Home » MODDING HQ 1.13 » v1.13 General Development Talk » (Potential) NCTH Code Change Discussion
Re: (Potential) NCTH Code Change Discussion[message #323260]
|
Fri, 26 July 2013 08:50
|
|
mmm |
|
Messages:63
Registered:May 2013 |
|
|
Then it seem that making weapon related AP consumption(ready, aim, attack, or even switch target) stance dependent will improve the realism. If not, it will at least offer some options.
The modifiers can be described as take away/add a fixed amount of angular error from/to the shots. They don't work well everywhere, one example would be stability related parameters. Some may be better off to take a percentage modification. Scopes basically work this way, which is ill suited.
The other thing is that NCTH is based on spread radius. That is 70% accurate is 30% spread radius, 90% accurate is 10% spread radius. When the CTH goes up, the difference in final hit probability is just immense. Were it done based on area rather than diameter, it would have been easier to balance. Also it will give the less competent shooters at least some chance.
NCTH was probably pushing too hard to make sure that you have a chance to miss even at close range.
Report message to a moderator
|
Corporal
|
|
|
|
Re: (Potential) NCTH Code Change Discussion[message #323264]
|
Fri, 26 July 2013 12:53
|
|
mmm |
|
Messages:63
Registered:May 2013 |
|
|
I guess the "proper attachment" means scope, shrinking the spread circle to 1/x is powerful indeed. But I'd better not go there again, it bores everybody out... Actually what I wanted to say is the difference between good and very good is a bit too great. For example, 70% and 90% CTH in area rather than radius would be roughly 3 time hit probability difference. Significant, but realistic amount of difference.
The possibility to miss(but also not impossible to hit) at any distance is presumably the thing that keeps people happy, it's the balance everyone talks about. I'd personally prefer a CTH system with a sound logic, even if it means exponentially increased CTH at closer range. That's realistic too because naturally short range fire fight is more dangerous, managing distance is of great importance. Of course the OCTH is the least preferred.
Stance need tweaking. But I'm afraid the existing modifiers won't suffice. I'd accept limited usefulness of certain weapons.
Oh, one more thing. Targets in prone position aren't really as small as they should be.
[Updated on: Fri, 26 July 2013 14:47] by Moderator Report message to a moderator
|
Corporal
|
|
|
|
|
|
Re: (Potential) NCTH Code Change Discussion[message #323286]
|
Sat, 27 July 2013 02:09
|
|
mmm |
|
Messages:63
Registered:May 2013 |
|
|
Well, the significance of handling is greatly affected by the magnification. I'll basically just repeat myself, to make sure I made myself clear.
Yes, handling affects the CTH, base and cap. The "CTH" number is basically the percentage of radius you shaved off from the outer aperture. And the outer aperture is shrunk by the scopemagfactor. For the same amount of CTH difference produced by handling, it's impact on accuracy is really proportionally reduced by magnification.
Quote:At 23 tiles my standing test merc gets 64 CTH vs. 59 or 43,31 aperture size vs. 59,16 with non-pimped weapons.
Didn't quite get the second part. I'll assume it's the size in terms of dispersion area. Again, you're the coder, you have the access to the debug info and the actual formula. I'm just relying on the observation and dumbed down(maybe outdated) version of the formula. You're more than welcome to clear up any of my misconceptions.
Then it didn't exactly match my conjecture((36/41)
[Updated on: Sat, 27 July 2013 12:07] by Moderator Report message to a moderator
|
Corporal
|
|
|
Re: (Potential) NCTH Code Change Discussion[message #323380]
|
Sun, 28 July 2013 20:33
|
|
silversurfer |
|
Messages:2793
Registered:May 2009 |
|
|
Let's assume for a moment that we drop this whole magnification division (no matter where the divisor came from - scope, target or laser pointer). What exactly should scopes and pointers affect and how do we ensure usefulness of scopes and pointers?
Here is a list of what shooting modifiers we have at the moment:
- AP used
- BP used
Initial Target Aquisition: - BaseChance by skills
- BaseChance by fixed bonus from weapon and attachments
- BaseModifier by effects on the shooter (injury, moral, drugs etc.)
