Jump to content

Weapon- And Alpha-Balancing: Real Mech Combat With Gcds!


114 replies to this topic

Poll: Weapon- And Alpha-Balancing: Real Mech Combat With Gcds! (117 member(s) have cast votes)

Do you like this idea?

  1. Yes! (99 votes [84.62%] - View)

    Percentage of vote: 84.62%

  2. No, because... (18 votes [15.38%] - View)

    Percentage of vote: 15.38%

Vote Guests cannot vote

#21 Phaesphoros

    Member

  • PipPipPipPipPipPipPip
  • 513 posts

Posted 04 July 2013 - 08:59 AM

This is somewhat similar to Homeless Bill's idea. But it is much simpler due to a harsher penalty (cannot fire instead of loss of accuracy). I'm curious as to what HB thinks about this.

I'd guess the numbers would need some adjustment, as .5 seconds is not much for dual Gauss / dual AC/20.

#22 MustrumRidcully

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 10,644 posts

Posted 04 July 2013 - 09:17 AM

View PostPhaesphoros, on 04 July 2013 - 08:59 AM, said:

This is somewhat similar to Homeless Bill's idea. But it is much simpler due to a harsher penalty (cannot fire instead of loss of accuracy). I'm curious as to what HB thinks about this.

I'd guess the numbers would need some adjustment, as .5 seconds is not much for dual Gauss / dual AC/20.

It doesn't really need to have a large number. 0.5 seconds is enough to force you to to "reaim" and just too little that you could just torso twist away, so you're Dual (or hypothetical Triple) Gauss or AC/20 boat has to spend as much time facing the enemy as a laser use, and also risks not hitting the same spot with both (or all three) shots.

The biggest challenge FuDup alludes to with global cooldowns was finding cooldowns that are not too long, but also not to short to be inconsequential. This approach decides that not all weapons need the same - for example, lasers might not need one at all, AC/2s can only have a very small one (if they need one at all, boating ACs is really not about alpha strikes.).

Edited by MustrumRidcully, 04 July 2013 - 09:25 AM.


#23 Stat1cVoiD

    Member

  • PipPipPip
  • Bad Company
  • 83 posts

Posted 04 July 2013 - 09:38 AM

View PostBillyM, on 04 July 2013 - 08:24 AM, said:

Like it, but a few caveats...

Six...
Large...
Lasers...

Six...
SRM...
Sixers...

Gotta get more granular than just "global or not"...

-3LL shooting at once, ok... 6LL isn't
-3srm6's shooting at once, ok... 6srm6's isn't.

...so, weapon#-based global cooldown is required IMO

--billyM


With Heat, we already have a system that is supposed to punish 6 LL-Boats... I don't say that my GCD-suggestion would be the ultimate solution to fix ALL balancing issues in this game, since the weapons themselves have still to be tweaked and adjusted.
It is just about denying the current problem of people stacking up on high-pinpoint-damage weapons and delivering a focused punch for up to 60 dmg to one single component, which just breaks the game's mechanics and is way too easy to (ab)use.
Even if you fire 6 LLs at once, you have to focus them on a single component for a whole second to maximize their effectiveness and the enemy pilot can still start to torso twist, to make this pretty much impossible.

SRMs, as I already said before, have already huge drawbacks, regarding that their maxrange hardcaps at about 300 meters... Therefore i think they should have an advantage on their small range (Which they have currently clearly not) and given how they work they are also VERY likely to spread their damage over multiple components...

My proposal is not about reducing the overall DPS, it's just about stopping this horrendous "One-Shot"-Gameplay which is currently going on.

And i think "simple is best" :). That's because i didn't want to make specific extra rules of CDs for every single weapon in the game, because it would make it way too complicated, from my point of view.
The game still has to feel intuitive imho... :)

View PostPhaesphoros, on 04 July 2013 - 08:59 AM, said:

This is somewhat similar to Homeless Bill's idea. But it is much simpler due to a harsher penalty (cannot fire instead of loss of accuracy). I'm curious as to what HB thinks about this.

