Home » MODDING HQ 1.13 » v1.13 Idea Incubation Lab » New CTH system - Presentation
Re: New CTH system - Presentation[message #250553]
|
Wed, 28 April 2010 16:07
|
|
Headrock |
|
Messages:1760
Registered:March 2006 Location: Jerusalem |
|
|
I agree, the robot should be unable to use scopes, as you said, and also a serious penalty for moving targets. So the robot would be pretty weak at long range or dynamic combat, but can be a serious help taking care of hiding enemies in close quarters - just open the door, take a hit or two, and autofire the crap out of them. And yeah, since the robot is on tracks, and probably weighs a lot more than a person, it would be considered automatically as having both the Expert Heavy Weapons and Expert Auto Weapons skill.
Dunno about bolt action. I guess MadLab could rig the robot to cock its own weapon. After all, the robot comes pre-installed with whatever weapon you chose for it, right? So some customization is to be expected when installing that weapon.
Report message to a moderator
|
|
|
|
Re: New CTH system - Presentation[message #250561]
|
Wed, 28 April 2010 16:50
|
|
Forral |
|
Messages:71
Registered:May 2009 Location: Sweden |
|
|
Xavhum something from UFO type game:
-in reaction (interup in JA2) don't allow lots of aim clic, it's a fast reaction to something, you don't have the time to aim if an enemy just run and pass in the window in front of you, the same when he begin to be see just after a house corner ect -> caracters have just the time to fast move his gun and shot
-> made aim clic cost lots more
-> made clic to add shot in auto fire cost more
or
->made the first clic cost a lot (made in fact auto fire react with less shot than in your turn and aim shot without lots of aim clic)
or radicaly:
-> aim clic is not allowed and just few shot in auto mod is possible in interuption (ufo like game: reaction fire is always instant no aim shot)
(and don't allow to switch the gun or the thing you have in your hand, don't think it is doable (oh!! an enemy, hum go to my bagpak schearch the handgun, aim during long time this enemy, shot him, reaim reshot, he is dead good (and made all of this in less than 1/32 second ) )
In reaction sequences I would argue that it's not a matter of characters acting during a period of time equivalent to that of a 'blink of an eye'. Rather, it's turn-based games' way of simulating real-time combat. What interuptions or reaction sequences simulate is something more along the lines of simulataneous actions, as opposed to instantaneous ones. So rather than pulling out your gun, taking aim and firing at your foe during a brief second, you're doing this all *while* the enemy is running past. The fact that the enemy is frozen in position is merely a limitation of the visiual interpretation of turn-based games. This is further supported by the fact that you suffer penalties to CTH in accordance to to the enemy's movement - ie, the enemy, despite seemingly standing still, is actually running past while you act.
It may create some odd scenarios in which you're capable of stringing together a fairly lengthy collection of actions in situations where you could only feasily have seen the enemy during a very brief window of time, such as when spotting a foe through a window from the opposite side of the room. Should you however be standing at the window yourself, firing at the the foe running past outside, it makes a lot more sense as the area in which the enemy could be feasily spotted and tracked is considerably greater.
Even so, suppose an enemy *ends* his turn standing in front of that window. Realistically he won't actually just be standing there. He's in the process of moving past it. So your reaction or interuption round isn't really going to be any different from that of a normal turn. By that I'm saying that I don't feel interuption sequences should be given any penalties or bonuses that differ from those of a normal round.
However, AP costs for inventory management is something I'd be in support of. It's something that would prevent some of the nonsense of the situations you described. I've lobbied for such ideas quite a few times actually. Sadly it's probably not related to this particular project.
Report message to a moderator
|
Corporal
|
|
|
|
Re: New CTH system - Presentation[message #250677]
|
Thu, 29 April 2010 15:07
|
|
Headrock |
|
Messages:1760
Registered:March 2006 Location: Jerusalem |
|
|
Quote:Interesting read yes. But i believe the source of evil here is the CtH calc. Having such thing in the game is just too good.
As explained in the article above, the problem is a little more complex. The CTH Bars are too good because they give you a perfect indication about your chances to hit. You're not going to waste your time firing a bullet when the chances are low (unless on auto or just trying to suppress), and can min-max your AP spending by aiming just enough to hit the target. However, if CTH Bars are turned off, it's more guesswork than actually needed.
