Balancing The Alpha Strike With A Reactive Reticle
#101
Posted 27 June 2013 - 01:10 PM
However as I understand it, removing convergence completely as the original post suggests will turn Assaults into damage sponge titans since no one will be able to accurately hit them anymore unless they fire one weapon at a time. At the same time to kill a Light mech you just fire an Alpha Strike or two and enough weapons will hit to strip most armor and weapons since instead of the pinpoint aiming point your weapons will fire through a large (large enough) window guaranteeing a hit that would have normally missed. Please don't just dismiss this because it will likely require a complete rebalancing of all the mechs. It will be one of the results of changing group fired weapons into a narrow cone.
#102
Posted 27 June 2013 - 01:21 PM
And, because the convergence speed is faster against slower moving targets, assaults will actually have enemy weapons converge faster on them.
The fact that a light 'Mech moving fast slowing convergence of weapons fired against it will also mean that the group weapons fired at fast 'Mechs will either be spread out, with some weapons in certain locations missing completely, negating the chance of one shot kills.
Perhaps a comprehension failure on your part.
Edited by DocBach, 27 June 2013 - 02:09 PM.
#103
Posted 27 June 2013 - 02:13 PM
#104
Posted 27 June 2013 - 02:15 PM
Don't we have something like that already with multiple reticles? I have never understood why the arm reticle would ever trail the torso reticle, but it does by half a second to a second, and gets worse when moving over uneven terrain. So we are talking longer for the anti-alpha fix, I guess.
And you want to penalize long range and shooting from the hip, and buff brawling ranges. I am not sure anyone wants to go back to huggy-mechs from closed beta.
Anyway, you are talking about removing alot of the dynamism of the game to save players from accuracy. I just think there are better ways to save players from an over-rapid destruction of their mechs. Ways that do not complicate Mech operation more than it already is. You know many players run from MWO as soon as they see the two bouncing reticles. They never even learn to torso-twist, or well, understand what is happening when they do.
#105
Posted 27 June 2013 - 02:15 PM
Easy
#106
Posted 27 June 2013 - 02:15 PM
#107
Posted 27 June 2013 - 02:23 PM
KAT Ayanami, on 27 June 2013 - 02:15 PM, said:
Easy
so how much heat would dual gauss boats put out? How about SRM boats? Wouldn't it be strange to have a weapon that does 1 heat when fired by itself all of a sudden shoot up to 30 heat because it fired a second one?
Edited by DocBach, 27 June 2013 - 03:48 PM.
#108
Posted 27 June 2013 - 02:43 PM
You all need to consider the downside of forced fuzzy aim more. Do you really want to have your butts shot off while your weapons gyrate helplessly? That's what would happen in most of your defeats. You'll be sitting there forced to plink away at some positioned mech who has you zeroed in, unable to do anything about it. You take for granted that you can turn that battle now.
One of the hard things about balancing is that to kill one problem you can often create many more uglier ones.
Good Luck and see you in the game.
#109
Posted 27 June 2013 - 02:48 PM
#110
Posted 27 June 2013 - 02:52 PM
Lightfoot, on 27 June 2013 - 02:43 PM, said:
And that's the mentality I can't understand. Forced? Do you suddenly get cemented onto the ground when you're shot?
The correct response to 'I'm being shot by someone I can't see' is not 'Stand out there and get shot some more while trying to figure out where he is'.
This is why we have such bitter arguments on these forums. It will be impossible to come to any sort of agreement when people have such divergent views on how forgiving the game should be.
Snipers may be useful in killing things. But their real value is is pinning down an enemy force who is either forced to dig them out, or flank around.
IMHO, its ridiculous that people think its fair or reasonable that this is too boring or that the game needs to allow stupid decisions so we can run up to snipers and punch them in the face.
Edited by Valore, 27 June 2013 - 02:55 PM.
#111
Posted 27 June 2013 - 03:45 PM
Lightfoot, on 27 June 2013 - 01:10 PM, said:
However as I understand it, removing convergence completely as the original post suggests will turn Assaults into damage sponge titans since no one will be able to accurately hit them anymore unless they fire one weapon at a time. At the same time to kill a Light mech you just fire an Alpha Strike or two and enough weapons will hit to strip most armor and weapons since instead of the pinpoint aiming point your weapons will fire through a large (large enough) window guaranteeing a hit that would have normally missed. Please don't just dismiss this because it will likely require a complete rebalancing of all the mechs. It will be one of the results of changing group fired weapons into a narrow cone.
That is partially true. It makes all mechs damage sponges.
Which is as it should be.
The old battletech lore is a slugfest, not a twitchfest. There is a significant difference between the two - and that is the surviveability of the mechs. Large mechs have the armor to withstand, while a lighter can dodge, those in the middle can do a combination depending on the mech.
But it doesn't mean your fears are either unjustified nor is it really true. An Assault still stands out like hell - it gets hit, and as it is now a fast Light is the best there is to dodge and can do it well, but its still as flimsy as can be. The deviation of aiming won't be as drastic as I think you fear either. We are talking about cutting off torso convergence and you are worried you still can't hit an Atlas. Think of the size, even with no convergence something like an Assault still will receive most of the hits in one part. Would be no different than my Catapults - the CT and Arms are targets you can't miss by a meter or so. And for a light? You might be surprised that deviation might mean another hit you would have normally missed.
Also, you argue the Highlander's variant firing angle? What would be so hard about it being any different than now? You have the CT aim, the "good arm" aim and the invisible "off arm" aim. Wow, sounds alot like what you explained.
So I don't see any issue in this adjustment, it shows promise that it can work for the SPECIFIC PINPOINT removal, and instead give a bit more coverage which should work. Some tweaking obviously needed, but there's nothing to say it can't be done so far.
#112
Posted 27 June 2013 - 04:35 PM
DocBach, on 27 June 2013 - 02:48 PM, said:
I've got a similar suggestion running here - give it a look and tell me what you think.
#113
Posted 27 June 2013 - 04:48 PM
#114
Posted 27 June 2013 - 06:53 PM
DocBach, on 25 June 2013 - 07:27 AM, said:
Homeless Bill's system was a good idea, but it was incredibly complicated and over complex
This proposal fully fleshed-out, it's comprehensive, it doesn't leave any messes to clean up, it doesn't force massive weapon-rebalancing, and it think it would be really cool.
The way I look at it, your solution is perfect for the former mindset, while mine is perfect for the latter. Yours has a large effect on the style and pacing of combat, pushing it towards a true Battletech simulation; mine requires very little adjustment by the player and won't affect combat save for the elimination of extreme, pinpoint damage.
Honestly, I'm okay with either direction; I just want this **** fixed.
The one point I take issue with is the supposed complexity of my proposal. For all its detail, numbers, and overly-thorough explanations, the impact on the player is truly minimal: set up your weapon groups intelligently and slightly stagger fire for accuracy or alpha strike away to pour on inaccurate pain. And the complexity (or lack thereof) of implementation on PGI's side was covered in the full article.
#115
Posted 27 June 2013 - 07:18 PM
#116
Posted 28 June 2013 - 04:52 PM
The main problem for the system, however, is that it is very, very, very-very noobie-unfriendly
#117
Posted 28 June 2013 - 05:38 PM
#118
Posted 29 June 2013 - 07:31 PM
#119
Posted 29 June 2013 - 11:54 PM
#120
Posted 30 June 2013 - 09:03 PM
I don't think torso weapons ought to converge at all when group fired (unless they are lasers - maybe).
giving them a preset convergeance would be ok though - something like 200 meters?
Edited by Thorasta, 30 June 2013 - 09:31 PM.
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users