I'd guess the numbers would need some adjustment, as .5 seconds is not much for dual Gauss / dual AC/20.


Thats absolutely true. These numbers need testing! But thats what a Beta is for, isn't it? ;)
As mentioned above I didn't want to complicate the current system more than required, because of accessibility and how hard it would actually be to implement.
But those numbers are surely not carved in stone... ;)
My chain of thought behind this is exactly what MustrumRidcully wrote in the post above this one:
It doesn't have to be a large number. 0.5 sec might still be enough to raise the skill-requirement by a lot and forcing the AC/20-pilot to keep the focus longer on the target than he would like to. ;).
However, from my point of view, the AC/20 has to stay a scary weapon, since its short range and slow bulletspeed deliver already some decent drawbacks to make up for the high damage.
Therefore i chose the 0.5 secs to make hitting 2 of 'em not guaranteed anymore, but also not a matter of luck... But you are right: If this is the right number for this weapon or not, has to be tested, because I am not omniscient... ;)

Edited by Stat1cVoiD, 04 July 2013 - 10:09 AM.


#24 Kvitta

    Rookie

  • 1 posts

Posted 04 July 2013 - 11:50 AM

Good Very Nice. :) Jarrrrrrrrrrpppppppppp

#25 Prouf

    Rookie

  • Bad Company
  • 3 posts

Posted 04 July 2013 - 11:52 AM

View PostStat1cVoiD, on 04 July 2013 - 03:38 AM, said:

I do not think that turning a TT into a FPS can be achieved by holding on to all the mechanics and rules of the board game.
BUT:
PGI altered one core mechanic and idea of BTech which now screws the balancing up big time:
(Either by misinterpretation or by purpose, but from my point of view it obviously didn't work out)

How can it be in the TT, that two similar, parallel aligned weapons, mounted on the same bodypart, hit different locations if fired at the same time?

A: It is impossible.

Epic Illustration:
Posted Image



That means: Mechs are not supposed fire all their weapons at once!
They fire their weapons one after another.

Epic Illustration:
Posted Image


But why would they do that?
Well, I am not a huge BTech-Geek, but I know some people who are and they said:
the targeting system is rubbish and the reactor cannot handle the energy spikes generated by multiple weapons, fired at the same time.

But from my point of view there is one simple reason why they do that:
It looks way more awesome!

If you imagine a Mech Fight, do you think of huge battle robots constantly spiting out bullets, rockets and lasers or just firing one big blast, followed by running in circles while they are waiting for their CDs to finish?

Exactly!

So how can this system be implemented into MW:O and how would the game profit?

How to implement?
The most obvious solution would be to force chainfire on ALL weapons, which would mean that you would have to wait 0.5 seconds to fire the next weapon after you fired ANY weapon.
Doing this however, would just break the game for obvious reasons.

So how to do it?
Two words:
Global Cooldowns.
Add global cooldowns to different types of weapons.

Weapons without global cooldown (can be fired at any time and AT THE SAME TIME, if the respective CDs are ready):

SL
ML
LL
LRM
SRM
SSRM
AC/2
Flamer
Machine Gun

Why those weapons are not affected:
- Lasers already spread their damage, given how they work.
- how many missiles can be fired at once is already limited by the number of missile tubes on your chassis.
- AC/2, MG and Flamer for very obvious reasons.


Weapons affected by global cooldown (The time behind the weapon's name shows how long they will trigger a global CD on all those weapons after being fired):

AC/5 (0 sec)
Ultra AC/5 (0 sec)
AC/10 (0.25 sec)
LB 10-X AC (0.25 sec)
AC/20 (0.5 sec)
Gauss (0.5 sec)
PPC (0.5 sec)

What does that mean:
You cannot fire those weapons at the same time anymore. After having fired one of the above mentioned weapons you have to wait for 0.25 to 0.5 seconds before being able to fire off any other of those weapons.
The only exception are the AC/5s. You can fire multiple AC/5s at the same time or along with another weapon of this group, but it is still affected by the global CD.

