Pilot skill or probabilistic hit locations?
#221
Posted 12 March 2012 - 03:55 AM
#222
Posted 12 March 2012 - 04:24 PM
To do it right it needs to be indipendant of movement, the cone itself is already simulating movement just by being there.
The cone should only expand of there is no target being lit up by the crosshairs.
And should only contract if a target is being lit up by the crosshairs.
It should contract quickly, around 1-1.5 seconds at most, All the CoF will really be there for is to make snap shooting a little moar random so lighter faster mechs last a little longer in combat.
The fully expanded CoF should be about 15-25% bigger than the outside area of the torso crosshairs in the interview video.
Finally the fully contracted CoF should be about the size of the arm's circle crosshair in the interview video.
What this would mean is the snap shots you take would be a little out of sync with where your aiming, but after a second you will be able to pick where you want to aim, this would really only help aginst alfa-striking and legging.
#223
Posted 12 March 2012 - 10:55 PM
Otherwise, it should be some sort of probablity equation ...
#224
Posted 13 March 2012 - 07:53 AM
#225
Posted 13 March 2012 - 08:50 AM
mekabuser, on 11 March 2012 - 06:55 PM, said:
gravity , of course. ballistic drop....
By "environment" I meant all other battlefield conditions outside of the firing mech itself; that could be heat, insane sensor interference, etc. It's a pretty large bit of stuff.
Quote
Um, did you miss that I was making reference to what he posted?
Belisarius†, on 11 March 2012 - 07:23 PM, said:
COF is not an attempt to take into account all factors; it is an attempt to account for how well a 'Mech can handle it's weapons... while I don't like the COF idea, it is not a "cop out."
Quote
The most insanely, ludicrously capable pilots in the BTUniverse (the bounty hunter, kai allard liao, morgan kell, etc) are not capable of making all of their weapons hit the exact same point on a target under all conditions... in fact, under pretty much any other condition than shooting at an immobile target, and here's why...
It is the 'mech thant actually has to aim the weapons and handle the calcuations, and while getting a single weapont to hit a single point is easy in BT and for us, getting multiple weapons to hit a single point is not, in BT or in real life. That is simply the way the BTU is... it is not a japanese mecha game with mechs capable of insane shooting skills.
Quote
No pilot can be any better at gunnery than his mech is ultimatley capable of.
As for the RNG phobia ... it's just that, an irrational fear:
Quote
Speaking of RNG, where the TT gives these they can be expressed as probabilities, and can otherwise be mathematically mapped out to provide raw hard data on the various capabilities of BattleMechs and/or the various weapons they mount. The RNG mechanic need not be used, but the data that it expresses is invaluable and irreplaceable and that data can be expressed in video game rules that do not take anything, skills or otherwise, away from the player that the player should be handling in a MW video game... nor does it necessarily render game play mysterious and nonsensical and player skill pointless. It all depends on how the Baseline Data from the RNG is handled.
Link
The only places a MW video game would be using anything even like a RNG would be in simulating how capable a 'mech is at handling it's weapons or environment... NOT for gunnery or piloting skill rolls, which an MW video game would not even have.
pesco, on 11 March 2012 - 08:00 PM, said:
I don't know; my computer won't play mwll; but all online games based on perception time/reflexive reaction time suffer from that effect to some varying extent.
If your computer is only rendering at 32 frames per second and you're at a 300-500ms ping and playing on an otherwise fast server, the people rendering at 60+fps with 50-80ms pings are going to pound you into the dirt. It's thenature of the beast in twitch games.
Quote
EDIT: NM, I missed your point.
If your target is immobile, sure, it's easier, and the rules reflect that with an easier to hit roll.
If your target is closer but mobile, it's *harder to hit!*
No, seriously. The closer your target is, if it's moving, it's velocity across your FOV goes up a LOT.
Which is easier to track, a light mech going 110 kph laterally at 500 meters or at 10 meters?
Quote
_______________
Back to the big picture: If a Mech is supposed to be able to hit anything at all at max range of its long range weapons, any imprecision in the weapon must be very near negligable at shorter ranges.
Complete, total, and utter straw man. The point I was making:
Quote
Link
In other words, the data is there, expressed by the rules, for how capable a 'mech is of handling it's weapons under battlefield conditions. We don't need to guess at it or make arbitrary systems with God only knows what kind of unitended consequences. It's a known factor.
