339 replies to this topic

[Melissia]

I admit, I like the crosshairs swinging based off of movement, at least THAT is dependent on the player's sense of timing and aim, rather than luck. Piloting a 'mech should be a test of skill, not luck.

Edited by Melissia, 20 November 2011 - 04:06 AM.

[Kudzu]

J Echo, on 20 November 2011 - 01:55 AM, said:

Not only "might as well be," but is. Hoping that a random number generator decides in one's favor = hoping that one is lucky. Kudzu doesn't want put forth the effort to learn how to aim. For this reason, he'll say anything to try to make the game into a game of chance instead of a game of skill.

Oh, nice strawman! But yeah, I totally don't want to learn how to aim, that's why I've spent so much time playing (and being ranked in) FPS's like the old Infiltration mod for Unreal and America's army that use iron sights... Has it not occurred to you that perhaps I prefer aiming systems that stay true to the settings of the game?

In fact, it seems you are the one who doesn't want to "learn to aim" since having all shots perfectly land wherever the crosshairs point is about the most simpleminded system ever invented. There is no skill to be found in that-- you want to turn MW:O into this: http://www.freewebar.../deadeye-logun/

Quote

For example, it is an inarguable fact that the player has zero control over where the bullet goes within a random cone, not "full control" as Kudzu falsely claimed in his edit of that diagram.

If the cone was static and stayed the same no matter what a player did you'd have a point but, as usual, you are wrong. Since the size of the cone (thus the area your shots land in) is directly affected by your actions and choices you have FULL control over it. I also advocate for a centered weighted model of CoF-- which means that a shot is more likely land in the center, so even the cone itself isn't fully random.

[Creel]

Melissia, on 19 November 2011 - 08:58 PM, said:

Wat ain't no country I ever heard of.

Old movie references aside, I meant what I said. Raw skill obviously matters, and the clanners were all trained to very high standards-- but the inner sphere generally had better tactics, and so once they began adjusting their tactics and strategy to account for the clanners' skill and technology, they began winning battles. The battles were still hard fought, but then again, battles in game should be hard fought too.

Oh, I understand that... I'm just in awe of your agricultural expertise. If I ever i buy a cherry farm, I want you to train my employees.

[Creel]

J Echo, on 20 November 2011 - 01:55 AM, said:

Not only "might as well be," but is. Hoping that a random number generator decides in one's favor = hoping that one is lucky. Kudzu doesn't want put forth the effort to learn how to aim. For this reason, he'll say anything to try to make the game into a game of chance instead of a game of skill. For example, it is an inarguable fact that the player has zero control over where the bullet goes within a random cone, not "full control" as Kudzu falsely claimed in his edit of that diagram.

What, exactly is involved in "learning how to aim". With the previous systems, aiming was slightly more difficult than an old fashioned ADHD test, the instructions for which can be summed up as "click this button when the circle is in the right place". The idea that aiming in previous mechwarrior games has taken any significant amount of skill on the part of the player is a little silly.

[Creel]

J Echo, on 20 November 2011 - 11:47 AM, said:

Aiming was hard in Mechwarrior 3 and even harder in Mechwarrior 4. I constantly miss my shots, and I've been playing all sorts of games (from hardcore-realistic sims to strategy games) for my entire life. Leading the target is required for all non-laser weapons--and not only lead, but inverse-lead as well. Both of which change when you or your target change velocity (speed and/or direction). It is complex--you have to factor in [your speed] x [your direction] x [your enemy's speed] x [your enemy's direction] x [changes in any of these] x [your gimbal limits] x [any knocking from enemy gunfire] x (in Mechwarrior 4) [tipping of your mech from turning]. That's only eight of the factors, each of which interacts with every other for a total complexity of 64 possible interactions to adjust for. Again, that's the minimum. I'm leaving some out.

And while in Mechwarrior 3, the better-modelled mass of the mechs made for slower movement changes (and thus easier tracking), it still wasn't easy at medium and long ranges, and in Mechwarrior 4 it was as difficult as any F.P.S. (which is, of course, about what M4 was), because the mechs changed velocity so much more quickly.

The fact that new players do poorly at it, while players with a lot of experience at it do well, is irrefutable proof that the aiming in Mechwarrior games requires skill. You people who say that there's no skill in aiming either haven't tried it, or aren't good enough at it to realize how bad you are at it, or else aren't being honest about it because you have an agenda (e.g. wanting kills you don't have to work for).