- BaseModifier by weapon handling (partly stance specific)
- BaseModifier by target influence (shooting at same target, height difference, invisible target etc.)
Aiming at target: - AimCap by skills
- AimCap by traits
- AimCap by fixed bonus from weapon and attachments
- AimModifier by effects on the shooter (injury, moral, drugs etc.)
- AimModifier by weapon handling (partly stance specific)
- AimModifier by traits
- AimModifier by target influence (height difference, invisible target etc.)
- AimModifier penalty for using a scope below optimal range
- AimBonus per aimclick by fixed bonus
All of the above form muzzle sway, something that somehow represented CTH in the old CTH system.
Here comes modifiers that influence the point where the bullet will go.
- Magnification Factor at target range <- this is what we want to get rid of or at least use it differently
- Muzzle Sway - a random point in our aperture where the bullet will go
- Movement Offset - deviation because of target movement and the try to compensate; according to dinglehopper scope magnification factor should have a big influence on this
- Recoil - only interesting in burst and autofire
- Range Compensation if target is out of gun range
- Bullet Deviation - depends on gun accuracy and target/gun range ratio
- target facing penalty - shooting at a target from the side is more difficult than from front or back because the possible hit area is smaller
I hope I didn't forget anything important.
Report message to a moderator
|
|
|
|
Re: (Potential) NCTH Code Change Discussion[message #323407]
|
Mon, 29 July 2013 05:24
|
|
mmm |
|
Messages:63
Registered:May 2013 |
|
|
uh, actually I prefer none of them. In my vision sights of all forms, be it iron, reflex, telescopic, laser, will be error producing rather than reducing(hopefully that's an easier to understand way to describe it). The difference lies in the amount of error. However the amount is the inherent property of the sight(dot size, reticle thickness+magnification, etc), which is unaffected by the shooter's skill. So it would be like the naccuracy stat... Or maybe a fixed amount of CTH difference if the maximum aperture is constant under all situation. Although I do believe we'll need to use the aperture divisor for something(if not as a replacement for handling), and probably in another way. Otherwise the game will be unplayable. Maybe it's good to consider not to provide the full divisor at low level of aiming?
But I could foresee a few possible problem with such implementation. To let anyone less than prefect to have a reasonable hit probability at sniping distance without raising the overall accuracy too much(DEGREES_MAXIMUM_APERTURE), bipod/resting will have to provide a large stability boost, at least something similar to current 2x~3x scope(but the bonus may be better to be somewhat base weapon independent, a CQBR with bipod is not necessarily much more stable than a heavier sniper rifle, especially consider the extra stable stock, cheek rest, light trigger pull etc). It will be rather powerful, and it will force people to drop to the ground if they want to hit anything(which is realistic). Those who hate to do so will complain. And if the stability required to hit target afar is provided through resting, hitting targets up close will be dramatically easier, since I don't think applying accuracy penalty to closer target is the way to go. It will be reasonable to penalize the speed though, but the options are limited now, the difference will still be somewhat limited too.
[Updated on: Mon, 29 July 2013 19:39] by Moderator Report message to a moderator
|
Corporal
|
|
|
Re: (Potential) NCTH Code Change Discussion[message #323549]
|
Thu, 01 August 2013 23:42
|
|
LootFragg |
|
Messages:349
Registered:August 2009 Location: Berlin, Germany |
|
|
I think most stuff that makes me a bit unhappy about shooting in JA2 is data related, so that doesn't matter here. One thing I would like to see changed code-wise, however, is how aiming APs are calculated. Shooting has originally been a process that takes the same time for Speedy McFast and Crawly McWheelchair. While the former has more APs, he also consumes more APs when firing his gun so both can perform the same regardless of AP difference when not moving but shooting.
However, aiming APs are flat and therefore, a fast merc will have more time to aim than a slow one. This needs to be based on a merc's initial APs to prevent high AGI mercs from outperforming others on the shooting field.
Edit: Also, I have been thinking about the discussion about scopes but I am too tired to understand the details, so I won't contribute yet.