Besides which, to address the red herring of aiming for the head and getting the toe ... you wouldn't hit the target's toe. The shots would concentrate across the top of the target mech, and the hit tables + called shot already represent that.
___________
You appear to be confusing two different factors with each other.
Convergence has nothing to do with firing any single weapon; it has to do with how well a 'mech can use *multiple* weapons to hit a single point.
Accuracy: hitting a point with one weapon. Precision: hitting a point with multiple weapons. Not the same thing.
Edited by Pht, 13 March 2012 - 10:00 AM.
#226
Posted 13 March 2012 - 09:18 AM
Take a Pen Light, or 2, and place them under your arm pits. Now jog, trot or run down a hall way, torso twisted to face the wall. Now watch the pen lights track.
Now don't forget, the machine you just Torso mounted those Laser weapons on has the BEST Gyro system in the known universe, along with a damn fine Balance module. How's the Tracking?
Now place the Lasers in your hands and run it again. More stable when mounted in the arms? Better Tracking. Same Gyro, same stabilizing system in place remember.
Now take your machine outside and find a wall that can be fired upon that has an earthen (grass surface) on which to jog, trot or run across. How did that work out, Armpits and then hands?
Good Luck!
Edited by MaddMaxx, 13 March 2012 - 09:21 AM.
#227
Posted 13 March 2012 - 09:38 AM
MaddMaxx, on 13 March 2012 - 09:18 AM, said:
Take a Pen Light, or 2, and place them under your arm pits. Now jog, trot or run down a hall way, torso twisted to face the wall. Now watch the pen lights track.
Now don't forget, the machine you just Torso mounted those Laser weapons on has the BEST Gyro system in the known universe, along with a damn fine Balance module. How's the Tracking?
Now place the Lasers in your hands and run it again. More stable when mounted in the arms? Better Tracking. Same Gyro, same stabilizing system in place remember.
Dude, seriously? How about you make that analogy somewhat useful to the discussion by introducing necessary additional facts into the equation first? For starters... are we talking about shaved or unshaved armpits? If you leave totally out the potential detrimental effects that pubic hair could have on lasers and what not....
(j/k)
Now more seriously again... what I find personally most interesting in this whole thread so far is how close we can/could get to find some common ground here and where. As most won't argue seriously it should be 100% skill-dependent or 100% randomized. And you can bet your bottom dollar the devs will keep an eye on this thread, as the exact ratio of skill/randomization touches the very core of gameplay, and thus community feedback is like free market research.
It also seems to have somewhat evolved into a sideshow "theater of war" for the FPS vs. Simulation conflict. Which IMHO shouldn't detract us too far from the balancing/technical issues, though. After all it would be more productive to find some "common ground" where both sides can agree upon stuff.
Edited by Dlardrageth, 13 March 2012 - 06:29 PM.
#228
Posted 13 March 2012 - 09:39 AM
Belisarius†, on 11 March 2012 - 04:02 PM, said:
Sounds like someone might be bitter because he got killed by snipers all the time.
Convergence is a nice in between, and it adds an extra element to gameplay that's nonetheless controllable to a good pilot. If you're intending to take a shot at an enemy from cover, I can see players lining up a random piece of terrain to set convergence at the right point before exposing to take their shot. I kind of like that.
The one good argument for cone of fire I've seen is with respect to lights. Every 'Mech game thus far has been crippled by lag, but lag also gives forcefields to small, fast targets at range. Abusing that forcefield was an essential part of being a scout.
I'm a little worried that in a game with good netcode and accurate weapons, scouts are stripped of their forcefield and might be too easy to cripple. That's a legitimate objection, as it could remove a major tactical and strategic element. "I don't like snipers", in my mind, is a lot less legitimate.
Of course I hated getting potted by snipers [note: in MPBT 3025, shots landed exactly where the crosshair was pointing...regardless of whether you moved into cover while the shot was 'in flight'] but that isn't the point that I am making.
In reality, the precision needed to make a head hit on moving target at range is 'godamned hard' (I know - I have held an expert rating for about a decade - and those are *stationary* targets). When you have 'infinite zoom' and 'lasers' there is *no* reason you couldn't hit any target from here to the horizon, exactly where you aimed. (IF you had perfect alignment...).
