16 replies to this topic

[Duke Nedo]

Been listening to town hall and reading posts this morning, and what just comes to mind is a rather simple solution. I am sure others had similar or even identical ideas, sorry if it has been posted before.

My point of view is that the intented result of ghost heat is much needed, but like many others I think it was implemented in a unconsistent and unintuitive way. Imo the ghost heat mechanism fails because of two reasons:

• weapon groups are inconsistent
  
• it does not address alpha from low-heat weapons

My suggestion would be to address both these points.



1. Inconsistent weapon groups.
Instead of these groups, I suggest grouping ALL weapons by heat. What ghost heat wants to achieve is basically a stacking penalty, but the arbitary groupings make it too easy to circumvent, while at the same time ruining some builds in the process. If we instead group all weapons that are fired within the same 0.5s, add up their heat and scale that value to an exponential function. That is, when you alpha, the generated heat will not be linear, but exponential, exactly as for ghost heat, but with the big change that any heat generating weapon will contribute. Something like the figure below:



Scaling here is arbituary. Point is that every weapon combination creating heat will contribute to this value, no longer any free or forbidden combinations. This will for example address PPC+large laser+medium laser+srm+lrm combinations, but will not take projectiles fully into account since they create little heat per damage, especially AC5 and gauss.

2. Alpha from low-heat weapons
In order to address these, I'd suggest adding a small recoil bump. An old idea, but it would help with the low heat high damage weapons. This would use the same principle that all weapons fired simultaneously add up their recoil (use impulse?) contributions and give the aim a bump in corresponding size. Fire another group before it has returned to equilibrium will bump it further before starting to relax again. The whole recoil-relax process need to be short, like 0.1 to 0.2 seconds. Just a small bump is needed. This will address all combinations of ballistics fired together with lasers, the recoil will spread the lasers around if you fire them first or simultaneously, or you will have to wait for recoil to relax before firing the laser if you fire the ACs first. Here you can tweak so that arms recoil independently from torso aim, and arms could recoil more than torso.

The only thing that may not be directly affected here is the gauss+ppc combo, and that one is difficult to address since the gauss absence of heat is mainly balanced against its weight. They both should have some recoil though so firing gauss+ppc when they are not in the same group will make the recoil of one affect the hit of the other. If you group them, the charge mechanics of the gauss will prevent alpha completely. The key is to prevent the window where both can be manually fired simultaneously to avoid the recoil kick. One could try to make the kick near instant, or prevent other weapons to fire just after the gauss charge release (0.1 secs or so), or combine this entire suggestion with delayed convergence. At least as I see it, delayed convergence can't replace ghost heat or any other stacking penalty alone, but having convergence doesn't exclude other solutions.

All in all, I don't think the heat mash-up will take much coding at all. It's more like tweaking the current ghost heat groups and some constants. The recoil needs some work but probably not much. There is impulse shake in the game, just apply some to your own mech when firing projectiles.

Now, back to work, flame away!

EDIT: Adding a 3rd point here from below further below, and highlighting some misunderstandings.

3. Convergence

True range convergence was tried before, like Hoax described below, and it makes it difficult to impossible to converge reliably on mech legs or on moving mechs (with netcode and hsr) - giving us a strong reason to always aim for the centre of mass = boring and less skill involved.

Instead, the second best choice would be to force unlocked arms, and I suggest also separating the aim reticules for the left and the right arm. That would always give you 3 aiming circles instead of the 2 you have now with unlocked arms. In practice, this would mimic the wanted effect from delayed convergence - i.e. that you have to be a bit patient and wait for perfect alignment before firing away if you want to hit where you are aiming. This also opens up for many interesting tweaks, for example:
- Individual recoil for arms and torso
- Individual speed for arms depending on how much stuff you loaded them with
- Different movement restrictions for forehand and backhand motions, allowing you to pivot out to the right with your right arm for example where your left arm doing a backhand motion will not be able to follow because of a clash with the torso. Different speeds in forehand and backhand motions are also required to separate the convergence of the arms from each other in the first place.


