Jump to content

Targeting without expanding reticules of dumb CODess


125 replies to this topic

#81 UncleKulikov

    Member

  • PipPipPipPipPipPipPip
  • Bridesmaid
  • 752 posts

Posted 12 November 2011 - 06:17 PM

View PostPht, on 12 November 2011 - 06:11 PM, said:



Look to the tabletop for performance profiles of how well 'Mechs can concentrate their fire and how well the differing weapons types can hit things.

Take those parameters and put them into the game - have it calculate hit/miss and what is hit when you hit for each weapon on a 'Mech that fires based on those performance factors.

IE: At the extreme range of the normal range rating of a weapon, you stand more of a chance to miss - and if you are overheated and shooting at a fast moving target in cover, you're lucky if you hit.

No need for cones.

And as far as "Expanding reticules?" .... yeah, they're silly.

Do it like the lore does it. Color code the reticule for lock quality and use audibles. Changing the reticule shape or making it dance or having multiple pips for multiple weapons ... yech.

That would suck, relegating my personal accuracy to a statistical program. The tabletop game doesn't account for rate of fire affecting accuracy, or plenty of other interesting factors.

A expanding cone of fire makes a lot of sense, considering the tabletop rules for movement affecting accuracy.

#82 Pht

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,299 posts

Posted 12 November 2011 - 06:26 PM

View PostUncleKulikov, on 12 November 2011 - 06:17 PM, said:

That would suck, relegating my personal accuracy to a statistical program.


Mechwarrior is about making a game that puts us in the cockpit of a battlemech, as they exist in the battletech universe.

The pilot does not aim the weapons for the 'Mech - the pilot designates the target and the 'Mech tries to hit what the pilot is designating.

"gunnery skill" in the BT universe is more about putting your 'Mech in the best condition for it to be able to bring it's weapons to bear on the target you're indicating - things like taking into account your heat levels, damage to your 'Mech, the behavior of your target, what range it is at in relation to how well your 'Mech can hit with those types of weapons at that range.

"personal aiming" where you directly aim your weapons is what FPS's are built for - MW is a first person armored combat unit simulator.

Quote

The tabletop game doesn't account for rate of fire affecting accuracy, or plenty of other interesting factors.


If you go with this setup than you have an easy way to add parameters and predict and control their effect.

#83 omegaclawe

    Member

  • PipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 100 posts

Posted 12 November 2011 - 09:56 PM

View Postdiana, on 12 November 2011 - 04:38 PM, said:

Well, easy to learn, hard to master, for example, choosing target, movement and position to decrease the time it takes your weapons to align and fire, or even hitting a new target while your arm is in the middle of moving to the side, is a good thing, but HUD must be trustworthy.

View PostDraco Argentum, on 12 November 2011 - 05:42 PM, said:

Alright... I might have given the wrong impression. I'm not saying the hud should out-and-out lie to you, so much as omit precise data in favor of the more sweeping generalizations. Force the player to do some information gathering on their own to get a perfect result.
How exactly are the players going to know the windspeed all along the shot path? This would make your idea the worst of both worlds. The skill only guys will hate the random element. OTOH its a random element that is much more computationally intense than a simple gaussian distribution.

Much more computationally intense? Do you have any idea how complex many random generators can be? Your computer likely calculates more than 3 billion things in a second. If it needs to calculate an extra hundred, even (worst case scenario) per shot, the effect on performance is utterly negligible. We're not living in a time where we are strapped for resources in simple calculations like that.

That, however, is completely besides either of our points. So onto more important things, wind speed is not a random number generated individually with each shot; it will stay somewhat consistent throughout the period of a match, and can be entirely consistent if they don't bother to add gusts in. I would much rather see wind speed curve a shot randomly, though, than a shot curve at a 10 degree angle from the barrel it was shot out of immediately after leaving it. Admittedly, more of a visual problem, but still. In this situation, after the first shot, a player could conceivably know how it will affect his other shots, and adjust accordingly. If need be, they could also add a wind-speed gauge if it were that big of a deal... though it shouldn't be, considering that in real life, wind only provides significant effect to bullets traveling over 500m or so, and we're usually talking a foot or less; not enough to make a dead-center shot miss a battlemech.

That's my rant on the wind-speed part, anyway. It could also be dropped entirely without detriment, though, honestly. It was just listed as one factor that is not necessarily random that can be used to mess with aim.

Since you seem to be confused as to what my position is, allow me to clarify:
I am advocating the removal of random and immeasurable factors from the process of aiming and shooting. It should be possible, though not necessarily easy, for a player to account for all factors (excluding opponent action) that are affecting his mech, and use this information to make an accurate shot, regardless of what those factors are. It should be far more difficult, however, with more extreme factors.