While that might be realisitic, it might not be a whole lot of fun.
The 'donut of death' or 'shot cone' is an attempt to use probability to make the end result look alot more like a 'real' distribution of shooting. The problem with a shot-cone is that it essentially limits players to only ever getting as good as the probability curve allows. What this means is that the dispersion of the shot cone is then added to the players 'normal' dispersion. Bleh.
I think there would be several good ways to limit the effectiveness of the alpha strike and 'long range fire' without resorting to a probability cone.
1st - limit the amount of 'with crosshair' zoom - don't allow people to zoom in so much that picking out a location on the target mech is super easy. Find a balance between too hard and too easy
2nd - don't eliminate 'motion effects' while zoomed in. Make the crosshair jiggle with stepping *exactly* the same amount as when zoomed out (why would the crosshair bounce less just because you hit 'zoom'). This makes it virtually impossible to deliver precision long range DIRECT fire while moving. LRMs would still probably work pretty good, though
3rd - model the ballistics of each weapon differently. AC20 fly slower than AC2 etc which are slower than PPCs which are slower than instant-lasers. This helps with the 'alpha strike' issue at medium and long ranges.
4th - model lasers as damage over time, rather than delivering all damage 'instantly' on hit - pretend that for every 1/10th second the target location is hit continuously, the damage increases (this is backwards from how real penetrating lasers would work...but meh). Then, make each level of laser have a longer pulse duration. IE, small laser firing time is .3 seconds, ml firing time is .6 seconds, and LL is 1 second.
5th - model convergence, either make it automatic but don't give the player any indication of 'how converged' his weapons are. IE, make it take some small amount of time for the weapons to align from 1000m to 50m - without telling the player 'ding, converged', OR allow the player to set 'convergence range' for specified weapons groupings. OR set convergence to match the distance of the currently selected target. Either way, that makes it harder to 'snap' onto a close scout mech and deliver an alpha strike to the same hit location.
6th - include a *very* small dispersion amount - I would suggest around 0.25deg maximum deviation. This is to make long range precision shots 'hard' but not impossible - as an example, if you were shooting at a mech cockpit 1km away (cockpit is say 3 m high) assuming a 3m long barrel (rather large for a laser) the tip of the barrel can move less than 5mm, or about the thickness of a pencil.
Edited by docmorningstar, 13 March 2012 - 09:55 AM.
#229
Posted 13 March 2012 - 01:48 PM
#230
Posted 13 March 2012 - 06:20 PM
Diana, on 13 March 2012 - 01:48 PM, said:
thank god right? there would be nothing more ludicrous than having your difficult deflection shot negated by probability .
Unfortunately , there are several members in this thread who just love to argue , and will continue to do so , never stopping.
Most who are for probability have more than likely not played more than a few rounds of mwll, if any at all, so Have no concept of how a modern mw games weapons can be modeled.
its not pinpoint.
it has different travel times, and ballistic drop for slugs.
it has different travel times to ppc and gauss.
it has duration lazers.
It has zoom that makes it difficult to hit anything but assaults where you want .When theyre not moving.
All of which makes hitting your target not easy, not to mention pinpoint. But no, they dont see that.. They talk about what mechs can do and how they really work and others talk about targeting computers, and cof and it just winds up being (something that I have no interest in). No matter how many times (in my opinion) they get proven wrong , or (my idea of) logic is shoved in their faces, they just keep at it because frankly they have nothing better to do and are
Looking forward to destroying you on the bf.
Edited by Mason Grimm, 13 March 2012 - 09:59 PM.
I came, I saw, I edited.
#231
Posted 14 March 2012 - 03:27 AM
I don't *know* how LL plays, but I do know how MPBT:3025 played and how MW4 played out-of-the-box. In 3025 you would see the AC/2 blackjack and vulcans *dominate* the battlefield simply because they had the longest range. In MW4 you used to see ERLL or LRM or GR 'boats' because combat effectiveness (at that time) was max damage X max range. This wasn't 'fun' for me.
If I wanted to play silly buggers warp speed combat, I would play COD or something.