Final note: Ghost heat nor Convergence nor Cone of fire nor Recoil will alone be able to achieve the wanted effect - but a combination of 1) uniform heat stacking penalty, 2) Recoil and 3) Forced unlocked arms/separated left and right arms convergence could do the job.

The shape suggested here does require minimal coding so I think this suggestion is actually realistic.

Edited by Duke Nedo, 04 October 2014 - 01:29 AM.

[DEN_Ninja]

Duke Nedo, on 03 October 2014 - 04:24 AM, said:

Been listening to town hall and reading posts this morning, and what just comes to mind is a rather simple solution. I am sure others had similar or even identical ideas, sorry if it has been posted before.

My point of view is that the intented result of ghost heat is much needed, but like many others I think it was implemented in a unconsistent and unintuitive way. Imo the ghost heat mechanism fails because of two reasons:

• weapon groups are inconsistent
  
• it does not address alpha from low-heat weapons

My suggestion would be to address both these points.



1. Inconsistent weapon groups.
Instead of these groups, I suggest grouping ALL weapons by heat. What ghost heat wants to achieve is basically a stacking penalty, but the arbitary groupings make it too easy to circumvent, while at the same time ruining some builds in the process. If we instead group all weapons that are fired within the same 0.5s, add up their heat and scale that value to an exponential function. That is, when you alpha, the generated heat will not be linear, but exponential, exactly as for ghost heat, but with the big change that any heat generating weapon will contribute. Something like the figure below:



Scaling here is arbituary. Point is that every weapon combination creating heat will contribute to this value, no longer any free or forbidden combinations. This will for example address PPC+large laser+medium laser+srm+lrm combinations, but will not take projectiles fully into account since they create little heat per damage, especially AC5 and gauss.

2. Alpha from low-heat weapons
In order to address these, I'd suggest adding a small recoil bump. An old idea, but it would help with the low heat high damage weapons. This would use the same principle that all weapons fired simultaneously add up their recoil (use impulse?) contributions and give the aim a bump in corresponding size. Fire another group before it has returned to equilibrium will bump it further before starting to relax again. The whole recoil-relax process need to be short, like 0.1 to 0.2 seconds. Just a small bump is needed. This will address all combinations of balistics fired together with lasers, the recoil will spread the lasers around if you fire them first or simultaneously, or you will have to wait for recoil to relax before firing the laser if you fire the ACs first. Here you can tweak so that arms recoil independently from torso aim, and arms could recoil more than torso.

The only thing that may not be directly affected here is the gauss+ppc combo, and that one is difficult to address since the gauss absence of heat is mainly balanced against its weight. They both should have some recoil though so firing gauss+ppc when they are not in the same group will make the recoil of one affect the hit of the other. If you group them, the charge mechanics of the gauss will prevent alpha completely. The key is to prevent the window where both can be manually fired simultaneously to avoid the recoil kick. One could try to make the kick near instant, or prevent other weapons to fire just after the gauss charge release (0.1 secs or so), or combine this entire suggestion with delayed convergence. At least as I see it, delayed convergence can't replace ghost heat or any other stacking penalty alone, but having convergence doesn't exclude other solutions.

All in all, I don't think the heat mash-up will take much coding at all. It's more like tweaking the current ghost heat groups and some constants. The recoil needs some work but probably not much. There is impulse shake in the game, just apply some to your own mech when firing projectiles.

Now, back to work, flame away!

This...is actually interesting. I remember seeing something similar back in the day but you are making some sense.

If this could be done right it would certainly make a lot of sense and is easily explained.

[Duke Nedo]

Thanks!

Perhaps a few more comments I forgot in the OP.

- Like you discussed in the town hall meeting, forcing unlocked arms would fix Gauss+ppc pinpoint for all chassi except the DW, wouldn't it? That would be much easier to implement than true convergence + HSR...

