Author Topic: Is maximum range even a sensible attribute within current ranged weapons system?  (Read 2132 times)

Offline Tamior

  • Zombie Food
  • *
  • Posts: 78
    • View Profile
Tarburst is correct about the underlying rationale, the max range is an attempt to model different guns having different effective ranges based on velocity loss due to aur resistance.

I'd argue that the way it works now creates more problems than it solves.
1) Fixed max range prevents you from AIMING at targets out of it even if you just want to steady the shot and fire when the target IS within range.
2) Reduced accuracy at range (due to dispersion) already creates something very similar to velocity drop (i.e. average damage is reduced with range), fixed max range just literally introduces a situation where projectiles can go from full energy and 0 energy within a single map tile.
3) Do we even need to model velocity drop given that reality bubble is something like 50-100 meters in radius?

I mean, thing about it this way:
 - for a combination of weapon and skill where you can't really hit at max range removing that max range will not make it viable at longer ranges — dispersion already limits your effective range before max range does
 - for a combination of weapon and skill where you CAN really hit at max range — max range works not as a "velocity drop" thing, but as a magic wall that completely prevents you from targeting and firing once your target crosses it even by an inch.

Does this need to be a hard cap on range? No, it could be an effect that causes accuracy and damage to drop past the effective range, but as Coolthulu pointed out, we need some effect to handle the situation where a gun is extremely accurate at short ranges, but loses effectiveness at range for various aerodynamic reasons.
Is dispersion mechanic not exactly that? An effect that makes various weapons lose effectiveness with range at different rate?
« Last Edit: July 08, 2017, 08:32:55 AM by Tamior »

Offline tarburst98

  • NPC
  • **
  • Posts: 212
    • View Profile
the other problem is you can, with the right setup, have 0 dispersion. which if there is no max range lets you shoot literally forever.

Offline Tamior

  • Zombie Food
  • *
  • Posts: 78
    • View Profile
the other problem is you can, with the right setup, have 0 dispersion. which if there is no max range lets you shoot literally forever.
You are still limited to your reality bubble and line of sight.
And many weapons already have max range longer than visible reality bubble.

I mean, heavy rail rifle is already auto-learn recipe with 60+ max range.

And, again, if you have a 0 dispersion weapon the "magic wall of max range" makes it WORSE since now an inch of extra distance makes a sure-fire hit suddenly impossible to even fire.

P.S. That being said, no weapon should be allowed to have 0 dispersion if you ask me.

P.P.S. Can you give a couple of examples of these "0 dispersion weapons"?
« Last Edit: July 08, 2017, 09:52:40 AM by Tamior »

Offline tarburst98

  • NPC
  • **
  • Posts: 212
    • View Profile
Barrett rifle with .50 BMG

Offline Tamior

  • Zombie Food
  • *
  • Posts: 78
    • View Profile
Barrett rifle with .50 BMG
A weapon with 45+ max range that would not be effected by abolishing max range much, if at all.
« Last Edit: July 08, 2017, 01:30:08 PM by Tamior »

Offline Coolthulhu

  • Contributor
  • Survivor
  • ****
  • Posts: 3753
    • View Profile
P.S. That being said, no weapon should be allowed to have 0 dispersion if you ask me.

P.P.S. Can you give a couple of examples of these "0 dispersion weapons"?

Example: Accurized mossberg 500 with ergonomic grip. Currently 6 max range with 00 shot.

The problem isn't just strictly 0 dispersion weapons. It's that dispersion drops linearly, meaning dropping it from 1100 to 1000 is a smaller change than dropping it from 100 to 80.
Weapons with 15, 30 etc. dispersion are also amazingly long-ranged, just not accurate enough to circumcise a fly.

3) Do we even need to model velocity drop given that reality bubble is something like 50-100 meters in radius?

No, reality bubble is not 100 meters. It's ~60-70 TILES.
Tiles aren't meters. They can't be translated to meters without context. Their shape changes depending on what they need to represent right now.
If you want to translate it to meters, you will have to find the distance at which a healthy adult person can no longer see a car-sized object, assuming fully clear line of sight, clear weather conditions, flat surface and so on. That range will be 61 tiles - at 60 tiles you would still see the car-sized object.
That's the "long range tile" size, one used for sight and firearm ranges.