I played 3 pretty extensively, and did very well in league play. I played 4 for a short time, but I didn't like the feel of it. I've never felt like the pulling the trigger when the crosshair was on target was all that challenging, nor should it be. It is, after all, pushing the button when the circle is in the right place. It's the other aspects of the game that make it stand out. Tactics, coordination, and teamwork should be what we focus on. I don't think that getting rid of auto-convergence will immediately make people think more about the other battlefield aspects, but I think that it will make these aspects more important. Especially with the dev's "4 pillars".

Autoconvergent weapons were put into the game, because there wasn't a better way to do it 20 years ago. Now there is, let's use it.

[Pht]

mithril coyote, on 17 November 2011 - 02:32 PM, said:

...but many have argued, the boardgame isn't a piloting sim, so a new approach has to be taken.

Does not follow - fallacious. No, the pen and paper isn't a first person piloting sim; but that doesn't require the conceptual approach to how capable 'Mechs are of hitting things with weapons fire be changed.

is there a point between these two styles that forms a happy medium? a style that retains the randomness of the boardgame but the simulator feel of the mechwarrior games?

Quote

in fact there is. the Battletech/mechwarrior novels, which are the medium that established the setting...

Um - no. The Boardgame showed up two years before the first novel; and the novels followed the boardgame in how things behaved.

And as far as using the novels - while they're rather consistent on the topic, they vary on the topic and are ambiguous at best.

Any attempting to get hard performance for 'Mechs and their weapons out of the novels is bound to be an ambiguous process with unpredictable results.

Quote

what is required to match the feel is randomized hit locations.

I don't think randomized is the right thing to go for; and that scares people away (and should!) - no, when you miss, it shouldn't be in a manner that makes no sense. When that happens you have no way of understanding why you missed and therefore you have no way to correct it!

Quote

in the novels, even if your aiming for a specific spot, your likely to hit anywhere on the target. even mechs with clan targeting computers find it difficult to obtain sniper type accuracy, instead merely obtaining lesser shot spread. in the novels, the closer you are the more likely you are ot hit where you are aiming, while firing at targets farther away results in a wider spread of damage over the target.

Actually, there are conditions in which a 'Mech can pretty well hit anything the pilot wants - shooting at an immobile target. This actually introduces a problem for the cones idea.

Quote

the "cone of fire" system, where your shots are randomized in an area around your reticle, matches this feel. by having this area larger against targets towards the end of the weapons range, and smaller against targets close to you,

If your misses are randomized like this than you can miss in ways that make no sense given how your target is behaving and how you are piloting your 'Mech.

As far as closer/further - a target that is very close to you can very well be missed, given certain (reasonable to happen) conditions.

Your target can be a VERY fast moving light 'Mech that is running in and out of cover; or your 'Mech can be overheated badly; or you could have taken massive damage... etc, etc.

The upshot is that to be able to make the cones system work in an intutive manner that doesn't scare people away you have to induce a lot of unnecessary complexity when all you need to do is model the performance of the 'Mechs and weapons from the already existing system. You pick up the performance/capability from the pen and paper - not the brute rules, but what they point to and illustrate - and put that into the game.

[wolf74]

With 11 theard talking about it. I see 4 basic groups of thought:

Cone of fire Type R (Random) (What most Video Gamer think)

Cone of fire Type E (Effected by mech move and being hit, bit of heat slowing correction down) (what a lot of other want)

Pin-Point Accuracy fire of weapon (Old MW4 way, the weapon always Converged under the crosshair)

Manual Control of the Convergence point +-5m to 15m (talked about very little) (Player picks a range point for his weapon to be converged at and on board systems adjust it +-5m to 15m to a Selected Target)
　



See the threads below for more information:
http://mwomercs.com/...on-convergence/
http://mwomercs.com/...of-dumb-codess/
http://mwomercs.com/...racy-indicator/
http://mwomercs.com/...-hit-locations/
http://mwomercs.com/...on-convergence/
http://mwomercs.com/...-balancing-cof/
http://mwomercs.com/...-medium-lasers/
http://mwomercs.com/...tch-the-setting (You Are Here)
http://mwomercs.com/...t-rng-nonsense/
　


to a lesser topics:
http://mwomercs.com/...-weapon-damage/
http://mwomercs.com/...really-matters/

Edited by wolf74, 20 November 2011 - 12:24 PM.

[Pht]

Wolf. Methinks you missed my posts simply because I haven't started a thread! ... I've been posting a different concept all over the place in these threads.

[Nik Van Rhijn]

You could also set your different ranged weapons groups to converge at different ranges, both horizontally & vertically.

[Pht]

J Echo, on 20 November 2011 - 12:29 PM, said:

I prefer the idea of in-mechlab convergence setting--meaning it cannot be adjusted during the mission. Like Second World War fighters--if you set your guns to converge at 300 yards, then you do the most damage when hitting targets at that range.

