Realistic Weapons Modification Thread - L.U.R.K. mod for S.T.A.L.K.E.R. Shadow of Chernobyl

I've noticed alot of people complaining about the extreme recoil of some of the weapons in the forums, and after playing for a bit I too decided it needed a little tweaking. Not only was I unhappy with the recoil, but I also had some issues with the accuracy and damage comparisons between the weapons. After many hours of messing around with the .ltx's I found the perfect set of parameters to balance out the weapons.

Before I begin, I want it to be known that I own both a WASR-10 (Romanian AK-47 clone) and a Bushmaster M4 A3
(among various other guns) and have fired both on full auto with the help of a bumpfire trigger. Needless to say while it is hard to accurately control prolonged automatic fire, it's no where near as jumpy as depicted in this mod. Most modern assault rifles have specially designed muzzle-breaks intended to vent the escaping gas in a way that stops the barrel from climbing. In fact, the AKM and my WASR had a canted muzzle break that would actually force the front of the rifle down if fired too fast. Oh, and the M4 has such little recoil it's not even worth mentioning.
After hours of messing around with the values in the weapons directory, I think I've found a fairly accurate system for realistic weapon balancing and recoil control. Most of this is based off the caliber, the length of the barrel, the stock (and other ergonomics), and the overall quality of the weapon design (ie. a rifle made by H&K has better overall recoil-control and accuracy than the cheap Soviet era crap spewed out by Russia).
In the "gamedata/config/weapons" folder, select the weapon you want to edit. For me I have to either copy/paste the file to desktop and move it back later, or run notepad as administrator for these changes to actually be saved. Now first, for the damage and dispersion values, scroll down to where it says
";-----Ballistics, Handling & Accuracy-----; "
under here you should see "hit_power " and further down "fire_dispersion_base ". This is your damage and base accuracy. I had to completely change these values for most of the weapons, they just seemed off. Note: if you change the weapon and the ammo stats too much it will really throw of the balancing (I just did weapons). Now, if you go further down you get to these lines:

cam_relax_speed =

cam_relax_speed_ai =

cam_dispersion =

cam_step_angle_horz =

cam_dispersion_inc =

cam_dispertion_frac =

cam_max_angle =

cam_max_angle_horz =

This is the recoil control area, which affects the jumping of the camera to simulate recoil. Basically, all you need to know is "cam_relax_speed" has to do with how fast your weapon recovers from recoil, but really on makes a difference with semi-fire only. You'll notice the AI's default cam_relax is 360, this is one of the reasons why they are so ungodly accurate. "cam_max_angle" and "cam_max_angle_horz" control how far the camera can actually shift during recoil (assuming to don't move the mouse while firing), they have it set to 180 degrees as default, which means you can actually recoil yourself to look strait up really fast. This is stupid and pointless, so I adjusted those numbers close to what OL has. Now, dispersion, step_angle_horz, dispersion_inc, and dispertion_frac all have to do with how much the gun moves every shot and where it moves. Not knowing exactly which did what, I just futzed around for many hours until I found a good combination. I'm using cam_dispersion as a changing value for the different recoil weapons have, and the other 3 are basically multipliers that stay the same.

To give you a rough description of my value range here is my values for 3 very common weapons in 3 very different calibers:

;AK47;

hit_power = 1.10

fire_dispersion_base = 0.65

cam_relax_speed = 12.5

cam_relax_speed_ai = 25

cam_dispersion = 2.5

cam_step_angle_horz = 1.0

cam_dispersion_inc = 0.075

cam_dispertion_frac = .75

cam_max_angle = 15.0

cam_max_angle_horz = 30.0

;M16A3;

hit_power = 1.00

fire_dispersion_base = 0.35

cam_relax_speed = 20

cam_relax_speed_ai = 40

cam_dispersion = 1.25

cam_step_angle_horz = 1.0

cam_dispersion_inc = 0.075

cam_dispertion_frac = .75

cam_max_angle = 10.0

cam_max_angle_horz = 20.0

;MP-5a2;

hit_power = .85

fire_dispersion_base = 0.70

cam_relax_speed = 27.5

cam_relax_speed_ai = 55

cam_dispersion = .75

cam_step_angle_horz = 1.0

cam_dispersion_inc = .075

cam_dispertion_frac = .75

cam_max_angle = 7.5

cam_max_angle_horz = 15.0

I took things into account such as: barrel length, cycling-system, ergonomics, gun weight, even full-stock vs folding/collapsing stock. While alot of these differences are opinionated, you can see that the values change in very tiny amounts, but it makes all the difference in the world. With the AK's hit power it will drop a bandit with 2-3 shots in the chest, which is pretty realistic (to bring them down instantly!). With the dispersion rates, the M4 has almost half the dispersion as the AK, which means at roughly 50 yards, the M4's grouping would be about half the size as the AK's (also realistic based on my experience with both). Now as far as weapons of the same calibers go, I'm tired of seeing something like the M-4 doing 3/4 the damage as a G-36. They're the same freakin' caliber, the same bullet! Some guns achieve different muzzle-velocities because of minor things like longer barrels or crap
like direct-impingement vs. gas-piston, but that doesn't mean the bullet's going to do more damage in the long run (at least not
noticeably). With that in mind, it helps to just keep the hit_power relatively the same for same-caliber weapons (if you have to, just
change it by .02 or something). I didn't really mess with any of the other numbers, except for the Deagle I made the "hit_impulse = 100" and the "control_inertia_factor = 5" in the hopes it would send bodies flying back 20 feet (lol). There were some minor discrepancies with some of the ammo stats, but I wouldn't worry about it.