Examples:

- A Stalker with 4 PPCs and 2 MLs plays like he had forced chainfire on all PPCs, but can fire both of his MLs off whenever he likes.

- A Hunchback with 9 MLs can fire them all at once, at any time, given the ML's CDs are ready.

- A Highlander with 2 AC/5s, 2 PPCs, 1 LPL and 2 SSRM can fire the LPL and the SSRMs whenever they are ready. He could also do that with the AC/5s, as long as he doesn't fire the PPCs. If he fired 1 PPC he has to wait for 0.5 seconds to fire the 2nd or to fire the AC/5s.

How will the game profit?

First of all, this will solve most of the the high-Alpha-Problem, because all high pinpoint-damage weapons have to be fired successively, making it much harder for a player to hit the same component with all of his weapons.
Secondly, it will make Lasers and Missiles more attractive, since they can still be fired at once or along with the pinpoint-damage-weapons.
Thirdly, it will nerf the unfair defensive power of high-alpha builds, because you cannot fire your volley and immediately Torso-Twist anymore. Even PPC-builds will now have to face the target for a longer period of time, if they want to deal all their damage.
And last but not least:
The game would look and play so much better; instead of sniping one big volley, Mechs would unleash a Bulletstrom as intended.

What do you guys think of it?
And @PGI: PLEASE TAKE A LOOK AT IT! IT WOULD MAKE THE GAME SO MUCH MORE FUN TO PLAY, TO WATCH AND IT WOULD RAISE THE SKILL-CAP FOR COMPETETIVE PLAY!



this is a very well writen plan my friend and makes sense buuut your drawing suck

#26 MustrumRidcully

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 10,644 posts

Posted 04 July 2013 - 12:16 PM

If you want something more... detailed.
Lasers might also need a global cooldown, but probably lower than that of other weapons.

One approach I thought about was: Limit the potential DPS over a short time frame, say, 0.5 seconds. This is not the total DPS the mech has (you might deal 20 DPS in 0.5 seconds,but then 0 for another 3.5 seconds).

If we think that 0.5 seconds is a good delay for the AC/20, the timeframe could be chosen as 0.5 seconds, and the damage threshold be 20.

So for a Medium Laser, we want to press beam time worth 4 seconds in a 0.5 second time interval. Someone less tired and better at math than me can now certainly figure out the "correct" global cooldown for the Medium Laser.

You probably need some knowledge mathematical series to formulate the correct "math" behind this.

#27 PEEFsmash

    Member

  • PipPipPipPipPipPipPipPip
  • Philanthropist
  • Philanthropist
  • 1,280 posts
  • LocationLos Angeles

Posted 04 July 2013 - 12:24 PM

The current balance problems are not alpha-strike problems. The problem is that there is no balance between brawlers/snipers. Snipers win at everything because brawlers are relatively very weak. SRM buffs, laser buffs, seismic nerf, etc will make high-heat alpha strikes pretty poor right off the bat.

The OPs suggestion isn't idiotic because he realizes that lasers deserve no nerfs. However, in the intrests of simplicity, the game could be more easily balanced by making simple, one-value changes like the ones above.

#28 Stat1cVoiD

    Member

  • PipPipPip
  • Bad Company
  • 83 posts

Posted 04 July 2013 - 03:21 PM

View PostPEEFsmash, on 04 July 2013 - 12:24 PM, said:

The current balance problems are not alpha-strike problems. The problem is that there is no balance between brawlers/snipers. Snipers win at everything because brawlers are relatively very weak. SRM buffs, laser buffs, seismic nerf, etc will make high-heat alpha strikes pretty poor right off the bat.

The OPs suggestion isn't idiotic because he realizes that lasers deserve no nerfs. However, in the intrests of simplicity, the game could be more easily balanced by making simple, one-value changes like the ones above.


To make brawling more viable it would require a huge map rework imho. Just tweaking some weapons wouldn't be enough since good teams will always be able to deny brawling attempts, because the maps are too open in general and offer too few cover.
I absolutely agree with SRMs being in need for a buff and seismic requiring a nerf,... I opened and participated in threads about those and other issues in the past, but thats not the point of this current thread.