Offline Tamior

  • Zombie Food
  • *
  • Posts: 78
    • View Profile
The problem isn't just strictly 0 dispersion weapons. It's that dispersion drops linearly, meaning dropping it from 1100 to 1000 is a smaller change than dropping it from 100 to 80.
Weapons with 15, 30 etc. dispersion are also amazingly long-ranged, just not accurate enough to circumcise a fly.
How exactly is confidence of a shot calculated?
Is there a "base" (weapon-specific) value for a point-blank shot that is then decreased at longer ranges depending on weapon dispersion?
or
Is a "base" value for a point-blank shot also a function of the same weapon dispersion stat, not a different weapon characteristic?

No, reality bubble is not 100 meters. It's ~60-70 TILES.
Tiles aren't meters. They can't be translated to meters without context. Their shape changes depending on what they need to represent right now.
If you want to translate it to meters, you will have to find the distance at which a healthy adult person can no longer see a car-sized object, assuming fully clear line of sight, clear weather conditions, flat surface and so on. That range will be 61 tiles - at 60 tiles you would still see the car-sized object.
That's the "long range tile" size, one used for sight and firearm ranges.
We are hardly going to have all-round realistic game-mechanics if tiles represent vastly different range depending on context.
If 5-6 tiles are both a length of a car and a range where shotgun blast loses all kinetic energy, we already had to sacrifice realism/common sense somewhere. Going back to modeling velocity drop in this situation might just make the system even less sensible.

Offline tarburst98

  • NPC
  • **
  • Posts: 212
    • View Profile
how about, i personally prefer having max ranges so it FORCES you to train for different ranged weapons instead of just getting to level 100 with pistols and exploding hulks from the other side of the map.

Offline Tamior

  • Zombie Food
  • *
  • Posts: 78
    • View Profile
how about, i personally prefer having max ranges so it FORCES you to train for different ranged weapons instead of just getting to level 100 with pistols and exploding hulks from the other side of the map.
You would still have the same if weapon dispersion naturally limited different weapon to different effective ranges without an actual hard cap on max range.
Plus, let's be honest here: rifles right now have options for ALL ranges. So you still only need to train rifles if you go for pure practicality.

P.S. Can you give me an example of a handgun you would use over a rifle for longer ranges if both had no hard range limits?
« Last Edit: July 08, 2017, 08:20:27 PM by Tamior »

Offline Coolthulhu

  • Contributor
  • Survivor
  • ****
  • Posts: 3753
    • View Profile
How exactly is confidence of a shot calculated?
Is there a "base" (weapon-specific) value for a point-blank shot that is then decreased at longer ranges depending on weapon dispersion?
or
Is a "base" value for a point-blank shot also a function of the same weapon dispersion stat, not a different weapon characteristic?

Confidence is approximated actual hit chance. Don't trust it, the approximation is not good - just because confidence says you get x% chance to hit, doesn't mean the actual chance to hit will be that high/low.
Dispersion is currently calculated by summing up random rolls for weapon dispersion, skill dispersion, sight dispersion, recoil and possibly something more I forgot. Those can't be exchanged - if you have 50 weapon dispersion and 50 skill dispersion, you will not be as good at sniping as if you had 100 skill dispersion and 0 weapon dispersion. This is a major problem with the system, because it means you can't compensate for things, you need all parts of the dispersion roll near-perfect to actually hit things.

Once the dispersion value is rolled, it becomes an angle of miss, in minutes of degrees. So if you roll 60 on dispersion roll, you miss by 1 degree. This doesn't mean you miss the target totally, just its head.
From this miss angle, it is calculated (trigonometry) where you actually hit. If this hit is still within creature's size from the center, you hit the creature. If not, your shot goes wild, but you can still hit something else.
Whether you get a headshot, good shot, etc. depends on how little you missed, and on target's size. If you roll 0 on dispersion roll, it's always a headshot (if the target has a head).

After the dispersion roll, it's all just raw trigonometry. Point blank shots are simply covered by a wider angle.

EDIT:
Quote
P.S. Can you give me an example of a handgun you would use over a rifle for longer ranges if both had no hard range limits?

Honestly it's less about pistols, which are currently all either high recoil or sci-fi, and more about pistol caliber smgs/rifles, shotguns, pneumatic rifles and so on.
Shotguns are the worst offender because they are all very low dispersion and 00 shot is 0 dispersion. If there was no max range, consistent shotgun headshots would "only" require grinding up the skill.
« Last Edit: July 08, 2017, 10:13:38 PM by Coolthulhu »

Offline Mantar

  • Zombie Food
  • *
  • Posts: 80
  • NOMAD
    • View Profile
If you want to translate it to meters, you will have to find the distance at which a healthy adult person can no longer see a car-sized object, assuming fully clear line of sight, clear weather conditions, flat surface and so on. That range will be 61 tiles - at 60 tiles you would still see the car-sized object.
That's the "long range tile" size, one used for sight and firearm ranges.