- To this one could add separate aim for the two arms. In order to separate them, give slightly different speeds to forehand and backhand motions, and also modify arm speed by how much weight you put into each arm. This would "unpin" the DW by a tiny bit atleast.

- and... let's avoid the random cone of fire thing, that one I feel strongly against. If I pull of an insane headshot I'd like to hear "nice shot man" and not "you lucky *******!"

Edited by Duke Nedo, 03 October 2014 - 05:28 AM.

[kapusta11]

So... Another mechanic that hurts *all energy* Assaults and Heavies but leaves Twin Gauss and Pure ballistic builds alone? Oke.

[Hoax415]

Preventing alpha through heat is silly.

Preventing alpha through reduced pinpoint accuracy when firing all weapons at once is much less silly.

What is the real problem we are trying to address with ghost heat? Is it that mechs don't heat up (quickly) enough?

No. Ghost heat is trying to address the gameplay problems that arise when every pilot is putting all his damage in one exact place.

Therefore the correct solution is to address the problem itself directly in the simplest manner. Why should firing 4 PPC's at once be as accurate as firing a single medium laser?

Forget ghost heat. Its a band aid that very poorly sort of addresses the actual problem while creating a cumbersome secret system ripe for exploitation by mechs that have the layout to avoid it.

Edited by Hoax415, 03 October 2014 - 10:04 AM.

[Roadbeer]

kapusta11, on 03 October 2014 - 09:56 AM, said:

So... Another mechanic that hurts *all energy* Assaults and Heavies but leaves Twin Gauss and Pure ballistic builds alone? Oke.

Well, that could be simply addressed by re-implementing rearm, not rearm/repair, just rearm.
Lurmageddion = solved. Spamming 1440 missiles suddenly costs you money to drop, even on a win.
AC/? spam = solved (see above)

Make THAT the penalty for ballistics/missiles and not some arbitrary penalty for firing more than one AC/2

[Xanquil]

Easier solution (and one I've stated before)
1: Every time you fire there is a .5sec global cool down no matter if it is a single weapon or an alpha.
2: If you chain fire or fire a single weapon aiming isn't changed.
3: Firing more than one weapon causes them to start to scatter like SRMs.
3a: Two weapons is a minor (SRM2) spread, but the more you fire at once the "worse" it gets.
4: Targeting computer reduces spread by a % (10-15%)
5: This would apply to every weapon not just a select few.

This uses systems already in the game, simulates the standard battle computers attempt to comply with the pilots wishes, and fixes a lot of the issues we have now.

The reason for #1 is because of people like me who have a programmable keyboard/mouse.
#2 Is where your skill shots are. Being able to hit the same place with 4 PPCs when fired separately is skill, magic alpha is not.
#3 Yea it has RNG but honestly it needs to happen. (as much as I hate RNG) Alpha strikes should never be 100% predictable, some of the shots should have a chance to miss, otherwise it continues to be the much better option to firing single weapons. (And it shouldn't.)

The reason Alpha strikes are as big of a problem as they are is because the armor system used always assumed two things.
That all weapons fired in a round weren't always going to hit, and that all weapons that did hit were (almost) never going to going to hit the same location. MWOs alphas completely destroys that system causing all sort of game breaking issues.

This is all before heat scale starts messing with your accuracy.

[Duke Nedo]

kapusta11, on 03 October 2014 - 09:56 AM, said:

So... Another mechanic that hurts *all energy* Assaults and Heavies but leaves Twin Gauss and Pure ballistic builds alone? Oke.

I'd see this suggestion more as a model to apply a stacking penalty to high alpha uniformly over all weapons. I'd like to see it scaled to allow a bit more energy heat than now, for example if you consider pure heat output by this model, 2 ER-PPC (30 heat) and 4 Large Lasers (28 heat) would trigger about the same base level of stacking penalty to heat. More importantly, firing 2 ER-PPCs + 2 LL + 6 ML would trigger the stacking penalty to. That's the whole point with point #1, to make it apply uniformly to all combinations of weapons that produce heat. I don't like the term "ghost heat", this would be more like "heat overload".