And high Alphas however are a huge problem, because they are currently the easiest and most efficient way to kill enemy Mechs.
As long as you spread the damage over a Mech, you might require up to 8 (!) times the damage to kill it, as against just sniping a vital component. Additionally they offer a huge defensive benefit by being able to fire their volley and move back into cover or twist the torso like crazy.

Given how MW:O works, I assume that as long as you can fire multiple pinpoint weapons at once, pinpoint-damage weapons will always be in a huge advantage.
And even if you nerf f.e. PPCs into oblivion (Don't get me wrong, they need a nerf right now!) most players will just take the next fotm which is able to focus all damage on one single spot in one shot.

And as said in my opening post: It would play out way more interesting and look much better, if mechs would fire their weapons more constantly as against firing everything in single big blast. :)


View PostMustrumRidcully, on 04 July 2013 - 12:16 PM, said:

If you want something more... detailed.
Lasers might also need a global cooldown, but probably lower than that of other weapons.

One approach I thought about was: Limit the potential DPS over a short time frame, say, 0.5 seconds. This is not the total DPS the mech has (you might deal 20 DPS in 0.5 seconds,but then 0 for another 3.5 seconds).

If we think that 0.5 seconds is a good delay for the AC/20, the timeframe could be chosen as 0.5 seconds, and the damage threshold be 20.

So for a Medium Laser, we want to press beam time worth 4 seconds in a 0.5 second time interval. Someone less tired and better at math than me can now certainly figure out the "correct" global cooldown for the Medium Laser.

You probably need some knowledge mathematical series to formulate the correct "math" behind this.


Personally I don't see Lasers, esp. MLs, being in need for any sort of tweaking. I think they work absolutely fine how they currently are and require a good pilot to be a serious threat.
Another reason why i didn't added many weapon-types to the GCD is, that it might encourage people to build Mechs with more different sorts of weapons, instead of just sticking to 2 or 3,...
Right now ML- or even LL-Boats doesn't seem to be a Gamebreaking issue.
All this requires testing however and if Lasers would seem OP then, they could still be tweaked accordingly with a GCD... :)

Edited by Stat1cVoiD, 04 July 2013 - 03:33 PM.


#29 Homeless Bill

    Member

  • PipPipPipPipPipPipPipPip
  • The 1 Percent
  • The 1 Percent
  • 1,968 posts
  • LocationA Box Near You

Posted 04 July 2013 - 03:40 PM

I like your system better much better than the heat solution, but I think that my proposal addresses the fiddly bits better while also not so harshly penalizing a certain class of weapons.

My system operates on the principle that you can still alpha as much as you want, but your weapons (namely the ones dealing pinpoint damage) will start to take an accuracy penalty at a certain threshold. I think that preserves a valid tactical choice; alphas aren't necessarily a bad thing. I just see no reason that they should be able to deal all their damage to a single location.

My main problem: disallowing a tactic is lazy when there's a way to preserve and balance it. It's akin to removing shooting while jumpjetting to counter poptarts.

Additionally, I think my TCS numbers are a better way of balancing the details. I see no reason you shouldn't be able to fire 2xAC/10s together. But not 3 without a slight penalty. Same with PPCs. Enforcing chainfire on a certain class of weapon both lacks solvency for missiles and lasers (because at some point, there's a limit for every weapon), and over-penalizes pinpoint loadouts. It would be a huge nerf to snipers, and I'd rather leave the roles as balanced as possible. Why should you be able to fire off one last huge SRM barrage, but not with your ballistics?

Like most solutions, it sacrifices a small bit of complexity for a large bit of solvency.

Edited by Homeless Bill, 04 July 2013 - 03:42 PM.


#30 Urdnot Mau

    Member

  • PipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 501 posts

Posted 04 July 2013 - 04:58 PM

View PostHomeless Bill, on 04 July 2013 - 03:40 PM, said:

I like your system better much better than the heat solution, but I think that my proposal addresses the fiddly bits better while also not so harshly penalizing a certain class of weapons.