Anything past that block of text is me offering a methodology or set of factors to achieve that goal. It is entirely possible I have not thought these additional methods through, however, and they may be flawed in such a way that does not conform exactly to the ideal without further adjustment. This more or less means the idea requires adjustment. :)

#84 Xhaleon

    Member

  • PipPipPipPipPipPipPip
  • The Money Maker
  • The Money Maker
  • 542 posts

Posted 12 November 2011 - 10:59 PM

View Postomegaclawe, on 12 November 2011 - 09:56 PM, said:

Since you seem to be confused as to what my position is, allow me to clarify:
I am advocating the removal of random and immeasurable factors from the process of aiming and shooting. It should be possible, though not necessarily easy, for a player to account for all factors (excluding opponent action) that are affecting his mech, and use this information to make an accurate shot, regardless of what those factors are. It should be far more difficult, however, with more extreme factors.


Can you compile and put down a simple numbered list of what these accountable factors are, both normal and extreme?

#85 Draco Argentum

    Member

  • PipPipPipPipPipPipPipPip
  • 1,222 posts

Posted 12 November 2011 - 11:00 PM

View Postomegaclawe, on 12 November 2011 - 09:56 PM, said:

Much more computationally intense? Do you have any idea how complex many random generators can be? Your computer likely calculates more than 3 billion things in a second. If it needs to calculate an extra hundred, even (worst case scenario) per shot, the effect on performance is utterly negligible. We're not living in a time where we are strapped for resources in simple calculations like that.


Yes, much more. A random number generator is just a table lookup on a computer. You need two of those for a single shot's deviation. Actually simming wind on a bullet is much more than that. This is a problem when its PGI's servers that have to do it.*

You should check the lengths benchrest shooters go to to compensate for wind. Thats the level of effort for wind to be anything other than a random number.

On to actual position. I'm saying that it is physically impossible for a pilot to account for everything since the mech is humming and vibrating even when standing still. The random effect is needed to make the game simulate firing an actual cannon since those are never 100% accurate.

*It has to be their end that does weapon calclations or the game will be extremely hackable.

#86 Lasercat

    Member

  • PipPip
  • 23 posts

Posted 13 November 2011 - 02:15 AM

View PostDraco Argentum, on 12 November 2011 - 11:00 PM, said:

On to actual position. I'm saying that it is physically impossible for a pilot to account for everything since the mech is humming and vibrating even when standing still. The random effect is needed to make the game simulate firing an actual cannon since those are never 100% accurate.


Does the game need to simulate firing an actual cannon though? It's just strange to nitpick about details like that in a sci-fi game of giant robots. If it doesn't add or detract from the experience of piloting a fictional robot it's probably ok to leave out.

View PostXhaleon, on 12 November 2011 - 10:59 PM, said:

Can you compile and put down a simple numbered list of what these accountable factors are, both normal and extreme?


They'd have to be made up stuff, there's no obvious list it's just whatever serves the purpose well enough. I'll try to explain a possible one.

Say you're using a hunchback and standing still. The ac is centered exactly on the crosshairs. Then you start moving at 20kph, now the ac starts to point a little above the crosshair. Go faster to 50kph and the ac drifts off to the right. So the faster somebody moves the more their weapons are going to be displaced from center. To make it even better the drift can be affected differently by the amount of torso twist and current heat level.

Now this displacement is not random, it's 100% predictable, but not obviously so. So someone who is sharp and has practiced a lot in a hunchback has the ability to accurately tell where their shots will land at any given moment. If they upgrade later to an atlas maybe the pattern of weapon displacement is different and they'll have to relearn everything, but the point is there is no RNG that makes it worthless to learn how to aim past a certain point.

Edited by Lasercat, 13 November 2011 - 02:20 AM.


#87 Xhaleon

    Member

  • PipPipPipPipPipPipPip
  • The Money Maker
  • The Money Maker
  • 542 posts

Posted 13 November 2011 - 02:49 AM

View PostLasercat, on 13 November 2011 - 02:15 AM, said:

They'd have to be made up stuff, there's no obvious list it's just whatever serves the purpose well enough. I'll try to explain a possible one.


Which is to say that you vehemently don't want a cone of fire, yet have absolutely no concrete idea how to do it otherwise.
You don't speak for omegaclawe, so I'm still waiting for his answer.

Lasercat said:

Say you're using a hunchback and standing still. The ac is centered exactly on the crosshairs. Then you start moving at 20kph, now the ac starts to point a little above the crosshair. Go faster to 50kph and the ac drifts off to the right. So the faster somebody moves the more their weapons are going to be displaced from center. To make it even better the drift can be affected differently by the amount of torso twist and current heat level.