The pure AC-builds are tougher, that's why I suggested a recoil. That will address all combinations of ballistics and lasers as well as UAC double-tap boating, and the only pure AC-build that could be considered a problem would be the 2x AC20. Not the biggest of problems imo since it is short ranged and slow, but still. It would trigger one of the biggest recoils so it would at least be difficult to follow up with lasers, and forcing unlocked arms & separated left and right arms would help too. The only dual gauss build that is a problem is the 2x gauss + 2x cer-ppc and that one is discussed in the OP.

Hoax415, on 03 October 2014 - 10:04 AM, said:

Preventing alpha through heat is silly.

Preventing alpha through reduced pinpoint accuracy when firing all weapons at once is much less silly.

What is the real problem we are trying to address with ghost heat? Is it that mechs don't heat up (quickly) enough?

No. Ghost heat is trying to address the gameplay problems that arise when every pilot is putting all his damage in one exact place.

Therefore the correct solution is to address the problem itself directly in the simplest manner. Why should firing 4 PPC's at once be as accurate as firing a single medium laser?

Forget ghost heat. Its a band aid that very poorly sort of addresses the actual problem while creating a cumbersome secret system ripe for exploitation by mechs that have the layout to avoid it.

I strongly disagree with random spread... if that was the only solution to big alphas you would still see the PPC shotguns out there... I dont think that alone would do the job and I would hate to see fired rounds go where I did not fire them.

I do however like delayed convergence a lot and it can in every type or form be combined with my suggestions above. Problem is I read that the main difficulty with this is due to HSR, so that sounds like game over... the poor mans solution to this could be forced unlocked arms and the separation of the left and right arms in convergence. Most big weapons have to be put in different arms, except for the symmetric Direwolf then that has 2 in each arm.


The point I wish to make here is that this flavor of heat stacking penalty is uniform to all combinations of weapons, and it requires very little coding from PGI. Basically just grouping all weapons together and apply different scaling constants (and reducing medium lasers to 4 heat...).

The recoil thing is basically there as well, add some impulse when firing them big guns. Forcing unlocked arms is easy, the only thing requiring some coding would be separating the "convergence" of the right and left arms from each other and giving them different speeds due to direction and weight.

Edited by Duke Nedo, 03 October 2014 - 11:21 AM.

[Hoax415]

Duke Nedo, on 03 October 2014 - 11:18 AM, said:

I strongly disagree with random spread... if that was the only solution to big alphas you would still see the PPC shotguns out there... I dont think that alone would do the job and I would hate to see fired rounds go where I did not fire them.

I do however like delayed convergence a lot

You would not see 4xPPC being a top-tier build if a sane cone of fire system was implemented especially one where the cone is further affected by heat levels like it should be because this is a Battletech game and that's how heat works in BT.

Delayed convergence ala WoT will suffer from a lot of the same problems that the old convergence system suffered from. Mechs are not tanks, they have much stranger geometry. Aiming at legs becomes a massive headache and there is nothing in WoT that moves anywhere near as fast as a Spider.

Convergence was tried and it was awful because of light mechs and it removed ever aiming at anything but center mass for fear of convergence getting messed up.

Edited by Hoax415, 03 October 2014 - 11:35 AM.

[Belorion]

No... just no.

Any solution that will be accepted will be simple and effective. You lost on simple in your first sentence.

[Duke Nedo]

I feel for you if I lost you at the first sentence.

The current system is complicated. This suggestion is very simple. Add the heat of everything you fire together within 0.5s just like now, if its too much you get a stacking penalty on top. There you go!

[Felio]

I don't think recoil is really a fun thing. It wouldn't lower a FLD alpha anyway, and dealing with it on the AC/2... I don't think the weapon needs that nerf.