The new system, on the other hand, is very different in terms of CTH bars because they don't actually tell you how likely you are to hit, just how well you are aiming and holding the gun. The amount of 3D maths carried out after the trigger is pulled is so significant, that firing when the bar is full really doesn't guarantee a hit at all - it just indicates that you have somewhat MORE chance to hit the target than you would if the bar was half-full or empty. Whether that's enough chance to actually hit the target is not revealed to you, and based largely on distance, weapon, target stance, and a lot of other stuff that isn't factored into the CTH bar anymore.
BTW, CTH bars didn't exist in JA2, so I don't understand how you stopped using them then.
Quote:If i understand it correctly in order to compensate for extra given info you decrease overall accuracy and add some randomness?
Mainly I just give much less info - roughly the amount you'd actually get if you could ask your merc how much time and effort he's spent into aiming his gun and how accurate he FEELS. Also yes, there's a lot more randomness, which is gradually decreased with better aiming, better gun, and better shooting conditions. But never to a point where a full bar is a guaranteed hit, especially when it's a full bar using a pistol at 20 tiles.
Quote:well wouldnt that be frustrating. Most people expect go get a hit if the counter shows 9x%+ (in fact id say the frustration level is somewhere about 75%+ ). And i definitely expect a hit if its 99%.
Then don't. Or turn the bars off if you fancy, that's why the menu option is there. Also you're contradicting yourself there - you say you don't want any info, but are worried that what IS given would confuse you?
Report message to a moderator
|
|
|
|
|
|
|
Re: New CTH system - Presentation[message #250692]
|
Thu, 29 April 2010 19:44
|
|
Headrock |
|
Messages:1760
Registered:March 2006 Location: Jerusalem |
|
|
Quote:I commented on "What's wrong with the Old CTH system?", where i believe nothing is wrong it just isnt supposed to work with CtH indicators.
I don't really fully agree with that either. It was fine for JA2, but the types of weapons added by 1.13, as well as a whole slew of attachments that are far more powerful than the ones we had in JA2, make the calculation itself too lenient. So even if you do turn off CTH indicators, an MP7 is still amazingly powerful at night time, and an M24 EBR is way too powerful in daytime. You don't need to read the CTH, you just need to shoot at the right distance. Just ask people who play 1.13 with the CTH bar turned off, though there certainly aren't many for the exact reason that that's TOO LITTLE information.
Quote:I dont understand. There is no timelimit in combat, no pressure. Too much guesswork for what. For predictions? But isnt prediction exactly what makes certain things overpowered. On the other hand, stand up, fire a burst with a Kalashnikov, sit down. Like in real life.
There's a certain "perfect" point where a game is exactly both "fun" and "difficult" at the same time, and it's always a compromise. Having no data about the shot (a-la JA1/JA2) requires the player to pay a lot of attention to every little detail, some of which is in fact hidden completely and can't be taken into account. This has prompted lots of players over the years complaining about having no chance-to-hit indicator of any kind, because the game doesn't explain why a shot did or did not hit the target. On the other hand, having too much data about the shot (a-la 1.13) allows the player to spend less time micromanaging each shot if he wants to increase his hits, but in turn opens up a lot of exploits which pretty much every 1.13 player has discovered over the years.
The "perfect" spot is somewhere in between, having enough data to know whether a shot is "worth" trying, but not enough to know whether it's a sure shot. The AK user from your example has a lot of data despite how simplistically you portrayed him. He knows how much the weapon weighs in his hand, he knows whether he's got the sight aligned with the target, and how good his grip on the AK is at the moment, and so forth. What he DOESN'T know is what's going to happen after he pulls the trigger. But he doesn't have to think about each component separately, he just gets the "feel" of his aim, and pulls the trigger when it's good enough. That's what will be reported to the player - how sure the merc is about the shot he's about to take. Whether the shot hits or not is up to a lot of factors that he has no idea about, like how many cordite particles the bullet is going to encounter on its way out of the barrel, the exact wind speed during each centimeter of bullet flight, and whether or not the muzzle was aimed percisely enough at the target when the trigger was pulled.
So the new CTH system separates these things. It lets your merc tell you the "feel" for the shot, but doesn't tell you what your merc can't possibly know. That's what I meant by finding the balance between no info and too much info. You shouldn't have to calculate all the details yourself because that detracts from your enjoyment of the game. On the other hand, you shouldn't have all the data calculated for you (or even available at all), otherwise you can analyze the system to remove all the challenge.
Quote:Changing range/hitpercentance woulnt change a fact that players will continue to minmax shots if they have the possibility for that.