Now this displacement is not random, it's 100% predictable, but not obviously so. So someone who is sharp and has practiced a lot in a hunchback has the ability to accurately tell where their shots will land at any given moment. If they upgrade later to an atlas maybe the pattern of weapon displacement is different and they'll have to relearn everything, but the point is there is no RNG that makes it worthless to learn how to aim past a certain point.


So basically, you want the game to become Counter Strike 1.6. Wonderful.

How about the Hunchback pilot realizes that his gun spreads too much when firing at the maximum distance and while running, so he brings himself closer and to a walk, thus causing his cone of fire to be small enough to hit reliably? Is that so hard to work with?

#88 Lasercat

    Member

  • PipPip
  • 23 posts

Posted 13 November 2011 - 04:20 AM

View PostXhaleon, on 13 November 2011 - 02:49 AM, said:

Which is to say that you vehemently don't want a cone of fire,


No. I guess maybe nothing like that would end up in the game regardless of what anybody wants, as per Bryan's post in another thread saying they're avoiding random elements.

View PostXhaleon, on 13 November 2011 - 02:49 AM, said:

yet have absolutely no concrete idea how to do it otherwise.


Sure anybody can have their own "concrete" idea on how to do it. But there are many other ways to do it too. What I described was one possibility. I'm sure there are better ways.

View PostXhaleon, on 13 November 2011 - 02:49 AM, said:

So basically, you want the game to become Counter Strike 1.6. Wonderful.


1.6 is a great game in that it rewards players who try to develop their skills extremely well - which is something that any pvp fps-like game which wants to be successful should do.

View PostXhaleon, on 13 November 2011 - 02:49 AM, said:

How about the Hunchback pilot realizes that his gun spreads too much when firing at the maximum distance and while running, so he brings himself closer and to a walk, thus causing his cone of fire to be small enough to hit reliably? Is that so hard to work with?


You should have noticed from my description that what you just said would be sound advice. Since the amount of drift depends on a few things it's hard to predict where the shot goes while running full tilt with some arbitrary torso twist and heat level. It'd essentially look random even though it isn't. You would only get a better idea of what was happening through observation and experience.

So sir longtime ace hunchback pilot would have learned approximately where his mech shoots given certain conditions, and would have an advantage over a newbie hunchback pilot in the accuracy department. Makes sense. Again the end result is the game supports the player getting better as far as aim goes, and not through gimmicky things like rpg stat bonuses either.

Of course a players aim will get better naturally, but something like this gives a baseline inaccuracy for everybody to deal with, exactly like the cof many people seem to like. The difference is it's an inaccuracy you can actually learn to overcome rather than just be stuck putting up with, as a cof model would have you do. It pushes the skill ceiling that much higher, which is a good thing.

#89 Draco Argentum

    Member

  • PipPipPipPipPipPipPipPip
  • 1,222 posts

Posted 13 November 2011 - 04:35 AM

View PostLasercat, on 13 November 2011 - 02:15 AM, said:


Does the game need to simulate firing an actual cannon though? It's just strange to nitpick about details like that in a sci-fi game of giant robots. If it doesn't add or detract from the experience of piloting a fictional robot it's probably ok to leave out.


Yet you go on to ask for very detailed aim point differences as speed changes. Simming chassis vibration is good for realism and good for stopping excessive sniping.

#90 Xhaleon

    Member

  • PipPipPipPipPipPipPip
  • The Money Maker
  • The Money Maker
  • 542 posts

Posted 13 November 2011 - 04:51 AM

View PostLasercat, on 13 November 2011 - 04:20 AM, said:

1.6 is a great game in that it rewards players who try to develop their skills extremely well - which is something that any pvp fps-like game which wants to be successful should do.


The difference between 1.6 and Source was that the spread of weapons in 1.6 was a fixed pattern, while Source had them truly randomized.

No, I don't want that to happen to Mechwarrior, because then the strategy of the game hinges entirely on who can shoot better and who knows the exact pattern of the weapon spread, instead of the game rewarding success at maneuvering and positioning and success at planning and coordinating with teammates. It doesn't matter how well you can get behind the enemy, if they are able to turn around and put a perfect salvo into you because they knew exactly how the weapon would act. It doesn't matter how difficult it is to predict, because if it can be predicted, players will train to abuse it.

Simply put, I want the game to focus and reward macro "tactical" skills rather than micro "shooting" skills.

The developers may have stated that they are avoiding "random" elements, but have also described how LRMs will weave around to simulate the imperfect spread on the cluster hit table on the TT game, which IS a random element. I strangely can't find it now, though. I sometimes fear that they are being as vague as possible because what they have in mind is more sinister and pandering than anyone here can imagine.