I do support a more comprehensive heat penalty system, though. The current one is just too easy to work around. I would probably also not reset the heat penalty completely every .5 seconds. You can't do 30 damage and then 30 damage a half-second later. That's barely an improvement.

At the end of the day, I just think one-shot kills, or even one-shot maimings, are something that should never happen. If that means adjusting damage with the Locust in mind, then that's what it means. Either that or you adjust the Locust in some way.

[Hoax415]

Felio, on 03 October 2014 - 12:52 PM, said:

I don't think recoil is really a fun thing. It wouldn't lower a FLD alpha anyway, and dealing with it on the AC/2... I don't think the weapon needs that nerf.

Why would the AC/2 have high recoil?

You would obviously expect AC2's to have the least recoil out of all autocannons.

You are correct about one thing though, recoil does absolutely nothing about pinpoint alpha on its own. Because by definition recoil is something that happens after you fire. So presumably you could still alpha with perfect accuracy it would only be your second shot that it affected.

Recoil really would just be a way to balance weapons by giving another stat to play with. But its not a terribly useful one because obviously giving laser weapons recoil would feel pretty stupid.

Anyone suggesting to just add recoil and be done with it does not understand the actual problem of pinpoint alpha or they have a strange idea of what "recoil" means as a game system.

Edited by Hoax415, 03 October 2014 - 02:07 PM.

[Xtrekker]

My solution to this was to just introduce a new mechanic instead of trying to re-engineer existing systems. Add a maximum power output per chassis. Just like any power source, a fusion engine would have to have some kind of power regulation prior to actual weapon capacitors, and that results in a limit of total power at any given instance. Weapons, jets, etc., could be assigned values for power requirements, and if total weapons fired at one time exceeded that limit, they should either fail to fire or take damage due to insufficient amps.

I absolutely agree with recoil as well, with more of an effect introduced for arm-mounted ballistics.

So this would give us 3 systems to work with...heat, power, and recoil. A truly "skilled" player would be good at managing their build on multiple levels, not just heat all the time. And it would effectively deal with high alpha builds. If someone absolutely had to get that alpha out, they could override to "high output" and take the weapon damage.

Actually, I would go so far as to say this power output would primarily affect gauss and energy weapons. The recoil mechanic would work well for ballistics (and give an interesting effect to asymmetrical builds or chain-fired ballistics). I would assume volleys of LRMs would also have a recoil effect, although impact on play would really be negligent unless you were stunlocking someone as you charged in with other weapons.

Anyway, any discussion on the matter is good, because the current system is really unimaginative and allows far too much exploitation.

Edited by Xtrekker, 03 October 2014 - 03:50 PM.

[Rebas Kradd]

Hoax415, on 03 October 2014 - 10:04 AM, said:

Preventing alpha through heat is silly.

Even though that's exactly how lore does it?

Hoax415, on 03 October 2014 - 10:04 AM, said:

Preventing alpha through reduced pinpoint accuracy when firing all weapons at once is much less silly.

The ability to target components of a mech is a hallmark of BattleTech lore. I tear my hair out every time I see people asking for this to be removed for the sake of eliminating pinpoint alphas. It's throwing the baby out with the bathwater. Being able to isolate and focus on arms or torsos of mechs in order to disable critical weapons or systems is a hugely important nuance and one that would be sorely missed. Asking for anything else makes targeting arms/torsos inefficient, and turns "shoot the center torso" into the best bet for any 1vs1 situation.

"You don't like that Gauss rifle in my arm? TARGET THE ARM. Oh, but you can't do that, because your four medium lasers have been given a RNG spread (that is, a nerf). You notice that my left torso is cherry-red? TARGET THE TORSO. Oh, but you can't, because your medium lasers are scattering across the reticle and can't reliably hit the component you want. Talk about frustrating." This is a fundamental reason why pinpoint accuracy, or something approaching it, has to be preserved.