They won't. As explained above, they will not have all the data required to perform perfect shots, at least not without excellent conditions and the suitable weapon for that condition. The formula can no longer be used to ensure certain hits or min-max at all, but it CAN be used to determine the right time and place to take a shot if one wishes to conserve time and ammo.
In other words, putting more emphasis on having the right weapon at the right time, and not shooting when you've got the wrong weapon. That's emphasis on tactics and forethought rather than straight analysis of data and micromanagement.
Report message to a moderator
|
|
|
|
|
|
|
|
|
|
|
|
Re: New CTH system - Presentation[message #250728]
|
Fri, 30 April 2010 14:24
|
|
DepressivesBrot |
|
Messages:3658
Registered:July 2009 |
|
|
Seems like quite a lot of effort for one or two weapons that would actually use it. (The AN-94 and possibly the metalstorm systems, G11 seems to have a constant rate of fire in full auto)
But seeing as the xml has to be changed regardless of this little feature, it probably wouldn't matter.
If the xml tag for muzzleclimb would take values like <>2;2;4<>, then these could be split upon parsing and stored internally in some form of linear data structure (array, list, ... don't know what's the most efficient in c++)
The values would then be used according to the shots already fired.
The last (and often only) element would then determine the value for all following shots.
{Just a thought how I would do it, I'm sure there is a more elegant/efficient way, also the part about the xml is highly theoretical as I haven't found the place where those are read}
Alternatives would include a tag, either boolean or including the number of shots before recoil takes effect (default =0) or several tags (probably worst, could be confusing for modders)
On a sidenote: Obviously you understand this stuff better than me, but shouldn't the fourth round have a slightly higher value than the following as it takes (partially at least) the combined recoil of the previous three?
Report message to a moderator
|
|
|
|
Re: New CTH system - Presentation[message #250731]
|
Fri, 30 April 2010 14:57
|
|
CptMoore |
|
Messages:224
Registered:March 2009 |
|
|
The muzzle climb should be less if the character is stronger, or trained, and be less when foregrip is used, how good the balance is (including weight)...
You need to find a way which includes all these parts into a formula and then only store the gun part of the formula into the XML.
x, y seems to be too simple.
Report message to a moderator
|
Sergeant 1st Class
|
|
|
Re: New CTH system - Presentation[message #250733]
|
Fri, 30 April 2010 15:34
|
|
usrbid |
|
Messages:1506
Registered:December 2008 |
|
|
The G11 has recoil buffering for up to three shots. In addition when shooting a three round burst it always automatically places the three shots in a (slightly rotated) triangle. I believe the same is true when you full auto the G11, the first three are going to be placed in a triangle with recoil buffering. This is what the gun stock is for, it can absorb recoil up to three times.
So my x, y suggestion is already a simplification as the system is not able to place the first three shots in a triangle. But anyway, the G11 is not actively used, there are probably less than ten working prototypes.
However there are quite a few guns which use a much faster fire rate for the first two or three bullets, there is no recoil buffering, the bullets are fired so fast that the operator won't feel recoil until after the first two or three rounds hit the target. For later rounds the rate of fire is reduced.
Other than that, there are *many* ways to make a gun shoot the first 2-3 rounds without the operator feeling recoil, here are just a few examples. You can even configure pistols to shoot two rounds before you are off target.
Report message to a moderator
|
|
|
|
|
Re: New CTH system - Presentation[message #250738]
|
Fri, 30 April 2010 17:48
|
|
CptMoore |
|
Messages:224
Registered:March 2009 |
|
|
DepressivesBrot@ CptMoore
Please re-read the paragraph about 'Muzzle climb' x,y only determines the basic direction of recoil and therefore represents only the gun part. The operators stats are also figured into the formula, somehow. (Headrock hasn't specified that part en detail yet)
After doing that and going on Wikipedia I have some additional random thoughts:
A gun usually does not have any directional recoil, its the physical contact with the gun which gives most of the recoils direction. The gun gives only the possible contact positions for the shooter / environment, and the ammo type. The ammo type determins the energy behind the recoil.
The recoil direction changes when the contact to the gun changes, so folding stocks change the recoil, next to foregrips and bipods.
Bipods also reduce the recoil energy, making the recoil move the muzzle less.
So I would suggest using sx, sy to define the standard contact between shoulder and shooter. (sx = how much right to shooter / on the sholder, sy = how much over the contact center). Add recoil energy "re" based on the ammo type.