#91 Lasercat

    Member

  • PipPip
  • 23 posts

Posted 13 November 2011 - 07:06 AM

View PostXhaleon, on 13 November 2011 - 04:51 AM, said:

No, I don't want that to happen to Mechwarrior, because then the strategy of the game hinges entirely on who can shoot better and who knows the exact pattern of the weapon spread, instead of the game rewarding success at maneuvering and positioning and success at planning and coordinating with teammates. It doesn't matter how well you can get behind the enemy, if they are able to turn around and put a perfect salvo into you because they knew exactly how the weapon would act. It doesn't matter how difficult it is to predict, because if it can be predicted, players will train to abuse it.

Simply put, I want the game to focus and reward macro "tactical" skills rather than micro "shooting" skills.


Don't see any reason why the game can't reward a balance of tactical and shooting skill. In fact that's pretty much what Bryan said. I'll try to give a convincing argument as to why a non-random shot distribution doesn't result in that kind of situation:

What you said in your paragraph is hard to believe...you're suggesting a team which has the skill and coordination to completely outmaneuver another team and place themselves right behind them without them even noticing is going to have horrid "shooting" skill compared to the team that just sat there? Or that the team which has zero coordination and no situational awareness is going to be full of ridiculously good shots? Those 2 kinds of skill are almost never exclusive.

If you work in a team and have great coordination chances are you have a competitive attitude, and if you've got a competitive attitude chances are your aim doesn't suck either. If it does suck, then with your attitude being what it is you're not going to stand for having terribad aim (or your teammates won't :)), and you'll learn to make it better. Now if for some reason figuring out how a mech shoots isn't something you've got the patience for, you could still trade the hunchback for a jenner and learn the ins and outs of scouting, electronic warfare, and acting as some kind of command - things which are apparently going to be a big deal in this game. So it follows that even if you're the worst shot ever, there still is a crucial role for you to fill to give a winning advantage to your team.

With that being said, it's far more likely that any match between 2 teams will be against players who more or less have a grasp of all the game mechanics and understand how to out-shoot and out-think the other. The lopsided scenario you described would probably never happen because that kind of skewed skill distribution wouldn't exist for every member of either team. Instead the win would likely go to the team who was more skilled all around.

View PostXhaleon, on 13 November 2011 - 04:51 AM, said:

The developers may have stated that they are avoiding "random" elements, but have also described how LRMs will weave around to simulate the imperfect spread on the cluster hit table on the TT game, which IS a random element.


That's fine, we're not controlling the missiles after all. What would be silly is if the lock time was randomly generated, or x amount of your missiles randomly didn't fire once you pulled the trigger.

View PostXhaleon, on 13 November 2011 - 04:51 AM, said:

I sometimes fear that they are being as vague as possible because what they have in mind is more sinister and pandering than anyone here can imagine.


They seem to have some good ideas. Nothing to do but speculate.

#92 omegaclawe

    Member

  • PipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 100 posts

Posted 13 November 2011 - 08:54 AM

View PostDraco Argentum, on 12 November 2011 - 11:00 PM, said:


Yes, much more. A random number generator is just a table lookup on a computer. You need two of those for a single shot's deviation. Actually simming wind on a bullet is much more than that. This is a problem when its PGI's servers that have to do it.*

You should check the lengths benchrest shooters go to to compensate for wind. Thats the level of effort for wind to be anything other than a random number.

On to actual position. I'm saying that it is physically impossible for a pilot to account for everything since the mech is humming and vibrating even when standing still. The random effect is needed to make the game simulate firing an actual cannon since those are never 100% accurate.

*It has to be their end that does weapon calclations or the game will be extremely hackable.

Yes, it does have to happen on their end. And no, most RNG's are not lookup tables... that would be horribly memory inefficient, and much slower, due to the speed (or lack thereof) of the RAM. I'm halfway to a computer science degree, here. I know what I'm talking about.

Wind speed, as it happens, has a complexity of O(n) when calculated each tick, and is constant (O(1)) when only calculated upon firing the shot. The second case is so absolutely negligible that computer scientists often neglect to make any note of it whatsoever. The first case is positively minuscule, and will not create any noticeable overhead past the collision detection required to make a shot worth shooting (Collision detection >= O(n^3) in most cases in three dimensions).

So... what does all this tech mumbo jumbo mean? It means that, even if the servers were running on old Pentium II processors, you still would not notice a difference between pure random and wind speed/barrel position.

Real cannons, notably, are extremely accurate when care is taken to take proper measurements. The M1A1 Abrams is accurate to over 2 miles, while at cruising speed, against moving targets, due to a computer automatically adjusts the aim for all present factors. The Navy is testing a railgun accurate to over 100 miles. These weapons are not fundamentally different from the things we were using in WWII; we have only improved our ability to calculate the path of the shot, and extremely skilled gunners in that era could also make such shots by accounting for those factors on their own.