You mentioned alphas being ideal in focus-fire situations. Honestly, in claustrophobic 12v12 matches, it hardly makes a difference in focus-fire situations. The difference between seven mechs alpha'ing you and seven mechs scattering their fire all over you is the difference between dying in twenty seconds or dying in thirty seconds. You're dead either way, especially in an Assault. So I don't see the focus-fire issue as being all that relevant.

Your suggestion stomps on MechWarrior lore on at least two fronts. Heat-related solutions make the most sense for eliminating alphas.

Edited by Rebas Kradd, 03 October 2014 - 03:53 PM.

[Rebas Kradd]

Xtrekker, on 03 October 2014 - 03:37 PM, said:

So this would give us 3 systems to work with...heat, power, and recoil. A truly "skilled" player would be good at managing their build on multiple levels, not just heat all the time.

The tradeoff there is new players. They'll run screaming into the blue trying to figure out this system.

[Duke Nedo]

Felio, on 03 October 2014 - 12:52 PM, said:

I don't think recoil is really a fun thing. It wouldn't lower a FLD alpha anyway, and dealing with it on the AC/2... I don't think the weapon needs that nerf.

The recoil I suggested is not a nerf to ac2 in any way. I suggested a recoil-relax time of ~0.1-0.2 seconds and that is much less than the rof of ac2's. Maybe if you macro-fire them for shakes, but that would probably be a welcome "nerf" The take home message for AC2's in this suggestion is that they would not trigger ghost heat when boated and I don't think they should. With this generalized method of calculating heat stacking penalty, you could fire AC2's like 30 times within 0.5 secs without triggering much penalty... What the recoil suggestion DOES do is to prevent simultaneous 2x gauss 4x ER-LL builds for example by either bumping the lasers around a bit, or by forcing you do wait for the recoil to relax before adding the lasers -> less pinpoint alpha character on the volley.

Hoax415, on 03 October 2014 - 02:05 PM, said:

Anyone suggesting to just add recoil and be done with it does not understand the actual problem of pinpoint alpha or they have a strange idea of what "recoil" means as a game system.

True, see above. Recoil is mainly to offset ballistic+laser combinations and uac double taps. The only thing that is left out is pure ballistics boating, but imo that is not a huge problem. The 2x AC20 is a nasty hit, but the way I see that one is more like a high risk high reward weapon combo. It's not really gamebreaking. Perhaps the King Crab will change that, but since the clan version is a dot version there is no real problem right now. Dual gauss + ppc is the only difficult one imo, if you manualy manage to fire them exactly simultaneously, then recoil will fail to touch the ppc shot. Their projectile speeds are very different now though so I find it more efficient to fire them separately and lead aim differently as it is now....

This is also why I would like to promote forced arm unlock and separating the left arm from the right arm in the aiming reticule. That would give some sense of "convergence" without going back to the system that did not work very well. Like described above, aiming for legs or even aiming for a moving mech in general with half-crappy netcode so that yon dont really know if you are aiming at it or behind it makes it really difficult. I like the concept, but sounds technically difficult to implement in a good way, and it would also reduce the pace of the game which is already very low for many players.

Another important thing about convergence as I see it is also that alone it would not do the job to prevent the high alpha builds. If you have to wait for perfect convergence for 0.5-1.0 seconds that is not so much for hitting a heavy or assault. You would still get 6x ppcs pinpointed to your torso if that was the only mechanism in play. It would be much less likely for lights and moving mediums, but I still think we would see them quite a lot. Even convergence needs the backup from other mechanisms that work against the alpha builds.

Xtrekker, on 03 October 2014 - 03:37 PM, said:

So this would give us 3 systems to work with...heat, power, and recoil. A truly "skilled" player would be good at managing their build on multiple levels, not just heat all the time. And it would effectively deal with high alpha builds. If someone absolutely had to get that alpha out, they could override to "high output" and take the weapon damage.

May be a bit complicated with adding another system right now, and at least from a logic perspective, all AC's would have 0 power demand so it would basicly be a anti-gauss and anti-ppc mechanism?