arbitrary and primarily focus on the # of hardpoints instead of how those hardpoints are used and its effectiveness relative to its location.
Quote
Secondly, PGI has a policy never to modify the number of hardpoints. Ever. Even the worst mechs in the game, that would be easily fixed by adjusting the number of hardpoints, are locked. Quirks are the only balancing tool being used right now.
It's weird. But I don't think these things are ever going to change.
It's weird. But I don't think these things are ever going to change.
If it were to remove hardpoints preemptively, I could be OK with that. The Vindicator is an obvious example and the Oxide initially (rumored?) to have more missile hardpoints would be at the top of the list. However... we're actually adding hardpoints here that actually exist in the mech (although the Cicada hasn't gotten through the process yet). With the "legomech" system, it is actually possible. The question is whether there needs actual design considerations (visual ones) that need to be made.
For instance, the Raven has missile hardpoints that differ in size based on the stock hardpoints (the Raven-3L vs the other variants that use the Right Torso missile rack)... that needed a special consideration for obvious reasons.
The Cicada-3C having 4 hardpoints in one section (at least under my proposed buff) would have to make sure it would all fit nicely.... which probably requires some time and effort to make it happen.
Edited by Deathlike, 12 February 2015 - 05:35 PM.
#5
Posted 12 February 2015 - 05:58 PM
Deathlike, on 12 February 2015 - 05:19 PM, said:
If it were to remove hardpoints preemptively, I could be OK with that. The Vindicator is an obvious example and the Oxide initially (rumored?) to have more missile hardpoints would be at the top of the list. However... we're actually adding hardpoints here that actually exist in the mech (although the Cicada hasn't gotten through the process yet). With the "legomech" system, it is actually possible. The question is whether there needs actual design considerations (visual ones) that need to be made.

Confirmed. Along with the jumping Banshee.
There's also this curious thing from the image library on the website:

The "jr7-o-alt-DO_NOT_USE.png"
Edited by Mcgral18, 12 February 2015 - 06:02 PM.
#6
Posted 12 February 2015 - 06:35 PM
Mcgral18, on 12 February 2015 - 05:58 PM, said:

Confirmed. Along with the jumping Banshee.
There's also this curious thing from the image library on the website:

The "jr7-o-alt-DO_NOT_USE.png"
Considering the state of hoverjets... I'd wonder how the Banshee hero would have fared in today's game. I assume it would be like the older Highlander-732....
3 user(s) are reading this topic
0 members, 3 guests, 0 anonymous users