View PostXhaleon, on 13 November 2011 - 02:49 AM, said:

So basically, you want the game to become Counter Strike 1.6. Wonderful.


More or less. No arbitrary pattern, though, so much as battlefield factors. Keep in mind that 1.6 still has something like x10 the amount of active players as source (not to mention sales), and Valve is reverting back to a model similar to 1.6 for Global Offensive.

------

Anyway, Arbitrary, Incomplete, and Not necessarily Accurate list of factors:
  • The position the mech's body is in (Leg movements, torso twist, etc.)
  • The Forces acting upon the mech's body (e.g. Impact from steps, Turning/Torso Speed, recoil from previous weapons, knock from opponent's weapons)
  • (Related to previous 2) The weight distribution of the mech.
  • The position of the weapon upon the mech's body
  • The position of the actuators attempting to move the weapons to their proper aim
  • The heat of the mech/weapon (this is a real factor in real weapons; I'm not certain exactly how it affects the aim (less/more Velocity, perhaps?), but I do know the Abrams tank can properly adjust for it. Occam's Razor would probably just omit this one.)
  • The damage done to the mech
  • The forces acting upon the projectile (for those weapons applicable. Gravity, ambient wind speed, etc.)
These are all factors with real weapons, and all of them can be adjusted for properly by either a computer or an extremely skilled human being (commonly known as a "Sharpshooter").


Putting your skilled pilots in a position where they can do the most damage is a huge part of tactics, and even with such skills, inferior tactics can get them ganged up upon and destroyed before they can be effective. As long as information warfare plays a large factor in discovering what can shoot at you, and what you can shoot at, let alone other things, tactics will be huge. You can, however, only do so much about your team, and no matter what, you will end up with a number of teammates in any matchmaking system who do not try to use tactics in any regard, anyway. Those that do can prosper without being skillshots in the first place.

This is mechwarrior, not mech commander; it is not supposed to be pure tactics, anyway.

#93 UncleKulikov

    Member

  • PipPipPipPipPipPipPip
  • Bridesmaid
  • 752 posts

Posted 13 November 2011 - 11:35 AM

View Postomegaclawe, on 13 November 2011 - 08:54 AM, said:


Anyway, Arbitrary, Incomplete, and Not necessarily Accurate list of factors:
  • The position the mech's body is in (Leg movements, torso twist, etc.)
  • The Forces acting upon the mech's body (e.g. Impact from steps, Turning/Torso Speed, recoil from previous weapons, knock from opponent's weapons)
  • (Related to previous 2) The weight distribution of the mech.
  • The position of the weapon upon the mech's body
  • The position of the actuators attempting to move the weapons to their proper aim
  • The heat of the mech/weapon (this is a real factor in real weapons; I'm not certain exactly how it affects the aim (less/more Velocity, perhaps?), but I do know the Abrams tank can properly adjust for it. Occam's Razor would probably just omit this one.)
  • The damage done to the mech
  • The forces acting upon the projectile (for those weapons applicable. Gravity, ambient wind speed, etc.)
This is mechwarrior, not mech commander; it is not supposed to be pure tactics, anyway.

I would like that, with a cone of fire to represent most weapon's relative accuracy rating. Lasers could also factor in "particulate" matter to reduce their damage at range, to give them a weakness like "lead fire" of ballistic weapons.

And for fun, when a mech uses up ammunition, that should adjust his mech's abilities to be more spry since there is less load on the mechanisms.

Personally, I would rather wind not be one of the variables because it would be a small factor in the first place, and one that couldn't be predicted entirely along the flight path, which increases the amount of randomness that is out of the player's control.

That's my main beef with it - I'm for random distributions when firing, but the pilot should be able to directly influence those by (a pulsing the trigger, (b slowing his mech down, and so on. Wind resistance is going to be a (wait for it) maelstrom of inconsistent variables that could be very frustrating to deal with in the heat of battle.