I always felt that battletech/mechwarrior universe could be the thinking mans combat game. I just want the game to be fun, I want battles to be 'epic'. That is all I care about. Don't be a dick about it, man.
I am talking about concepts, and potential ways to get to those concepts. All you are doing is talking about your particular implementation. If you get the concepts right, then you can implement things in a variety of ways. If you screw up concepts, well, then your implementation is *always* going to be chasing good behavior.
#232
Posted 14 March 2012 - 04:34 AM
mekabuser, on 13 March 2012 - 06:20 PM, said:
That was somewhat hostile.
I don't know about you, but when I fire a cannon and it DOESN'T go where I'm aiming, I get angry. I'm the ********* pilot, I'm aiming the cannon. Not some two bit loaded dice.
(For reference, I am bias because I have bad dice rolls. All the time. nine 1s from ten D6s. And it flipping INVERTS for games where low dice rolls are desireable)
If I can see that shot, and I can make that shot by my aim, I should damn well be able to make that bloody shot. And, in addition, your 'ballistic' arguement all falls apart when you talk about lasers. They DON'T deviate. They will HIT, with a 100% chance, whatever they are pointed at. If I'm pointing at a target with a laser, and it misses for no reason, then the game has failed.
That is all.
#233
Posted 14 March 2012 - 04:46 AM
http://en.wikipedia....wiki/Contrarian
@ snow dragon. I wasnt being sarcastic. I was saying thank god they are not using probability. I said there would be nothing more silly than having your difficult deflection shot, that was going to hit<miss> because of some random generator .
Second, lazers with duration will hit where you point yes, but if the target is moving laterally it is much more difficult to maintain focus on one spot. Very difficult..
This will not be MW4 instant damage lazers/../It is another whole animal.
A proper heat mechanic will also balance out any boating people might like to engage in
Once again, something else MWLL has done damn near perfect.
#234
Posted 14 March 2012 - 05:11 AM
There are many variables in my opinion should be accounted for. Probabilistic may help a lot.
First of all I wa thinking about "hacking avoidance". On another thread has been discussed how the Hit location should be registered; wether Server side or Client side. It opened the chance of hacking a Client side aspect.
However, if fire is Probabilistic, then you cannot Hack on Client side something will be calculated as a probability on Server side (I hope).
I like the concept your reticule becoming larger or smaller related to movement, heat and other aspects impacting on your "machine reliability". Some Ppl said "I wanna hit where I aim", true, but your not using a rifle on your shoulder, where your body is the only responsible of your actions. You are using a Machine with tons of Electronic that need to compensate and offer you the "best chance" to put your shot "where you want". This also involves that using electronic, gyros and so forth as an important component and allocating "Tons" for them, will make sense to improve your hitting chances. You may choose to give those 5 Tons to electronic and no chance to put an extra Laser or an extra Missile rack or ammo. It make sense to me
Also i think Alpha Strikes , Chain Fire and Group fire should be differently effected by that. A single weapon fired in Chain should be way more precise and reliable than a Group Fire. Alpha should be seriously unprecise. Electronics should account for different weapons, behaviors and so forth, offering a freaking large reticule I really would love that !!
#235
Posted 14 March 2012 - 05:46 AM
For me it's 50/50 Pilot skill and probabilistic.
The former helps you negate the latter, but never eliminate it.
#236
Posted 16 March 2012 - 03:55 AM
Pht, on 13 March 2012 - 08:50 AM, said:
No, seriously. The closer your target is, if it's moving, it's velocity across your FOV goes up a LOT.
Oh, and here I thought you were worried about it being too easy for people to hit things! Now you're explaining that it's hard. ^^
Quote
Sorry, let me adjust my original remark. Say I'm at point blank range to that Atlas and manage to aim for its head, pulling the trigger in just the right moment. Would you like a die to be rolled in the background and the shot go to the middle left torso instead?!
Just to answer to the accusation of putting out straw men and red herrings: The point I was trying to get to was this: The player is going to notice you're rolling dice in the background and the shots "go where they want". There would be situations where this becomes very visible. I'd like the video game to not do that. And incidentally, I'm pretty sure it won't.
For the record: I wouldn't be strongly opposed to small imprecisions in the weapons, aka "cone of fire". But as I wrote earlier, I'm not terribly convinced we even need them. I'll let the developers duke it out with their ER Large Lasers for a while and decide whether it's easier to hit people than it should be.