You could then just translate (sx, sy) * re to x, y. Foregrips would reduce re a bit and remove some of the y based on strength. Bipods would make sx and sy almost zero. A folded stock would reduce the sx to zero and make sy way higher and add re, since energy isnt aborbed into the body (and of course a accuracy minus is added with a folded stock).
I would still seperate sx, sy for the XMLs so you can store and override sx, sy (gun contact) independently of re (ammo recoil energy). And delayed recoil needs to be somehow specified by telling how the recoil energy is generated (burst recoil: 1. 50% re, 2. 120% re; auto recoil: 1. 40% re, 2. 140% re, 3. 100% re, n. 100% re).
Report message to a moderator
|
Sergeant 1st Class
|
|
|
|
|
|
|
|
|
|
Re: New CTH system - Presentation[message #250879]
|
Sun, 02 May 2010 19:54
|
|
Headrock |
|
Messages:1760
Registered:March 2006 Location: Jerusalem |
|
|
UPDATE
Today I've finished writing the bulk of the shooting mechanism. That means that this system now works in game, although it's very basic at the moment. It can calculate Muzzle Sway, adjust for target movement, apply scope magnification, and apply bullet deviation once the shot is fired.
In addition, an INI file has been created that contains all the important coefficients for this system. This means that users will be able to tweak pretty much every factor in the calculation, from the "NORMAL" shooting distance to the effect of Fatigue on Aiming CTH, and so forth. There are over 50 coefficients currently, and possibly more will be added later.
Currently, the system is missing its drop compensation (shooting beyond gun range) and recoil. There is also the issue of special mechanics for the robot/tank/alien. Once these features are added, the shooting mechanism will be fully functional. I'll also alter the CTH bar to give a little more info than it currently does, namely the magnification factor being used for the current shot (as you remember, it's either the scope mag factor or (target's distance/normal distance), whichever is lower). That should make shooting much more intuitive.
At that point, of course, there's also the matter of the massive XML work required. This includes adjusting the Accuracy value for all guns to a 0-100 range (where pistols and shotguns occupy the lower echelons, and sniper rifles go all the way up to the 90's). Scopes have already been altered to have a "magnification factor" instead of a Minimum Range To Use. Finally, I'm going to need to figure out how each attachment works, and also decide which tags to use for such attachments. It's quite possible that tags would be enabled for use that allow attachments to "intervene" in many different parts of the CTH or Shooting formulae, allowing modders a maximum ability to model attachments on the actual effects of their real-life counterparts.
Finally, there's the matter of tweaking the system. The values I've set as default for the various coefficients seem to work nicely so far, but that might change once people actually start testing this in the heat of combat and with various weapons. Fortunately, the INI file will allow testers to mess with these values as they see fit without having to wait for me to change the code. Eventually, once good balanced values are determined, the need for the INI will be reduced and most if not all of the coefficients will be hard-coded. Possibly the most important coefficients, the NORMAL_SHOOTING_DISTANCE and MAXIMUM_SHOT_ANGLE will remain in JA2_Options.INI, allowing players to easily determine the accuracy of all shots simply by tweaking these two.
I expect that unless the XML work proves to be too much for me, I should have this done by the end of the month. Help with the XMLs would of course be greatly appreciated, if anyone's up to it.
And as usual, more ideas and feedback are always required. Specifically at the moment would be how attachments should influence this system. I've got Scopes working properly, and Bipods already have their effects written down, but things like Reflex Sights, Grips and Lasers are still "in the dark", so keep these suggestions coming.
Report message to a moderator
|
|
|
|
Re: New CTH system - Presentation[message #250881]
|
Sun, 02 May 2010 20:13
|
|
Headrock |
|
Messages:1760
Registered:March 2006 Location: Jerusalem |
|
|
Quote:A gun usually does not have any directional recoil, its the physical contact with the gun which gives most of the recoils direction. The gun gives only the possible contact positions for the shooter / environment, and the ammo type. The ammo type determins the energy behind the recoil.
Unfortunately, correctly modeling the contact point would be very difficult in a 2D game where weapons aren't actual physical models interacting with the character model. Therefore we've got to work with what we have, and that's a flat recoil value that tells us how much the gun pulls when held normally. Barring extreme circumstances, this simplified system is sufficiently realistic.
About ammo, again, most of our ammo is "standardized", meaning it shouldn't change the gun's recoil. However, if a modder does wish to change recoil using differently-charged ammo, that's what the recoil modifier tag is for (tracers currently do this if HAM isn't activated). The same thing would work for the attachments you mentioned (folding stock, for instance).