[Updated on: Thu, 01 August 2013 23:43] by Moderator Report message to a moderator
|
|
|
|
Re: (Potential) NCTH Code Change Discussion[message #323556]
|
Fri, 02 August 2013 11:08
|
|
silversurfer |
|
Messages:2793
Registered:May 2009 |
|
|
I was wondering - since there is probably nobody willing/able to do a complete rework of NCTH and incorporate some of the suggestions made here, how about easing the pain a little with iron sight shooting? Some say that scoped shooting is overpowered compared to iron sights. I won't agree that scoped shooting is overpowered, for me it works fine with stock 1.13. What I have to agree is that iron sight shooting sucks. So instead of making scopes less useful how about making iron sights more useful?
I'm already working on a percentage bonus for the shooting aperture when iron sights are used. It's a bit complicated because there is no "iron sight" in game. We only have magnification 1x that we use as iron sight. Unfortunately this also applies to shooting from hip. So if I modify the shooting aperture it will make iron sight and hip shooting more powerful. Anybody thinks that this will be a problem?
Another question is whether to modify the basic aperture or the final aperture. I'm leaning towards basic aperture at the moment. It's like modifying DEGREES_MAXIMUM_APERTURE for iron sights. I want to limit the values of the modifier from -50 to 50 percent.
-50 means that you will double your aperture for iron sight shots. Nobody probably wants that but we need to have a lower limit.
50 means half the aperture making iron sights basically as powerful as a 2x scope.
The default will be 0 which means no change.
I still have to figure out how to use lasers. Lasers have a magnification factor just like scopes (actually projection factor but it is used like magnification). You could end up with more accurate shots with iron sights only vs. iron sights + laser. That's not what I want. So I guess that I will have to introduce another modifier for lasers just like the one for iron sights. Laser projection factor will then only be used to calculate the effective range of the laser. I could make light level influence that... We'll see.
Do people use lasers for scoped shooting? I guess not. So I will probably make lasers only apply to shots that are done with magnification 1x. That means iron sights, shooting from hip and using reflex sights. Makes sense?
LootFraggI think most stuff that makes me a bit unhappy about shooting in JA2 is data related, so that doesn't matter here. One thing I would like to see changed code-wise, however, is how aiming APs are calculated. Shooting has originally been a process that takes the same time for Speedy McFast and Crawly McWheelchair. While the former has more APs, he also consumes more APs when firing his gun so both can perform the same regardless of AP difference when not moving but shooting.
However, aiming APs are flat and therefore, a fast merc will have more time to aim than a slow one.
This has probably been discussed a million times already. As far as I understand the concept of game turns everybody has exactly the same time available. It doesn't matter if merc A or merc B shoots a gun, the percentage of time used is the same for both. The question is - can the more nimble merc aim faster? I guess not. Better shooters already outperform bad shooters. They don't need to apply so many aim clicks to achieve the same hit probability. This makes them "faster" shooters because they don't need to spend so much time aiming.
Report message to a moderator
|
|
|
|
|
Re: (Potential) NCTH Code Change Discussion[message #323560]
|
Fri, 02 August 2013 12:33
|
|
DepressivesBrot |
|
Messages:3658
Registered:July 2009 |
|
|
GambigobillaI think problem with NTCH is every so often a guy tried to modify NTCH code but Headrock came and said "GTFO, my code!". We have not much coders, most of them doesn't want to mess with the NTCH code and most of the remaining is downed by Headrock. But it's about time to change things. Headrock is no more and if he ever comes back punch him in the face, put in an oil barrel, and dump in the ocean. So i'd say if you have ideas about the code don't hesitate, do it as you please, yet more go medieval! You can't break which is already broken. We have a saying: If you don't know what you're talking about, just say nothing for once.
What you are referring to is one incident where a coder who, while having done a lot of good for the 1.13 project, was also notorious for "improving" other people's code made some hackjob "improvements" to NCTH while merging it into the trunk. Afair the stuff went against the concept and was largely redundant to existing modifiers so it was reasonable to revert it with HAM5.