#94 Nik Van Rhijn

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,905 posts
  • LocationLost

Posted 13 November 2011 - 11:51 AM

I was a "sharpshooter" with a rifle. When taking into account all the variables on a flat calm day at say 500m it is still impossible to put 10 consecutive shots into the same hole using an accurised "sniper" rifle. Due to minor variations in such things as exact bullet weight, very slight variations in the propellant, even down to variations in the harmonics of the vibration of the barrel. As well as those (trained to be minimised) variations due to being a human. The pattern of holes will be roughly circular with a weighting to the centre. This is known as the circular area of probability ot CEP and is produced by the cone of error and for that situaion will remain reasonably consitant for loger renges when extrapolated. These ballistic calculations are well documented and fairly accurate approximations can be calculated by the computer very rapidly. This applies to a single weapon, just as with the Abrams above.
Torso mounted weapons are assumed (nothing in literature to confirm) to have some form of mounting which is capable of some (unknown)
amount of movement to change the point of aim.. Grouped weapons may well be on a single mount with regard to smaller energy weapons, in which case each would have it's own point of aim ie if mounted in a square then undeviated shots would all arrive as holes with the same spacing in a square. From the literature it has always seemed that separate weapons (or groups) all had their own reticules some of the aiming being accomplished by movement of the mech's torso or arms. It is highly unlikely that a group of disparate weapons, mounted in a number different positions, incuding on arms would hit in the same place. Certainly the ballistics of the AC's in the game bear no resemblance to our universe, but are consistant with the effects given in the books. Not even the elite pilots in the books hit everytime and certailnly not dead centre. Weapons on the left side of the mech tended to hit (or miss) depending on the range, how much both vehicles wer moving, the footing, the weather etc. Very few pilots in the mechwarrior universe achieved consistant hits on fast moving targets while moving fast themselves. The only time I can remember Alpha strikes hitting with most (not all) shots are when both mechs were moving slowly or standing still at close range.
We are in the Mechwarrior universe - don't expect all your shots to hit - ever, or at least all in the same location. It should be nearly impossible to "leg" a light mech running across your field of view at top speed at say 600m let alone core it.
I would like to see accuracies better that in the books, as I don't think the punters would accept it from the comments made here and elsewhere on the forums. At the same time it is totally unrealistic to expect to fire mutiiple weapons and hit all in the same "called" area.
It's down to the dev's to decide what they want to do.

#95 Phades

    Member

  • PipPipPipPipPipPip
  • 334 posts

Posted 13 November 2011 - 12:19 PM

View PostNik Van Rhijn, on 13 November 2011 - 11:51 AM, said:

I was a "sharpshooter" with a rifle. When taking into account all the variables on a flat calm day at say 500m it is still impossible to put 10 consecutive shots into the same hole using an accurised "sniper" rifle. Due to minor variations in such things as exact bullet weight, very slight variations in the propellant, even down to variations in the harmonics of the vibration of the barrel. As well as those (trained to be minimised) variations due to being a human. The pattern of holes will be roughly circular with a weighting to the centre. This is known as the circular area of probability ot CEP and is produced by the cone of error and for that situaion will remain reasonably consitant for loger renges when extrapolated. These ballistic calculations are well documented and fairly accurate approximations can be calculated by the computer very rapidly. This applies to a single weapon, just as with the Abrams above.
Torso mounted weapons are assumed (nothing in literature to confirm) to have some form of mounting which is capable of some (unknown)
amount of movement to change the point of aim.. Grouped weapons may well be on a single mount with regard to smaller energy weapons, in which case each would have it's own point of aim ie if mounted in a square then undeviated shots would all arrive as holes with the same spacing in a square. From the literature it has always seemed that separate weapons (or groups) all had their own reticules some of the aiming being accomplished by movement of the mech's torso or arms. It is highly unlikely that a group of disparate weapons, mounted in a number different positions, incuding on arms would hit in the same place. Certainly the ballistics of the AC's in the game bear no resemblance to our universe, but are consistant with the effects given in the books. Not even the elite pilots in the books hit everytime and certailnly not dead centre. Weapons on the left side of the mech tended to hit (or miss) depending on the range, how much both vehicles wer moving, the footing, the weather etc. Very few pilots in the mechwarrior universe achieved consistant hits on fast moving targets while moving fast themselves. The only time I can remember Alpha strikes hitting with most (not all) shots are when both mechs were moving slowly or standing still at close range.
Unlike match shooting, it is not like you are trying to hit a target smaller than your projectile or even the same size as your projectile. The only hit location on a mech that should be anything like that would be the cockpit.

We do have to acknowledge that the table top rules were designed for simplicity and created a lot of fluff to create reason for that simplicity. It also contained asinine rules for example prone mechs, but did not apply the similar rules for elevated shooters, relative size of the target, which direction the target's speed was relative to the firing position, and other very relevant factors to accomodate for pilot based judgement errors. It also never addressed the actual speed of which the hard point turrets or arms could move and what those maximums were at all, which would also be a very firm sticking point if you are going for realism.