Quote:One thought in terms of realism regarding compensating for muzzle climb: a lot of people tend to shoot low and to the left because they're improperly compensating for recoil and/or muzzle climb. Maybe someone with a lower skill might have a chance to throw rounds low and to the left when they use automatic fire or fire a semi auto that has a lot of recoil.
I'm about to publish another chapter about how recoil is going to work, after having come up with an interesting solution for it today. It works somewhat like what you guys have been saying here, but I think it portrays realistic recoil more efficiently, or at least would produce more realistic-LOOKING results.
Quote:The new system will be used for militia and enemy soldiers as well, will it? Wouldn't that require adjusting the AI system so that the AI knows how to make use of it? For example it could make a sniper switch between his sniper rifle and a sidearm when attacked at close range.
The AI is always a problem, even with the old CTH system. Knowing parts of the AI, I can say that it's possible that the AI would be encouraged to fire more often with the new system, given that low CTH is relatively rare now. Of course, more bullets are going to miss the target, so it's possible that the two will somehow cancel each other out, giving a combat environment filled with bullets but less hits. Then again, I have no way to tell how AI is going to handle this, that will only be revealed when testing. However I can say that I will try to adjust the AI to the best of my ability once we know better how it reacts to the changes.
Quote:
Will there be a factor which shows that someone is familiar with a certain gun?
Unfortunately no. Weapon familiarity is a much-requested feature, and the reason it doesn't exist is mainly because it's so complicated. You'd need a way to track how much time a character is carrying the weapon and/or how many times he's fired it, and given hundreds of weapons this can be complicated or at least memory-consuming, given over 150 NPCs.
Quote:Won't the new use of certain gun values require to modify the UI to represent their new meaning? Let's take recoil for example. It would be interesting to the player to know which weapon has lower recoil when playing an autofire merc. It will be hard to see this in a combat situation because the merc can't carry 5 guns with him to "test" their recoil?
As with the old CTH system, recoil would be displayed in EDB, just with X and Y values (that is, left/right and up/down). In combat, there is currently no plan to show recoil before taking the shot. However, the behavior of the recoil system (as will be explained later today) has many random factors and is heavily dependent on the shooter's skills. Therefore, there's really no good way I can think of to display recoil before the shot. If, after reading the new recoil system, you can come up with a reasonable way to display it, I'll take that into consideration.
Quote:By the way I would vote for adding new values to the existing item XMLs so that the old engine ignores the new values while the new engine will ignore the old values. It's up to the modders to decide which system they support or if they support both.
That might cause a rift in the modding community, and we're already battle-weary as it is. If the system is made optional (I.E. the player gets to pick which system to use), a mod with no support for one system or the other would receive only a fraction of the potential number of users, and would cause massive problems with people complaining that they don't understand why a mod doesn't work. This is possibly the LEAST favourable solution, although at the moment I have no idea what the best (or even just better) solution would be. It's possibly just as bad if we FORCE everyone to use this system, just imagine how many people will get pissed off and stop playing.
Quote:kevlar-grass has a wrong jsd
That's quite possible. Since I don't know what the problem is (not enough playing time lately) nor how the JSD system really works, I can't confirm nor deny. There's always the possibility that by fluke the new system (which disregards many things and treats others differently than the old system) would somehow bypass this problem. Remember that visibility is now much less important in CTH than it was previously...
Report message to a moderator
|
|
|
|
Re: New CTH system - Presentation[message #250882]
|
Sun, 02 May 2010 21:53
|
|
CptMoore |
|
Messages:224
Registered:March 2009 |
|
|
Headrock ... we've got to work with what we have, and that's a flat recoil value that tells us how much the gun pulls when held normally.
I assume you are talking about "AutoPenalty" and "Burstpenalty" as already available flat recoil. No changes to existing XML are always preferable.
Modelling my idea with almost correct physics isn't that hard, simple physics in high school should be sufficent for the logic behind it (mass of gun, mass of arm, energy of ammo, strength/force of left arm, strength/force of right arm) and C++ allows easy implemention of physics for one to n dimensional spaces due to operator overloading. But again it would require contact and ammo energy additions into the XML.
Of course it is possible to write a complete physical simulation with timing just for the heck of it and include it, but then many more parameters would be required anyway (mass distribution on gun, correct placement of foregrips, spring mechanisms inside guns and their timings etc..).