This is antithetical to how 'Mechs behave in the Lore and their capabilites in the Lore and doesn't, IMO, give any decent returns in gameplay/fun.

How tightly or loosely your hits cluster on your target should be controlled by what kind of weapons you're firing and how well your 'Mech can put them on the target you're painting for it given battlefield conditions.

[Melissia]

wolf74, on 20 November 2011 - 12:23 PM, said:

With 11 theard talking about it. I see 4 basic groups of thought:

Cone of fire Type R (Random) (What most Video Gamer think)

Cone of fire Type E (Effected by mech move and being hit, bit of heat slowing correction down) (what a lot of other want)

Pin-Point Accuracy fire of weapon (Old MW4 way, the weapon always Converged under the crosshair)

Manual Control of the Convergence point +-5m to 15m (talked about very little) (Player picks a range point for his weapon to be converged at and on board systems adjust it +-5m to 15m to a Selected Target)
　

You forgot this one:

Pin-Point Accuracy with reticle movement: Pinpoint accuracy, BUT 'mech movement, especially when moving fastter, throws off the reticle, as does being hit. This means you have to have better timing and fight the reticle to keep it on target in the middle of a battle, espeiclaly while moving (lighter 'mechs would probably have less of an issue firing while moving than heavy 'mechs, though). Bonus points for having the effect be amplified by heat. This is basically a sort of 'mech variation on the way recoil is done in many shooters.

Edited by Melissia, 20 November 2011 - 01:19 PM.

[Pht]

J Echo, on 20 November 2011 - 12:39 PM, said:

So?

What the hey, if it's ok to make arbitrary choices, why call it mech(anything)? ... just name it "generic mech game number (whatever)!

Or, better yet, let's pick and choose what we want and don't want based on nonsense, and expect to get a coherent game out of it!

Quote

And a cone of fire does give a return in gameplay/fun?

... ???

I'm not arguing *for* the cones.

Quote

First you tabletoppers complain that there's pinpoint accuracy, and then when people suggest reasonable methods of diffusing it, you shoot them down. Gargh.

... I know, let's introduce close-minded name calling into the discussion; that will really advance things

Besides which, I just posted a *more* reasonable method; in the post you've answered to here.

[mithril coyote]

since the wildly misleading 'comparison images have made a reappeance, i threw together some example images of what us cone of fire proponents actually mean.

the first thing to understand is that aiming will still occur as in previous MW games. you'll still have the reticle, you'll still have the reticle swaying around as you move, etc. those elements exist in the setting as defined by the novels.

what we're proposing is this: instead of hiting inside that central dot, the shots have a chance of hitting anywhere within the orange circle..which is not actually visable in play.

this is close range, the effective range of Small lasers and machineguns.



and this is the same distance, but the firing mech is moving. (only a 10% increase here)


medium distance, medium lasers, SRm's, AC20..

and moving (again 10% bigger)


Long distance - large lasers, AC10, AC5


and moving


very long distance - PPC's, LRM's, AC2


and moving



you'll note that at very long distance there is a small chance to miss entirely..this is consistant with the setting as described in the novels, where very long distance shots are lower accuracy than closer distance shots due to the lower precision.

you'll notice that this makes for a fairly minor change to the existing MW game aiming methods, the only effect being to spread damage out over the target slightly more.

this approach also makes the addition of later advanced technology systems like pulse lasers (which are supposed to be more accurate) and targeting computers (ditto) easier. neither of these have been well implimented in previous games, and under this system, they would shrink the scatter area by a few % each, making your shots more precise.

[dm5k]

Honestly I like the idea of a CoF for Mechwarrior Online. Mechwarrior 4 was just a FPS sniper fest and legging was frustrating. Whoever could pack enough firepower onto an assault mech and hit in the center torso first at long range wins. I think if CoF is done right then it would require actual skill to manage aiming. No more alpha strikes from across the damn map!

I believe that movement, heat level, and electronic jamming should effect aiming in some small way. Not to the point where your weapons completely miss where you're aiming. But where you want to be still or walking instead of running or jumping for better accuracy, or hide behind a building or hill to let your heat drop for a few seconds instead of continually firing with 100% accuracy in the red heat zone.

[Melissia]

Geeze, those are incredibly short ranges... it'd be almost impossible for a 'mech with short-ranged weapons to actually do much against other 'mechs with a range system like that. Even setting up an ambush the other 'mech would only have to take a few steps back from point blank range.

Your examples aren't that great (the "very long distance" is about where the very end of short and the very beginning medium should be), nevermind taht the cone of fire concept is really rather flawed to begin with...

Edited by Melissia, 20 November 2011 - 01:40 PM.