Report message to a moderator
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Re: (Potential) NCTH Code Change Discussion[message #323750]
|
Wed, 07 August 2013 17:27
|
|
LootFragg |
|
Messages:349
Registered:August 2009 Location: Berlin, Germany |
|
|
silversurferLootFraggI think most stuff that makes me a bit unhappy about shooting in JA2 is data related, so that doesn't matter here. One thing I would like to see changed code-wise, however, is how aiming APs are calculated. Shooting has originally been a process that takes the same time for Speedy McFast and Crawly McWheelchair. While the former has more APs, he also consumes more APs when firing his gun so both can perform the same regardless of AP difference when not moving but shooting.
However, aiming APs are flat and therefore, a fast merc will have more time to aim than a slow one.
This has probably been discussed a million times already. As far as I understand the concept of game turns everybody has exactly the same time available. It doesn't matter if merc A or merc B shoots a gun, the percentage of time used is the same for both. The question is - can the more nimble merc aim faster? I guess not. Better shooters already outperform bad shooters. They don't need to apply so many aim clicks to achieve the same hit probability. This makes them "faster" shooters because they don't need to spend so much time aiming.
As far as I understood it, you said the same thing I said. Running speed has nothing to do with shooting, hence all mercs, regardless of APs, can place the same amount of shots, spend the same time handling their guns. Right? Right. Original concept.
We're not talking about the quality of the shooter, both shooters are exactly alike, only difference is that one is physically fit to run a Marathon and the other one has a leg injury. Difference in AGI only which influences APs. More AGI, more APs, more tiles to run since running costs flat APs. More APs, more shooting cost, because shooting is scaled. This leads to the AGI bonus being ruled out in shooting, everyone shoots the same on snapshots.
Last time I checked (and that's been some time ago, admittedly), aiming APs per click weren't calculated as a percentage of the maximum APs but had a fixed value per aiming click. You can correct me on this one if it doesn't apply anymore. If it still holds true, that means aiming, as opposed to shooting, consumes flat APs (like running does), not dependent on your maximum APs.
And that means, fast runners who have plenty of APs can aim more often than slow mercs with limited APs. Unless of course this has already been revised.
Report message to a moderator
|
|
|
|
Re: (Potential) NCTH Code Change Discussion[message #323752]
|
Wed, 07 August 2013 17:47
|
|
LootFragg |
|
Messages:349
Registered:August 2009 Location: Berlin, Germany |
|
|
By the way, as for distances in the game, I agree with pheloncab. We need to accept unrealistic proportions in JA2 because the game, due to the way it is designed, CANNOT POSSIBLY be turned into a shooting sim. So while it has to be intuitive and kind of balanced, the actual ranges here are relatively unimportant. Judging from the size of the characters, pistols and grenades have a pretty realistic range but everything else is scaled down a lot to fit on the map. Neither 400 tile firefights with 7.62 nor 3 tile firefights with pistols would make much sense as far as gameplay is concerned.
Towards the laser debate, I like the way you're handling this and making lasers a short-range snapshot helper and that light influences the effectiveness. From my personal experience, when you're using a laser module, you're not using sights. You may glance over the sights to align the gun with your sight line but the laser dot already shows you the cross section of the target with the aiming line which is what you'd find out using the sights as well. I don't know if there are cases of low light encounters where it would be useful to combine scope and laser sight but generally, since the laser is adjusted for a small distance to target, hence producing aiming error at larger distances and since your (possibly illuminated) reticle already shows you your aiming line, there is no point in combining them from my point of view. Genuinely interested in how this'll turn out. The most basic approach of limiting the basic aperture sounds like the best one to me.
Report message to a moderator
|
|
|
|
|
Re: (Potential) NCTH Code Change Discussion[message #323778]
|
Thu, 08 August 2013 13:53
|
|
LootFragg |
|
Messages:349
Registered:August 2009 Location: Berlin, Germany |
|
|
Yes, you instinctively try to avoid that issue because good mercs are good in all stats and bad mercs are just, well. Bad. But there are those slow-as-crap mercs which aren't overly bad in everything else and I personally noticed it when playing with Bull. Bull is among my favourites but he is very slow. So slow indeed that he usually doesn't even reach the firefight before it's over. And when he does, he is really bad with guns as he can neither aim properly nor fire as many bullets as other mercenaries because auto-fire does the same thing and allows for longer bursts only on high AGI mercs.