#237
Posted 16 March 2012 - 05:08 AM
mekabuser, on 14 March 2012 - 04:46 AM, said:
Seriously now? LMAO... okay then...
Contrarian
Didn't know you were talking about real estate purchases or something else business-related. Looking at the proper definition and its source in Cambridge dictionary in fact... didn't know Mechs were a stock market commodity by now already. I humbly suggest if you want to contradict the moderation you do so with a proper source, not Wikipedia, like i.e. a dictionary, and try to do so in a PM probably.
And back to the topic, and probability and Wikipedia... just for general info :
Law of probability
Edited by Dlardrageth, 16 March 2012 - 05:11 AM.
#238
Posted 16 March 2012 - 05:37 AM
Lets remain cordial....everyone.
#239
Posted 16 March 2012 - 06:02 AM
pesco, on 16 March 2012 - 03:55 AM, said:
This, I think, is the biggest factor that is driving the 'skill' crowd. Most implementations of Cone of Fire tend to ... suck (see: World of Tanks). That breaks immersion (on an aside, I have a screenshot of a shell departing my 152mm derp cannon at 45deg in WoT - it is infuriating). That doesn't mean that CoF is a bad idea at all, it just means that badly implemented it can be an immersion breaker.
People aren't necessarily upset that your shot hits +/- 5in - they are upset when your targetting reticle completely covers the target, and you *still* miss. Or in MWO - you aim point blank at an atlas's head, and hit him in the foot.
To me it seems obvious that we want long range accurate shooting to be difficult to accomplish. I think that we have alot of tools in the box to accomplish this, and I think that limited CoF is one of them.
The TT rules kinda-sorta try to model this with different ranges for weapons. I have always thought that 'short' range in TT represents a 'high' confidence in accuracy, 'med' represents so-so confidence, long represents low confidence, and out-of-range represents *no* confidence.
It isn't that an AC20 shell won't fly 20 hexes, it is that you have slim-to-no odds of hitting a target at that range, because the inherent accuracy of the weapon precludes that. Lasers, you can think of differently - you can assume that the different quality of laser represents a higher degree of focal accuracy - or some such (else, why would large vs small have different shooting distances....)
Finally - the fact that weapons can shoot 'beyond' long range makes a CoF even more important. Without a CoF (and a different CoF for *each* weapon type) people will be *able* to take and make super long range shots with things like AC20s.
#240
Posted 17 March 2012 - 01:54 AM
docmorningstar, on 13 March 2012 - 09:39 AM, said:
Of course I hated getting potted by snipers [note: in MPBT 3025, shots landed exactly where the crosshair was pointing...regardless of whether you moved into cover while the shot was 'in flight'] but that isn't the point that I am making.
In reality, the precision needed to make a head hit on moving target at range is 'godamned hard' (I know - I have held an expert rating for about a decade - and those are *stationary* targets). When you have 'infinite zoom' and 'lasers' there is *no* reason you couldn't hit any target from here to the horizon, exactly where you aimed. (IF you had perfect alignment...).
While that might be realisitic, it might not be a whole lot of fun.
The 'donut of death' or 'shot cone' is an attempt to use probability to make the end result look alot more like a 'real' distribution of shooting. The problem with a shot-cone is that it essentially limits players to only ever getting as good as the probability curve allows. What this means is that the dispersion of the shot cone is then added to the players 'normal' dispersion. Bleh.
Just for the record, I agreed with everything in this post. We were on exactly the same page, so I'm sure why we were arguing. Your points (with the possible exception of #6) are exactly how I want the game to be modelled. I think there's more than enough variability in that to make cone of fire uneccessary.
I do think trying to put realistic explanations to the TT rules is a futile exercise that has put many good men in mental hospitals. The fact is that the game was designed to be fun in a tabletop setting, by people interested in having only the vaguest coincidence with high school physics. ACs are a good example; in reality, heavy calibre shells go much, much further than light ones, unless you physically attempt to throw them. MWO should worry about what's believable in MWO.
Edited by Belisarius†, 17 March 2012 - 02:01 AM.
2 user(s) are reading this topic
0 members, 2 guests, 0 anonymous users