That's working backwards from a game mechanic, though.  IRL you can see a car near the horizon line if you've got clear visibility conditions and all, especially if it's lit at night. That's about three miles away, and there's no way those 60 tiles could represent three miles.


The reality bubble is purely an engine artifact and has no exact real-world equivalent. You can see hordes and buildings and things beyond it, even for "hordes" of one zombie, so we shouldn't look too hard at the bubble when figuring out ranges.
 It'd be better to just arbitrarliy pick an official size for a tile that seems to line up about right, and then run with it., and start adjusting things to match where possible.

Offline Coolthulhu

  • Contributor
  • Survivor
  • ****
  • Posts: 3753
    • View Profile
It'd be better to just arbitrarliy pick an official size for a tile that seems to line up about right, and then run with it., and start adjusting things to match where possible.

No, it wouldn't.
Tile size has to express different scales because of multiple mechanics that require different scales.

Tile size has to be small enough that a car door can occupy it without it growing too weird and big enough that a relatively small reality bubble encompasses big enough area for it to not be claustrophobic.
It has to be small enough that a rat in it doesn't look like a single star on empty sky and big enough that hulk fits in it entirely, without squeezing.

And before you suggest obvious solutions like "just make the bubble bigger", "just get rid of bubble", "just make hulks take more tiles" - those were all suggested before and re-suggesting them won't help the tiniest bit. The problem is that they aren't implemented.

Offline Mantar

  • Zombie Food
  • *
  • Posts: 80
  • NOMAD
    • View Profile
Tile size has to express different scales because of multiple mechanics that require different scales.

I'm aware of that. But a first step towards fixing the wide variance between those scales would be to look at what those different scales are and see what tile "size" fits best with most of them. Then later you can try to adjust the others to more or less line up. (There'd naturally have to be a lot of wiggle room)

Quote
Tile size has to be small enough that a car door can occupy it without it growing too weird and big enough that a relatively small reality bubble encompasses big enough area for it to not be claustrophobic.
It has to be small enough that a rat in it doesn't look like a single star on empty sky and big enough that hulk fits in it entirely, without squeezing.

The hulk and other really big things are an outlier, because you don't have multitile creatures yet. That's another thing that'd have to be ignored when working out distances.
 And rats and tiny creatures will always have some weirdness since you can't pack more than one per tile. (Unless... nah, nevermind, that would be too opaque from an interface standpoint)

Quote
And before you suggest obvious solutions like

I wasn't going to suggest anything except "figure out a reasonable approximation of how big a tile is, and then begin range calculations from there."

Offline Coolthulhu

  • Contributor
  • Survivor
  • ****
  • Posts: 3753
    • View Profile
I wasn't going to suggest anything except "figure out a reasonable approximation of how big a tile is, and then begin range calculations from there."

I did figure it out: it's "as big as it needs to be". That's the only sensible approximation I've seen in those few years I've been working on DDA.
Anyone who tried to work out a single size always ended up trying to handwave very hard to fix and horrible problems such as single tile creatures, reality bubble having to have limited size, screen display having to have limited size and supporting weaker computers and so on.

Until we have multitile creatures, arbitrary sized bubbles (ie. bubble calculations so optimized that bubbles can be few hundreds of tiles in diameter), inter-tile walls for vehicles, and UI that manages to handle the last two without causing massive sanity loss, tile has to stay variable in size. And since the last two are unlikely to ever happen, so is fixed tile size.

Offline Weyrling

  • Survivor
  • ***
  • Posts: 521
    • View Profile
Dispersion is currently calculated by summing up random rolls for weapon dispersion, skill dispersion, sight dispersion, recoil and possibly something more I forgot. Those can't be exchanged - if you have 50 weapon dispersion and 50 skill dispersion, you will not be as good at sniping as if you had 100 skill dispersion and 0 weapon dispersion.
Are you saying that we actually do several rolls for accuracy and they each have enough weight to decide whether something is a hit or a miss?
Do we just start at 0 and add all the rolls, are some rolls potentially negative, or what?

I would assume that each 'dispersion' would be a positive to negative range, but if they're all "0 to Whatever" it would explain why accuracy is so weird.

I've been somewhat confused at how accuracy works in this game for awhile, because my hit ratio always seems arbitrary right up until I'm literally incapable of missing.