My system operates on the principle that you can still alpha as much as you want, but your weapons (namely the ones dealing pinpoint damage) will start to take an accuracy penalty at a certain threshold. I think that preserves a valid tactical choice; alphas aren't necessarily a bad thing. I just see no reason that they should be able to deal all their damage to a single location.

My main problem: disallowing a tactic is lazy when there's a way to preserve and balance it. It's akin to removing shooting while jumpjetting to counter poptarts.

Additionally, I think my TCS numbers are a better way of balancing the details. I see no reason you shouldn't be able to fire 2xAC/10s together. But not 3 without a slight penalty. Same with PPCs. Enforcing chainfire on a certain class of weapon both lacks solvency for missiles and lasers (because at some point, there's a limit for every weapon), and over-penalizes pinpoint loadouts. It would be a huge nerf to snipers, and I'd rather leave the roles as balanced as possible. Why should you be able to fire off one last huge SRM barrage, but not with your ballistics?

Like most solutions, it sacrifices a small bit of complexity for a large bit of solvency.


Both ideas want to achieve the same goal. I'm not sure what the differences would be, though. Also, your idea (as you mentioned) seem a lot more complex to implemente (you can also recall the delay it took to PGI come up with an answer to poptarting).
In my opinion, both ideas are valid and i would gladly test them.

#31 Aylek

    Member

  • PipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 292 posts
  • LocationBerlin, Germany

Posted 04 July 2013 - 09:10 PM

Very well though out proposal including epic illustrations. A system like this (added on top of our new heat penalty system) would probably solve the alpha strike meta even if the GCD starts after firing two weapons of a certain class you developed in short.

The only problems I see are the dev's approval and the complexity of implementation this short before launch.

#32 Kmieciu

    Member

  • PipPipPipPipPipPipPipPipPip
  • Urban Commando
  • Urban Commando
  • 3,437 posts
  • LocationPoland

Posted 05 July 2013 - 03:29 AM

TL:DR Global cooldowns for (ER)PPC, Gauss and AC20 (0,5 second).

I like that idea. Simple and elegant. No random factors are introduced to the game. Headshots will take more skill. Blowing up side torsos will require more hits if the target is torso twisting.

This idea requires no changes to the GUI. If you group 6 PPC and press fire once, you will shoot all of them, one by one (but it will take 3 seconds). If you've seen people chain firing at the beginning of the match you would know how awesome it looks and sounds.

People will adapt and set the PPC on chain fire by themselves. Global server-controlled cool-down will prevent any macros.

Try using a 4xPPC Stalker on chain fire, you suddenly realize it's not that overpowered against 5xLL Stalker using group fire. It takes 1 second of aiming to fire 5xLL. It would take 1.5 seconds to fire the 4xPPC. I think that would bring the high-alpha boats in line with other weapons.

SRMs should also be on the 0.5 second cooldown, but with the original 2.5 damage per rocket. That way we deal with the splat cat, while giving back the best brawling weapon to the light and medium mechs. Or the SRMs could be set to ripple-fire. http://www.youtube.c...DhoAEO-jo#t=14s

Edited by Kmieciu, 05 July 2013 - 04:11 AM.


#33 Nik Reaper

    Member

  • PipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 1,273 posts

Posted 05 July 2013 - 04:41 AM

I would only argue that engines have an weapon energy output rating so that it reduces global cooldown for weapons and in some larger cases allow for 2 of a large type to be fired at once... 2 ppc together from an assault is not that unbecoming.

#34 Nik Reaper

    Member

  • PipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 1,273 posts

Posted 05 July 2013 - 04:47 AM

View PostPEEFsmash, on 04 July 2013 - 12:24 PM, said:

The current balance problems are not alpha-strike problems. The problem is that there is no balance between brawlers/snipers. Snipers win at everything because brawlers are relatively very weak. SRM buffs, laser buffs, seismic nerf, etc will make high-heat alpha strikes pretty poor right off the bat.