You end up giving them fast, cheap snapshot guns without autofire or aim requirements. This also counts for old mercs who are still excellent shots apart from not being able to move in line with the young ones. It makes no sense. I believe I've talked about this before once and suggested that AP costs for aiming and bursts (and other shooting related stuff) be calculated as a float value that gets rounded for use in the game. I have forgotten why this was supposed to be complicated. Yet this is among the primary coding requests I'd have, making temp float values for burst & aim AP calculation and making standard turn APs a factor.
But this, together with changing the diagonal movement AP cost modifier towards SqR(2) so that diagonal movement costs only +50% (or +41.421356237%) APs instead of +100%, making movement cost the same for a fixed distance in any direction, counts among the more complicated issues, probably because it affects AI calculations in every way. A shame that what holds us back is ultimately the botched AI.
Report message to a moderator
|
|
|
|
Re: (Potential) NCTH Code Change Discussion[message #323807]
|
Fri, 09 August 2013 00:38
|
|
Flugente |
|
Messages:3509
Registered:April 2009 Location: Germany |
|
|
AP costs are, while not as much all over the place as other stuff, still rather untidy. Should you decide to alter the behaviour LootFragg suggested, it might be good to wrap them up in a simple function that also takes a SOLDIERTYPE* as an argument. That way any modification of AP costs depending on max APs can be done just once, while cleaning up the code as well.
However, it would first be necessary to decide which actions should scale and which shouldn't. For example, running should not scale - if you have more AP, you can run longer, as you are quicker/more agile. Aim clicks, as LootFragg pointed out, should scale, as there is no reason why a 80-year old should take longer aiming than a 20-year old if they are equal at shooting.
Report message to a moderator
|
|
|
|
|
|
|
Re: (Potential) NCTH Code Change Discussion[message #324437]
|
Thu, 29 August 2013 10:48
|
|
dzidek1983 |
|
Messages:92
Registered:June 2009 |
|
|
my opinnion after testing NCTH yesterday on 4870
Buzz missed a guy with a 10 bullet aimed burst from 5 tiles in Drassen
she missed all shots
ice being an auto weapon expert also had some problems
i gave them assault weapons from higher level mercs, they were healthy and rested
weapons even had the 2x scope
Buzz from like 5 tiles missed all 10 bullets they just all curved right from the target like there was som serious wind blowing she was shotting north so i could see exactly how they curve
she aimed perfectly in the middle, used the mouse scroll, so the inner circle was perfectly on the enemy torso
similar thing with ice, i aimed nicely and did a 5 bullet auto fire
now this is trickier... once he hit all 5 bullets and the second time all 5 flew in differnt directions around the enemy like he was holdinfg a 200 lb cannon and the weapon took over control over the shooters ability (strength) to hold it in line
Report message to a moderator
|
Corporal 1st Class
|
|
|
|
Re: (Potential) NCTH Code Change Discussion[message #324439]
|
Thu, 29 August 2013 12:28
|
|
dzidek1983 |
|
Messages:92
Registered:June 2009 |
|
|
the problem is you dont decide at witch distance the enemies pop up...
and franly the system works a bit to "linear" in my opinion cos when a bad dude pops around the corner a experienced soldier wont look at him through the scope as it would take to long and would, as you mentioned, be a hindrance...
i think when you take unaimed shots at point black (or in this case any unaimed shot) the game should never calculate optics bonuses/negatives "to-hit"
it would resemble you just poping a burst at a large target close
all those scopes, lasers, etc, should never be added to a non aimed shot
so Ice and Buzz at 5 tiles range would only depend on base weapon stats and own skills
Report message to a moderator
|
Corporal 1st Class
|
|
|
|
Re: (Potential) NCTH Code Change Discussion[message #324441]
|
Thu, 29 August 2013 13:00
|
|
dzidek1983 |
|
Messages:92
Registered:June 2009 |
|
|
ok so if i did unaimed shots they wouldnt be counted in and the scope would not give negative values?
Report message to a moderator
|
Corporal 1st Class
|
|
|
|
|
|
Goto Forum:
Current Time: Thu Mar 28 10:18:25 GMT+2 2024
Total time taken to generate the page: 0.03784 seconds
|