Report message to a moderator
|
Sergeant 1st Class
|
|
|
Re: New CTH system - Presentation[message #250885]
|
Sun, 02 May 2010 23:51
|
|
Headrock |
|
Messages:1760
Registered:March 2006 Location: Jerusalem |
|
|
RECOIL SYSTEM EXPLAINED
I've been toying with several possible ways to model recoil, especially compensating for recoil while the gun is firing. Today in the shower (of course) I got the idea that I'm about to explain below.
As those of us who've shot firearms before know, recoil is compensated for by applying force to the gun in the opposite direction of the recoil. When you're familiar enough with the gun, or have experience with this sort of thing, you tend to apply about as much force as would be required to cancel out the recoil. Of course, it's hard to apply exactly the required amount of force unless you're a great master, and sometimes the gun kicks more powerfully than you could possibly compensate for.
Even if you apply a reasonable counter-force, it's more than likely that the muzzle of the gun will shift anyway due to small discrepancies adding up. Therefore, as the autofire volley continues, it becomes imperative to intentionally apply more force towards the target's center, trying to compensate for all the accumulated muzzle deviation. Whether or not this extra force actually pulls the muzzle back to its original position is determined by whether or not we've applied just the right amount of force. Over-compensating is also possible - anticipating more recoil than is actually generated by the gun, causing the muzzle to go past the target center and end up on the other side.
To model this, I decided to use several key values, each of which depends on different factors. These represent the forces that are at work during the autofire or burst volley.
Recoil Force
This is the amount of force being generated by the gun as it fires. As explained earlier in this thread, recoil will have a certain X/Y pull, indicating the direction and distance that the gun's muzzle moves after each bullet given that no force is used to counter it. Recoil Force is simply calculated by pythagorean theorem: SquareRoot(X^2 + Y^2).
Max Counter Force
This is the maximum amount of counter force that the shooter can exert on the gun. It is determined primarily by strength, although attachments, bipods and shooting from a crouched stance can potentially increase the shooter's max force. If the gun's recoil force is greater than this max counter force, then the shooter can't possibly compensate for the entire recoil, and if the volley is long enough the gun will inevitably end up pointing far from the target.
Force Accuracy
By properly "reading" the gun, the shooter can know (intuitively of course) how much force to apply to counter the gun's recoil. Determined primarily by experience and Auto Weapons skill, this value tells us how accurately our shooter can predict the amount of force the gun is going to generate, and therefore apply a more accurate amount of force to counter the gun, and/or the extra force required to return the muzzle to its original position (pointing at the target).
Total Offset
Over time, assuming recoil hasn't been properly compensated, the muzzle moves further and further away from the target's center. As the distance grows, it becomes more imperative for the shooter to increase his applied force to counter this. In other words, overcoming not just the recoil from each shot, but whatever recoil has not been compensated for already.
How all this works together
The gun pulls in a specific direction with a specific force each turn. If no counter-force is applied, this is how much further away the muzzle will be, in the direction of the pull, when the next bullet is fired.
The shooter guesstimates the amount of force required to counter this force. The actual counter-force applied is randomal, although it's more likely to be close to or equal to the recoil force if the shooter has better Force Accuracy. Due to the randomality involved however, counter-force will usually be either less or more than the recoil force.
Actual force applied is then (Recoil_Force - Counter_Force). If this value is positive, the gun will indeed pull more, moving the muzzle off the target. If the value is negative, the muzzle moves in the opposite direction, possibly ending up on the other side of the target if it's TOO negative. The optimal result is of course 0, where the muzzle doesn't move at all.
The gun's Recoil Force never changes - it's always pulling with the same force in the same direction. However, the shooter does have the opportunity to adjust the amount of force he is applying, and this is the basis for everything else.
For the first X bullets of the volley, the shooter's force is static. If he randomly applied too much or too little counter-force, then the accumulated difference from those X bullets being fired would have moved the muzzle off the target center. Once these X bullets have been fired, the shooter gets the opportunity to adjust his force to try and eliminate this extra muzzle distance. Again, he may increase his applied force by a random amount, aiming generally to achieve a force that would move the muzzle back towards the center.
This repeats every X bullets. Each time, depending on the difference between Force and Counter-Force, the muzzle moves roughly in the general area of the target. The more Force Accuracy the shooter has, the less the muzzle moves. Contrarily, a shooter with less Force Accuracy will have the muzzle swing around back and forth over the target, spraying bullets more randomly.
The value of X (the number of bullets between force reassessment) is dictated primarily by the shooter's skills (esp. autofire skill).