If you don't want to go through the trouble of changing all the files yourself, and you like the way I did it, I uploaded my weapons folder here:
depositfiles (DOT) com/files/33co1pnt4
just unzip and move into your STALKER/gamedata/config folder and select replace the old weapons folder.

So this system seems to be fairly accurate to me, and it made the recoil much more realistic, but not Cowadooty easy. If you have any questions/opinions I'd be happy to hear them.

Edited by: Ichi_the_Blind

Thanks man, now the gameplay feels much, much better. The recoil was too excesive, I don't know that Holden was thinking when he configured the weapons to behave like that. Was for the sake of the so called balance, was it? Well, struggling with the recoil, even with weapons that kicks softly, isn't challenge or balanced, it renders the weapons useless and makes you think that Strelok is a girl that never shot a weapon before rather than a man, or let alone an experienced STALKER.
And you're right, some weapon stats are messed up.
For example: the CZ 52, it has crappy stats while the TT-33 has great stats. In real life both weapons are pretty much the same: same caliber (7.62x25mm Tokarev), mag capacity, single action trigger, average to poor ergonomics and the differences might be that the TT-33 is slightly more rugged and reliable whereas the CZ 52 is mechanically more accurate* and can whistand hot loads.

This reminds me of Red Orchestra: Ostfront, in where even a MP-40 kicks like a mule, the devs from Tripwire recognized this little mistake and fixed it for the upcoming Red Orchestra: Heroes of Stalingrad.

Again, thanks man.

*The CZ 52 operates with a short recoil, roller locking principle, since the barrel recoils linearly, that's why it's mechanically more accurate than the TT-33, that operates with 1911-style short recoil, swinging link, since the barrel tilts when recoil to unlock the the lugs from the notches machined in the slide, there's a slight dispersion and more play between the parts. Also, the czechs designed the CZ-52 so it can fire M-48 load (normally used on submachineguns) that is 20% more powerful than the original, russian made ammo. The only big cons of the CZ is the brittle firing pin (it can break if dry-fire the pistol) and the rollers, that's why it's less reliable, or rather less durable, than a TT.

Edited by: Subway2033

I messed around with these settings back in the day (read: back when 1.1 was first released) and found that I could only reduce the recoil on semi-automatic mode. Did you ever find a way to control recoil on other fire modes?

Also, not trying to bash your gun, but Romanians are POSes IMO.. I'd only use a real one if my life depended on it, and even then I'd toss it as soon as I got something better. My Yugoslavian M70B (Yugo improvement of the AKM) can actually get very accurate with the right ammo, a la Hornaday V-Match, on the contrary to all the hype about how an AK can't hit the broad side of a barn at 25 yards. Funny how most of the "tests" that were done comparing an AK to an M16 were using outdated mil-surp M43 7.62x39 compared to some of the more advanced 5.56x45 rounds you can find these days.

Edited by: Malikyte

Just reading over this now, it does look pretty well thought out as well as configured to be balanced, however when I had my experiences with Holden, he specifically pointed out one thing in that entire line.

cam_relax_speed =

That variable directly attributes to reducing recoil, however its more of an auto compensate feature rather than straight up reducing recoil, which makes the factor of recoil "pointless". I suppose in other words, you may as well just half the recoil instead of adding auto compensate, because it gives the gun an unrealistic feel when firing.
Anyways I'm not trying to bash your work. You did a very good job with these scripts, and should be treated as such. Furthermore, let me explain what the main variables do...

cam_relax_speed = Auto Compensationcam_relax_speed_ai = AI Auto Compensation
cam_max_angle = Maximum Distance the Camera is allowed to move up while firing. 180* means that if you fire your weapon for to long without moving your mouse down to compensate, you'll end up looking at the sky. I do agree 180* is a little to much, but 15* is just unbearable.Once again, good work on figuring it out, but apparently Pazer Draco has determined an algorithm for configuring recoil.

TehCourtJester wrote: Just reading over this now, it does look pretty well thought out as well as configured to be balanced, however when I had my experiences with Holden, he specifically pointed out one thing in that entire line.

cam_relax_speed =

That variable directly attributes to reducing recoil, however its more of an auto compensate feature rather than straight up reducing recoil, which makes the factor of recoil "pointless". I suppose in other words, you may as well just half the recoil instead of adding auto compensate, because it gives the gun an unrealistic feel when firing.
Anyways I'm not trying to bash your work. You did a very good job with these scripts, and should be treated as such. Furthermore, let me explain what the main variables do...

c

I changed some of these as well. I disagree about the pointless recoil thing. I have been shooting for years, and in a proper stance, after recoil the weapon does indeed return to point of aim by itself. I disagree with the Mak stats as well, I guess you guys did it for the "starter" pistol, but a real Makarov is very reliable and accurate. The Mosin rifles are very reliable as well.

I felt that most of the weapons in LURK have unrealistically high recoil, also, the scope sway thing was annoying and unrealistic. When I am in a shooting stance, I can hold the crosshairs pretty steady on an actual rifle. Other than these problems, which are easily fixed(modding reliability in .ltx files, ect) the mod is great, and I play it regularly. I havent figured out how to mod the ballistics yet, other than bullet speed. I can mod reliability, handling, and accuracy.