View PostNik Van Rhijn, on 13 November 2011 - 11:51 AM, said:

We are in the Mechwarrior universe - don't expect all your shots to hit - ever, or at least all in the same location. It should be nearly impossible to "leg" a light mech running across your field of view at top speed at say 600m let alone core it.
What mech moves 600 meters/s? If you meant 600meters/h, umm that's a turtle. A human can run a 6 minute mile which translates out to roughly 16,093 meters/h.

View PostNik Van Rhijn, on 13 November 2011 - 11:51 AM, said:

I would like to see accuracies better that in the books, as I don't think the punters would accept it from the comments made here and elsewhere on the forums. At the same time it is totally unrealistic to expect to fire mutiiple weapons and hit all in the same "called" area. It's down to the dev's to decide what they want to do.
The areas proposed for the mechs for hit sizes originally are like billboards on the side of the road in many instances. Convergence to the same hole would be impossible, but convergence to anywhere inside that billboard would be likely. One of the core game issues, that invariable ended up being a balance issue in many of the previous mech warrior titles, ends up being those areas are too weak to take a hit relative to the weapons being fired in tandem. The "easier" solution is to make those areas more segmented, so that you had to hit the same hole in order to get down into the inner structure. Make the segements smaller on the smaller mechs and larger on the bigger ones and guage hit area for the larger weapons to spread out, as needed, if they contact more than one section prior to penetration of outer to inner.

#96 omegaclawe

    Member

  • PipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 100 posts

Posted 13 November 2011 - 12:28 PM

View PostPhades, on 13 November 2011 - 12:19 PM, said:

The areas proposed for the mechs for hit sizes originally are like billboards on the side of the road in many instances. Convergence to the same hole would be impossible, but convergence to anywhere inside that billboard would be likely. One of the core game issues, that invariable ended up being a balance issue in many of the previous mech warrior titles, ends up being those areas are too weak to take a hit relative to the weapons being fired in tandem. The "easier" solution is to make those areas more segmented, so that you had to hit the same hole in order to get down into the inner structure. Make the segements smaller on the smaller mechs and larger on the bigger ones and guage hit area for the larger weapons to spread out, as needed, if they contact more than one section prior to penetration of outer to inner.

This might sound nice, but it brings with it the problem of making more or less everything ridiculously hard to kill when you can't do it in 1 or two shots.

...of course, if you have weapons which spread the damage more as opposed to penetrating deeply... such as LBX AC's, missiles, perhaps even PPC's... hrm... actually, that might just work.

Edit: Also, as for the sharpshooter thing... we are talking a whole different scale, here. In a handheld rifle, a 1% variation in bullet mass might be acceptable, whereas it would make a massive difference in mech-sized weapons, etc. May result in a reduction of factors which throw the accuracy off and, at the same time, firing a larger shell at a larger target means those variations in accuracy are less important. Deviation of shots about the size of a human torso on a battlemech is more or less no deviation at all.

Edited by omegaclawe, 13 November 2011 - 12:32 PM.


#97 Nik Van Rhijn

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,905 posts
  • LocationLost

Posted 13 November 2011 - 12:42 PM

I like the idea. As for AC's maybe they should be split into 2 groups as some fire a single shot (retain original name) while others fire burst and could have the suffix B. and do spread damage (LBX can fire cluster or slug). As has been suggested elsewhere we could work in the expanded heat scale etc from Solaris.

#98 GreenHell

    Member

  • PipPipPipPipPipPipPip
  • 543 posts
  • LocationGrandmas House

Posted 13 November 2011 - 01:23 PM

View PostVYCanis, on 11 November 2011 - 04:39 AM, said:

I agree with you Mchawkeye

I'm of the opinion that CoF is a heavy handed solution that would simply make mechs feel ridiculously goofy.
CoF works as a decent mechanic for hip shooting in FPS games simply by virtue of it being an unsupported weapon held by a human being. So when you are spraying and praying in your FPS of choice, it does not bump into the average person's suspension of disbelief.

however this breaks down when you suddenly have some 31st century combat vehicle that can barely hit the broadside of a barn, with a laser no less.
.
this is mechwarrior and we aren't driving human beings, we aren't hip shooting, all these weapons are hard mounted. We are controlling multi ton articulated turrets. Weapons are all supported here. Weapons should be firing fairly accurately, not spraying wildly. They aren't held on to mechs by bungee cords and duct tape (which honestly would be the only rational explanation for CoF at the short ranges that combat will be taking place). I mean, I love the tabletop, but i don't care that dice rolls mean that a Nova to discoball lasers every which where whenever it alphas, it would look silly as hell from a pilot's perspective.

That being said, if there is CoF it should be limited to ACs at extreme far ranges.