Here's a quick example:
Recoil Force = 5, gun pulls up and to the right equally.
Max Counter Force = 7, the shooter can't pull backwards more than this much.
Force Accuracy = 2.0, the shooter will compensate within 2 force points of the optimal force required.
First shot is fired with no recoil.
The next three shots will be fired with a recoil of 5. To compensate for this, the shooter needs to apply a counter-force of 5.
The shooter rolls his first Counter-Force, and comes up with 3.5. This is 1.5 points less force than the gun is exerting, and so is sub-optimal, or "under-compensation".
The second, third and fourth bullets are each fired with a recoil of 5-3.5 = 1.5 points. After those tree bullets, this puts the muzzle 4.5 points (1.5 * 3) away from the target center.
The shooter gets a chance to adjust his aim. Optimally, he would need 5 + 1.5 = 6.5 to compensate for the muzzle being off the target. He rolls against this, coming up with a value of 7, his Maximum.
The next three bullets are fired with -2.0 recoil (5 minus 7). Within three bullets, the muzzle has moved -6 points from its last position. Since the muzzle at was 4.5 points off the target, it is now 1.5 points from the target in the opposite direction. The volley has "raked" across the target.
Now the shooter can adjust his force again, and since the muzzle is pointing down and left of the target, he can relax his force. So when the next chance comes along, he'll try to reduce his counter-force to 4.5. This process repeats over and over until the volley is over.
This is just one example of course, and the variables in this example are all open for debate. But the general idea, as far as I see it, is sound.
Please comment.
Report message to a moderator
|
|
|
|
Re: New CTH system - Presentation[message #250886]
|
Mon, 03 May 2010 00:48
|
|
CptMoore |
|
Messages:224
Registered:March 2009 |
|
|
Better Naming for more intuitive understanding:
Force Accuracy -> Counter Force Accuracy
Missing variables:
Counter Recoil Per Shots = 3 (someone with more agility or dexterity might be able to adjust more often because of hightenes reflexes)
My Suggestion for stats binding:
Recoil Force Binding: AutoPenalty Burstpenalty Bipod
Bipod reduces recoil flat.
Max Counter Force Binding: Strength Strength*(Foregrip + "Auto Weapons" + Shooting + Exp)
Better grip, better force applied. Low strength makes you still suck even with skills.
Counter Force Accuracy Binding: Dexterity Bipod Dexterity*(Foregrip + "Auto Weapons" + Shooting + Exp)
Bipod ups accuracy very good. Dexterity more important than strength here. Foregrip wont help accuracy without dexterity.
Counter Force Per Shots Binding: Dexterity "RPM of Weapon"
Dexterity helps reaction time. Higher RPM works against dexterity.
I like it.
Report message to a moderator
|
Sergeant 1st Class
|
|
|
|
|
|
Re: New CTH system - Presentation[message #250911]
|
Mon, 03 May 2010 19:04
|
|
silversurfer |
|
Messages:2793
Registered:May 2009 |
|
|
HeadrockI agree, the robot should be unable to use scopes, as you said, and also a serious penalty for moving targets. So the robot would be pretty weak at long range or dynamic combat, but can be a serious help taking care of hiding enemies in close quarters - just open the door, take a hit or two, and autofire the crap out of them. And yeah, since the robot is on tracks, and probably weighs a lot more than a person, it would be considered automatically as having both the Expert Heavy Weapons and Expert Auto Weapons skill.
I'm not so sure about the scopes. Madlab used a video camera for optics and those usually have a zoom. It will probably be too complicated to implement an adjustable zoom but how about the player chooses the zoom level by using a scope as an attachement for the bot? This way you would have to decide before the battle if you want a long range scouting machine or a close range killing machine.
A few thoughts about the bot:
1. Base CTH: should be taken from the shooter with a penalty for aiming through a small display
2. CTH Cap: should be taken from the shooter with a penalty for aiming through a small display
3. Aiming CTH: should be taken from the shooter but influence of injuries should be taken from the bot (damaged) and higher penalty for moving targets
4. Muzzle Sway: should be zero since the gun is installed in a fixed position
5. Recoil: should be smaller than for a person but not zero (this is not a tank)
About the tank:
1. Base CTH: should be smaller than for a normal person (its harder to align the turret to a small target than it is to align a hand gun)
2. CTH Cap: should be smaller than for a normal person (its harder to align the turret to a small target than it is to align a hand gun)
3. Aiming CTH: high penalty for moving targets, shouldn't be allowed to fire at targets below a certain range (min radius)
4. Muzzle Sway: should be zero since the guns are installed in a fixed position
5. Recoil: should be zero for machine gun and a normal value for main gun
Report message to a moderator
|
|
|
|
|
|
Re: New CTH system - Presentation[message #250956]
|
Tue, 04 May 2010 17:10
|
|
Headrock |
|
Messages:1760
Registered:March 2006 Location: Jerusalem |
|
|
Quote:how about the player chooses the zoom level by using a scope as an attachement for the bot?