The OPs suggestion isn't idiotic because he realizes that lasers deserve no nerfs. However, in the intrests of simplicity, the game could be more easily balanced by making simple, one-value changes like the ones above.


the problem to consider is the impending 12v12 incoming and the fact that mechs unlike tanks were created with the idea of endurance over firepower, it takes time to blow one of the bigger ones up. Even if brawling weapons got the buff they deserve a team of snipers all focusing fire will melt a mech in a single alpha valley... while it might take 2 if the big guns were on forced chain fire :D .

#35 Stat1cVoiD

    Member

  • PipPipPip
  • Bad Company
  • 83 posts

Posted 05 July 2013 - 05:31 AM

View PostHomeless Bill, on 04 July 2013 - 03:40 PM, said:

Spoiler



One thing I really love about MW:O is the fact that your aim is always steady and there is no randomness (besides not intended Hit-detection problems) involved. Hitting or not hitting is entirely skill-based and not a matter of luck.
That is not usual for the FPS genre and sets this game apart from other titles.
Being able to aim steady while moving f.e. adds a lot of dynamics to the fights and shouldn't be changed imho.
Your system implements "luck shots", that means even if the Mech is very hot he COULD still hit, but not because of the player's skills but because of a random algorithm.
Therefore i like my idea more... :blink:

View PostKmieciu, on 05 July 2013 - 03:29 AM, said:

Spoiler


TL;DR?! Despite my beautiful illustrations... :ph34r:

The rest of your post follows my chain of thought behind this very well. :D

As you said:
My suggestion is not about nerfing pinpoint-damage weapons into oblivion, it is about bringing them in line with the other weapons and the game's combat system in general.

If you raise the SRMs DMG to 2.5, one could think about adding a GCD to those aswell. Imho SRMs are indeed in need for a buff. 2.5 however, could be a little bit over the top, but regarding the fact that the horrendous splash damage is gone, it could work out. Just requires some testing and we would know... :o

View PostNik Reaper, on 05 July 2013 - 04:41 AM, said:

I would only argue that engines have an weapon energy output rating so that it reduces global cooldown for weapons and in some larger cases allow for 2 of a large type to be fired at once... 2 ppc together from an assault is not that unbecoming.


I don't think that this would be necessary. If GCDs would be implemented, people would adapt. By adding more and more special rules and exceptions, balancing might easily be broken again or the system might become too complicated...

Guys, it's only 0.5 secs... it's basically chainfire. It doesn't mean that you will never be able to fire 4 PPCs, you will just have to aim a bit better :P

Edited by Stat1cVoiD, 05 July 2013 - 05:33 AM.


#36 General Taskeen

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • 5,737 posts
  • LocationCircinus

Posted 05 July 2013 - 05:43 AM

But chainfire itself doesn't even work right. It just cycles through each weapon in turn, rather than firing the next available weapon (it unfortunately does not offer modes of fire selection).

So basically there would be two chain fires that are the same thing, unless if I am understanding this correctly if you fire multiple weapons, at the same time, they all fire eventually you just wouldn't be able to stop them all from firing in turn once you click that alpha button or your 'grouped' weapons.

Either way one or the other has the be fixed, especially chain fire, because its lame as is. And alpha-firing is definitely easy, it always has been in every Mech Warrior game, but can the devs think outside the box to resolve issues that have always been in MW games.

I'm not particularly for or against this idea, just neutral. I wouldn't mind seeing slight randomness to alpha-firing, where as single fire is pin-point. I see other issues with the game as well where whole equipment and weapons need serious overhaul. Alphaing and pin-point are also a 'thing' in MW:LL too, but after play testing that for so long, it is also relegated by proper heat balance.

Edited by General Taskeen, 05 July 2013 - 05:50 AM.