Beyond that, the real issue should not be making weapons so over the top inaccurate that you can't hit stuff, the real issue should be to make it where weapons are accurate, but your aim is difficult, weapon convergence while consistent and predictable, takes effort to achieve, and lastly that weapon damage by virtue of how it is dealt, is not easily concentrated. (laser damage spread over its duration, ACs spreading fire in their bursts....)

i.e. If you are driving a laser boat mech. what would make more sense to you?

A. Having to deal with sway, knock, momentum, and other interference as you struggle your crosshairs over a moving target, having your lasers trying to align to the bouncing crosshairs, shooting lasery death, and having the lasers track a mostly concentrated line of damage from their right leg up to their left arm. of the lasers that miss, it was due to the arm they were mounted in getting hit by a missile at the last second, fouling their aim by a few degrees

or

B. aim at the general direction of my target, my firing cone is still kinda wide, alpha lasers, they discoball everywhere like a laser blunderbuss, even ones that are mounted in the same location or are parallel.



You bring up a good point friend:
Is a CoF really proper at the urban ranges that the game devs are hinting at?
Would a CoF really even matter if you plan on engaging in urban warfare?
Would it even make a bit of difference either way?

At max range, I could see weapons having a "TINY" bit of CoF, but then you could blame it on recoil, movement, damage, etc...
In the end, I believe a strong simulation will bring out the natural aiming difficulties of firing large weapons from a walking platform.

#99 Woodstock

    Member

  • PipPipPipPipPipPipPipPip
  • Mercenary
  • Mercenary
  • 1,166 posts
  • LocationKrakow

Posted 13 November 2011 - 01:44 PM

View PostVYCanis, on 11 November 2011 - 04:39 AM, said:

I agree with you Mchawkeye

this is mechwarrior and we aren't driving human beings, we aren't hip shooting, all these weapons are hard mounted. We are controlling multi ton articulated turrets. Weapons are all supported here. Weapons should be firing fairly accurately, not spraying wildly.



However the canon of the IP states pretty bluntly that is exactly what the technology of the time did. Weapons clustered ... a bit ... but basically you fired a volley of missiles, lasers AC's and PPC and they all hit different parts of the target!

Personally I think if we are firing a single weapon then sure it hits where the gun points. But the moment you add more weapons to the volley you should get an expanding reticle. The more weapons the wider it gets. With more than 5 weapons being roughly the size of the upper body of the target.

So you can aim but 'for the torso' or 'for the legs.'

It does not have to be worked out that way though. The mechanic does not have to be so obvious. maybe you do have a small crosshair and sure one weapon will always hit there ... but have the other shots be assigned hit locations by the system.

#100 Amechwarrior

    Member

  • PipPipPipPipPipPipPip
  • Legendary Founder
  • 729 posts
  • LocationHawaii

Posted 13 November 2011 - 01:54 PM

View Postomegaclawe, on 13 November 2011 - 08:54 AM, said:

Anyway, Arbitrary, Incomplete, and Not necessarily Accurate list of factors:
  • The position the mech's body is in (Leg movements, torso twist, etc.)
  • The Forces acting upon the mech's body (e.g. Impact from steps, Turning/Torso Speed, recoil from previous weapons, knock from opponent's weapons)
  • (Related to previous 2) The weight distribution of the mech.
  • The position of the weapon upon the mech's body
  • The position of the actuators attempting to move the weapons to their proper aim
  • The heat of the mech/weapon (this is a real factor in real weapons; I'm not certain exactly how it affects the aim (less/more Velocity, perhaps?), but I do know the Abrams tank can properly adjust for it. Occam's Razor would probably just omit this one.)
  • The damage done to the mech
  • The forces acting upon the projectile (for those weapons applicable. Gravity, ambient wind speed, etc.



How does effect a laser boating 'mech, standing, at long range? You know, like the problem we have had in the past with previous games? Not all maps will be urban. The factors you and others have listed against cone of fire does not negate the resulting near pin point accuracy of a long range, unmoving, laser heavy design. Only the actuators you listed or the "air particulates drain damage" like others have posted effect this scenario, I know you said the list is not perfect. What factors could you use to make this situation less accurate for the standing, long range laser situation that do not resemble cone of fire? Increased heat is not the answer, because he is overheating, and you are dead. He still wins and now lasers in more balanced designs are all jacked up with extra heat..

Unless this can be overcome, the best thing in every non urban map would be to get a laser heavy assault 'mech, stand and snipe on the lip of a hill and just reverse downhill every time someone tried to LRM you. I do not really want to debate this in a cone/expanding reticle vs other factors, I have said enough in other threads. I just want to know what the solution for the MW4 style long range laser boating is without cone of fire and expanding reticle. All I ask are your ideas for this, I do not intend to reply, see my post history if you want to know what I think.





6 user(s) are reading this topic

0 members, 6 guests, 0 anonymous users