IIRC, you can't open the Robot's inventory.
Quote:1. Base CTH: should be taken from the shooter with a penalty for aiming through a small display
2. CTH Cap: should be taken from the shooter with a penalty for aiming through a small display
3. Aiming CTH: should be taken from the shooter but influence of injuries should be taken from the bot (damaged) and higher penalty for moving targets
4. Muzzle Sway: should be zero since the gun is installed in a fixed position
Muzzle Sway is the result of Base CTH+Aiming CTH no greater than CTH Cap. So if Muzzle Sway is zero, the others are, by definition, also zero.
Quote:5. Recoil: should be zero for machine gun and a normal value for main gun
Since the main gun can only fire single shots, recoil is irrelevant. But for the machine gun I guess 0 recoil would be possible. Of course that would make that MG very very powerful.
Quote:i think [dexterity] should be factored in to the whole equation to some degree. otherwise strength which is pretty important allready would become even more so.
Dex is factored into most of the actions in the game, so whether you realize it or not it is already very important. Strength less so, actually. Regardless, DEX will be used for force accuracy, and STR for max force applied.
Quote:
about guesstimaating direction and force of muzzle climb, this should take wisdom into account.
As Warmsteel said, WIS is not going to be very important for force accuracy, as it happens too quickly and is more intuitive (by feel), hence DEX, AGI, and EXP Level. WIS will be more important for Drop Compensation (shooting beyond the gun's range), which I have yet to get around to, and Movement Tracking.
-----------------------------------------------
UPDATE
The new system is going to open up a lot of possibilities regarding the accurate modeling of attachments, as I mentioned earlier. Items are now going to be able to change each part of the shooting formula separately.
The following tags will be added for use by attachments:
- Increases Base CTH by a flat amount, BEFORE taking other factors into account (like injury or morale).
- Added to the sum of percentage-based factors after Base CTH has been calculated. So -5 here means we lose 5% of our base CTH after it's been calculated.
- Increases or decreases a character's CTH cap by percentage.
- Added to the sum of percentage-based factors after Aiming CTH has been calculated. So +10 here means we gain 10% of our Aiming CTH after it's been calculated.
- Already implemented, this causes the maximum muzzle sway to become smaller based on distance. 2 here means that at 2x normal distance, we are as accurate as we would be at normal distance without the attachment. Magnification suffers below its optimal distance, giving penalties instead.
- ,
- Increases or decreases the recoil of the gun, using X and Y as percentages, or possibly as flat modifiers. Not sure yet which is better.
- Increases or decreases the character's recoil accuracy by percentage. +10 means the soldier is 10% more accurate in applying counter-force.
- Increases or decreases the character's maximum recoil counter-force by percentage. -50 means the character can only apply half as much counter-force as normal.
- Increases or decreases the gun's maximum bullet deviation by percentage. -20 makes the gun 20% more accurate (but not the shooter).
- Increases accuracy when trying to fire upwards to compensate for shooting beyond range. This is unlike a flat range bonus (which will still be used for things like match bullets or extended barrel) as it doesn't increase the range but allows shooters to compensate better for it.
These modifiers will replace some of the existing modifiers that no longer have any effect in the system, such as "To-Hit Bonus".
Naturally, we'll need to rethink attachments to see what tags they should have.
Also, things like bipods will need to be reconsidered. In this system, either I keep a "" tag that has its own complex effect and is only applied when prone, or I make a new "" tag that means the attachment as a whole is only active when prone. If the original setup is kept, bipods will remain hardcoded and difficult to mod (giving a set number of effects based on one value). With the second system, grippod-type attachments will become impossible since they won't be useable when standing up.
Ideas, suggestions, etcetera, are welcome as always.
Report message to a moderator
|
|
|
|
Goto Forum:
Current Time: Wed May 15 06:00:24 GMT+3 2024
Total time taken to generate the page: 0.03533 seconds
|