#37 Stat1cVoiD

    Member

  • PipPipPip
  • Bad Company
  • 83 posts

Posted 05 July 2013 - 06:33 AM

View PostGeneral Taskeen, on 05 July 2013 - 05:43 AM, said:

Spoiler


Yeah, I totally agree with you. Chainfiring different weapon systems together or firegroups with many weapons with a short CD doesn't work intelligent and needs some fixing.
To get a feeling how my system would be supposed to work, you should just enter a Mech with 3 or 4 PPCs, assign them to one weapon group and activate chainfire:
The Mech will continue to fire PPCs as long as you push down the button for that group.
That means if you would tap the "Alpha" button, just one GCD weapon would be fired. You would have to hold the "Alpha"-Button down to fire all your GCD-weapons, until the last one has been set off.

To manage GCDs with multiple different weapons, the game has either to chose smart (f.e. smallest CD first, as long as OFF COOLDOWN), or by row (if OFF COOLDOWN). If the game choses by row, the player needs to be able to switch the weapon's positions in the UI.

Example:

You have an AC/5, an AC/10 and 2 PPCs mounted on a CTF.

Scenario 1 (smart):
All weapons are assigned to G1. All CDs are ready.

You hold the button down:
-Fires AC/5, along with AC/10 (AC/5 GCD = 0 sec; AC/10 GCD = 0.25 sec)
:: 0.25 sec later ::
-Fires PPC #1 (PPC GCD = 0.5 sec)
:: 0.5 sec later ::
-Fires PPC # 2

-> All weapons fired. 0.75 sec passed. Nex weapon ready: AC/5 in 0.75 sec

Scenario 2 (smart):
All weapons are assigned to G1. AC/5 still on CD for the next 0.5 sec. PPC #1 still on CD for the next 1.75 sec. (hypothetical)

You hold the button down:
-- AC/5 on CD --
- Fires AC/10 (AC/10 GCD = 0.25 sec)
::0.25 sec later::
-- AC/5 on CD --
-- PPC #1 on CD --
- Fires PPC #2 (PPC GCD = 0.5 sec)
::0.5 sec later::
- Fires AC/5
::1 sec later::
- Fires PPC

-> all weapons fired. 1.75 sec passed. Next weapon ready: AC/5 in 0.5 sec.

Scenario 3 (by row):
All weapons are assigned to G1. All CDs are ready.

Assigned:
PPC
PPC
AC/10
AC/5
(hypothetical Player' choice)

You hold the button down:
- Fires PPC #1
::0.5 sec later::
- Fires PPC #2
::0.5 sec later::
- Fires AC/10
::0.25 sec later::
- Fires AC/5

-> all weapons fired. 1.25 sec passed. Next weapon ready: AC/5 in 1.5 sec.

Scenario 4 (by row):
All weapons are assigned to G1. PPC #1 still on CD for the next 1.5 sec. PPC #2 still on CD for the next 2 sec.

-- PPC #1 on CD --
-- PPC #2 on CD --
- Fires AC/10
::0.25 sec later::
-- PPC #1 on CD --
-- PPC #2 on CD --
-- AC/10 on CD --
- Fires AC/5
::1.25 sec later::
- Fires PPC #1
::0.5 sec later::
-- PPC #1 on CD --
- Fires PPC #2
(Firing the PPC now, delays the next AC/5 shot by 0.75 sec)

-> all weapons fired. 2 sec passed. Next weapon ready: AC/10 in 0.5 sec.

...................
::0.5 sec later::
-- PPC #1 on CD --
-- PPC #2 on CD --
- Fires AC/10
::0.25 sec later::
-- PPC #1 on CD --
-- PPC #2 on CD --
-- AC/10 on CD --
- Fires AC/5
....................


These are ways how the game would handle the GCD with different weapons and different CDs, resp. GCDs, in a single group.
Naturally, it would be much more efficient to assign different kinds of weapons to different fire-groups and by that manually chosing more efficiently,...
But you see how I think that could work out.

Edited by Stat1cVoiD, 05 July 2013 - 06:46 AM.


#38 Koniving

    Welcoming Committee

  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • The Guide
  • The Guide
  • 23,384 posts

Posted 05 July 2013 - 07:10 AM

View PostStat1cVoiD, on 04 July 2013 - 03:38 AM, said:

What do you guys think of it?
And @PGI: PLEASE TAKE A LOOK AT IT! IT WOULD MAKE THE GAME SO MUCH MORE FUN TO PLAY, TO WATCH AND IT WOULD RAISE THE SKILL-CAP FOR COMPETETIVE PLAY!