[Creel]

J Echo, on 20 November 2011 - 01:29 PM, said:

That's what the comparison images were showing to begin with. Nothing misleading about them--they were showing exactly what you showed. And it sucks. Not only is it ridiculous and unrealistic and illogical, but it would make for terrible gameplay.

I find your statement ridiculous, and unrealistic, and illogical, and it makes for a terrible argument.

Ridiculous, is completely subjective and not at all meaningful in context, but it is the first thing that you have managed to demonstrate conclusively. You obviously view it as inspiring ridicule. Not useful in the argument, but I can understand you wanting to score points where you can.

As for unrealistic, yeah, battle robots in outer space...

Illogical? Not at all, there've been piles of arguments on both sides, all of them can mostly manage to stay logically consistent; internally at least.

[Pht]

Creel, on 20 November 2011 - 01:46 PM, said:

Illogical? Not at all, there've been piles of arguments on both sides, all of them can mostly manage to stay logically consistent; internally at least.

it's the basic presuppositions for the various arguments that need to be brought to light and examined to see if they can support any thinking based on them.


On the cones: lets give an example.

Say, your mech is stationary, and you're shooting at a target that is considered "long range" for the weapons you're using. This target is moving laterally - left to right across your field of view, and it is doing so very, very quickly.

You're piloting, lets say, a black hawk (lots of identical lasers). You shoot at the target, but due to the COF concept, instead of missing behind the fast moving target - your lasers go ... randomly all over the place.

The same problem crops up on what hits - they give you no idea of why you hit by where they hit.

These factors will be critical for newcomers to the BT lore in understanding how a 'Mech behaves and performs.

These problems are entirely possible in a bare COF system!

So, what do you do to fix this? ... well, you have to change the shape of the cone to make the missed shots make sense: you have to apply that re-shaped cone to the target offset in the direction that makes sense for how both the firing and target 'Mechs are behaving and the environment.

if you give all the weapons the same aggregate cone, than your weapons that are more accurate lose all advantage - so you wind up at least having to put all similarly performing weapons into their own cones, and those cones will have to change shape separately.

But ... what about weapons that are just slightly damaged? ... well, than you have to have a cone for each and every weapon.

To make things worse, you have to change the volume of the cone, and not by range, but by the firing 'Mechs condition - when your 'Mech is hot, the myomers and other systems cannot bring the weapons to bear as they should, so you wind up having to also change the voume of the cone. Larger cones for worse targeting conditions and smaller ones for better conditions.

Than there's the issue of imitating shooting at immobile targets that are smaller than your cones at any given range; and a 'mech pretty well can choose what it wants to shoot on an immobile target at most ranges; so you have to know when to shrink those cones to almost nothing in volume.


It's an over-complicated concept that has nothing over looking at the rules in the pen and paper and using those as a baseline for the performance of the weapons and the performance of the 'Mechs in using them.

Edited by Pht, 20 November 2011 - 02:03 PM.

[dm5k]

Those cones are a bit extreme from what I imagined CoF being. I think your mech should be fairly accurate but range, movement, and heat should play a role in accuracy though.

Edited by dm5k, 20 November 2011 - 02:05 PM.

[Miles Tails Prower]

The more control you take a way from the player, the more frustrating the game becomes. Pin point accuracy is only the attacker's part of the equation, the defender should be maneuvering and rotating their torso to avoid and "shoulder" damage aimed at their center torso.

If more variance in the skill for each weapon is what's needed, the add in gravity. Missiles and cannon shells arc as they travel to more distant targets while lasers do not, requiring more learned skills for the behaviors of each weapon. Just like how an FPS has weapons of varying kicks, RPMs and iron sights.

[Pht]

Miles Tails Prower, on 20 November 2011 - 02:04 PM, said:

The more control you take a way from the player, the more frustrating the game becomes. Pin point accuracy is only the attacker's part of the equation, the defender should be maneuvering and rotating their torso to avoid and "shoulder" damage aimed at their center torso.

This is really one of the core problems that needs to be resolved well!

The players, I think, by far and large, don't know what it is that a MechWarrior controls; they don't know that the MW controls what is being aimed at directly, but they only indirectly control how well their 'Mech can hit it's target.

Putting the reticule on what you want is the direct part; but the rest of it - how cool is your 'Mech running, how fast your target is moving across your field of view, what weapons do you choose to fire at what ranges and conditions... do you take your time, brace your arm with a weapon you want to fire on a structure and allow your 'Mech to really chew on getting a good firing fix ... these are how a 'Mech pilot "controls" how well or badly their 'Mech can hit what they're targeting.

Edited by Pht, 20 November 2011 - 02:25 PM.