It starts great, as it's pretty true. Though the reason is mainly because the mechs could not handle the heat. For example in tabletop 6 ER PPCs generate 90 heat. Your threshold is 30. But you can fire those 6 ER PPCs in a turn from a custom made stalker with an XL 340 engine and x heatsinks, because it spreads the fire out over the turn (10 seconds) and even then it rides on 80% heat or more (24 to 26 out of 30 typically). Even so it comes with high risks such as falling over, pilot losing consciousness, massive damage from falling over, loss of limbs, etc. Sometimes they can never get back up. If you're lucky you just sit for a turn or only walk to avoid blowing yourself up.

In MWO, we had to get to 150% heat to receive any damage. 200% heat to start receiving any real damage. 300% heat to explode. When, do you ever, reach 300% heat?

Luckily this latest patch has made it so that 120% causes CT damage, 150% seems to kill you. However this only addresses the massive
Spoiler

spree people are on.

Next comes convergence. However since your goal was to handle alpha strikes, I think the job's partway done. To further it along, remove the raising threshold so we can't have heat thresholds of over 60. Last tested we had a threshold over 90 with 22 DHS. There's a reason for the huge gap between standard and double heatsinks.

Much simpler. If you want to Alpha, try it three times with 3 ER PPCs. I dare ya. Blew myself up without an override.

That said, I don't like the idea of a global cooldown. Without the rising heat threshold, we simply won't have the problem because your alpha limit for DHS would be the same for 10 Standard Heatsinks. Try it with 10 standard heatsinks and see what happens. Want to shoot once or twice and shut down, be my guest. I'll take you apart with my chainfired weapons.


It was simply a better time back then, when our heat threshold was so limited that alpha strikes came at such a high risk that everyone chainfires their weapons.

At the time, going over 110% was instant death. No gradual damage. Just....

Boom.

Edited by Koniving, 05 July 2013 - 07:15 AM.


#39 Fraegar

    Rookie

  • Ace Of Spades
  • Ace Of Spades
  • 6 posts

Posted 05 July 2013 - 07:18 AM

This is very similar to an idea I had. The main difference is the only weapons I think should be able to fire together are ones located in the same location of a mech, ignoring flamers and MG.

For example the HBK-4P can fire its 6 ML in its hunch all at once but the arm lasers would have to wait and be fired separately from each other with a fraction of a second wait.

This way only certain mechs and variants can actually take advantage of group fire.

Besides flamers and the MG I think every weapon and group should have a cooldown before it can be fired after another weapon. All weapons ranging from .25s to .5s where pulse lasers should be half that of normal lasers.


This to me feels like the best blend between battletech and mechwarrior of where in BT each weapon is always rolled for hit individually but you are allowed to fire multiple weapons at once in mechwarrior.

Edited by Fraegar, 05 July 2013 - 07:19 AM.


#40 Nik Reaper

    Member

  • PipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 1,273 posts

Posted 05 July 2013 - 07:34 AM

@Koniving

But how does this help the peeka-boo meta that we have and that is by all rights the optimal way to do battle? ( minimal exposure max burst as short as possible and back to cover )

If there is a way to unload all the damage at once without blowing up or getting shutdown from 1 alpha it will still be the best way to play, you will just have to cool down a bit longer , witch does give the target time to close on you , but here's the kicker, what happens when 9~12 mechs unload alpha on you at the same or near same time ( the time for all of them to climb the cover or clear the corner ) ?
Near instant death with out no real opportunity to do much (and we know that you only need 2-3 pinpoint 40~45 alphas to drop dead or be blood red CT and that only if you are in an atlas , the other 8~10 shooters are there just for the worst case scenario ;) )

Edited by Nik Reaper, 05 July 2013 - 07:35 AM.






6 user(s) are reading this topic

0 members, 6 guests, 0 anonymous users