Author Topic: Is maximum range even a sensible attribute within current ranged weapons system?  (Read 2125 times)

Offline Coolthulhu

  • Contributor
  • Survivor
  • ****
  • Posts: 3748
    • View Profile
NO. IT. WAS. NOT.  That system literally threw away important data about dispersion in the name of "making it easier to understand", while doing nothing of the sort.  It simply threw all sources of inaccuracy into a bucket and then massaged them to get a result that vaguely resembled the output you were looking for, but mostly ignored the inputs.  Additionally, any theoretical gains from "simplifying" the distribution were negated by an over-complicated and self-referential system for setting maximum effective gun ranges that was riddled with bugs and oversights, and didn't actually verify anything important about accuracy.

The "throw the error sources in a bucket and treat it like an arbitrary distribution" system is a dead end, you can't do anything meaningful with it except force the distribution to be what you want it to be, becuse you've thrown any additional inputs that you could use away.

That's all a bunch of assertions targeting the implementation details that only show you didn't think about it.

Ignores the inputs? This is what everything else in the game does: extracts a final value and uses that. It's not like your mess of a system does anything else - it only delays the final calculation instead of summing them up on the spot.

The "over-complicated and self-referential system for setting maximum effective gun ranges that was riddled with bugs and oversights"? It was actually amazing at informing the player how effective the gun will be, helping NPC AI, providing the most important data about gun accuracy as a single nice value. It tells the PLAYER what PLAYER needs to hear, something so lacking in this game that without lurking github and forums or reading code you can't feasibly begin to understand what is happening in many of the systems in the game.
Instead you replaced it with an intentionally bad approximation, so that anyone not checking github will be at an even greater disadvantage than before. Ironic, considering some of the 0.D milestone PRs were accessibility related - even people without working eyes get to be misled! And all of that with no warning that it's an approximation too.

If you didn't like the clean, natural distribution system because it ignored inputs, you could have passed them along to do something interesting with them.
Instead you reverted it into a mess where Simo Häyhä equipped with a standard issue police pistol is about as effective at ranged combat as a jackass with a piece of metal hammered from broken lockers tied to a stick. And it only gets worse when you compare equal skills.

Oh and you didn't do anything interesting with the inputs either. You only use them to mess up the distribution in such a way that a single bad variable (which you didn't bother to reveal to players in any way, instead choosing intentionally misleading display) pretty much guarantees inaccuracy. This certainly is harder to understand than throwing them all together and just treating them all in the same way.

Quote
The problem with accuracy stems from the fact that dispite my repeated demands to establish some kind of baseline accuracy tests (as a prerequisite for overhauling how dispersion worked), mugling instead decided to add a bunch of code to the game to output graphs to convince themselves that the accuracy numbers at one specific point in time were good, and then YOU, Coolthulu decided to just merge the bullshit changes instead of holding them to that requirement.

I did fuck up a bunch of times in the development. I misunderstood you and others, missed some moderately important details (limb re-breaking, for example!), implemented things that turned out badly thought out or badly implemented.
But merging that amazing change (set of changes) was not a fuck up. Even with all its shortcomings, it was the single best thing that happened to ranged combat since 0.A (and probably earlier, but I haven't seen much of that).

The system unfucked the mess that the ranged combat was since aiming update and maybe even before. It made ranged combat WORK.
And now you regressed it all into... what we have now. Without considering the consequences at that - clearly evidenced by you saying "As a result the player-visible change to accuracy should be minimal to nonexistent.".
A ton of work that went into statistical analysis to ensure things don't suck? Gone, with nothing to replace it. And shockingly, things suck now.

Honestly, at this point it looks more like some sort of personal vengeance or dominance assertion. When a whole host of github lurkers and contributors ask you not to fuck things up, you fuck things up and skip the usual no self merging thing.
When you're given a set of numbers to fix this - which you asked for - you ignore them and all you have to say are accusations and excuses that all avoid the central point that things are a giant mess now, in pretty much every way worse than what was before, with examples that cover both game mechanics and real life.
All you have to say about the system targets some internal details, none of your arguments translate into meaningful examples - unlike the system you accuse of "not making sense", which handles them just fine.

You weren't like this before. Around ~0.B, it seemed like you genuinely want to make this game better, not "more yours".

Offline Tamior

  • Zombie Food
  • *
  • Posts: 78
    • View Profile
Aaanyway, let's try to see if I understand correctly what kind of system we want in general.

So in broad terms:
- every ranged attack should have 2 components "where the weapon is facing" and "how much the projectile veers off course from where the weapon was facing"
- the "where the weapon is facing" part is influence by player skill and actions — some weapons may help acquire target faster, some may allow for more precise aiming,  but in the end it may point right at the head of even a very distant target for any weapon
- the "how much the projectile veers off course from where the weapon was facing" part is NOT influenced by player skill and is the part that makes some weapons intrinsically inaccurate at long ranges no matter the skill.

Is that the framework we want in general or am I missing something?
« Last Edit: July 11, 2017, 12:43:06 AM by Tamior »

Offline Kevin Granade

  • Administrator
  • Survivor
  • *****
  • Posts: 5497
  • I code dead people.
    • View Profile
Aaanyway, let's try to see if I understand correctly what kind of system we want in general.

Details

Is that the framework we want in general or am I missing something?

That IS roughly the system we have, though theres no split, its just, "where the player wants the projectile to go", which then has errors added to it to determine where it actually goes.

Coolthulu, if you want to start making personal accusations, we have nothing to discuss.
Its like a fun family cookout, except your family is burning in flames while trying to eat you. -secretfire
I'm more excited than a survivor on meth and toast'ems. -Nighthawk
The the giant wasp is slammed through the zombie brute!

Offline Coolthulhu

  • Contributor
  • Survivor
  • ****
  • Posts: 3748
    • View Profile
Coolthulu, if you want to start making personal accusations, we have nothing to discuss.

I provided the necessary examples in the github issue, with relatively in-depth rationale. If needed I can make them into unit tests, expand the examples to cover more cases.
I don't insist on it being achieved in a particular, specific way. But so far there was no achieving anything. Anything positive.
I don't see any direction here, so I provided a familiar one: the last good solution.

Offline Tamior

  • Zombie Food
  • *
  • Posts: 78
    • View Profile
That IS roughly the system we have, though theres no split, its just, "where the player wants the projectile to go", which then has errors added to it to determine where it actually goes.
Ok, then the question is: what exactly is wrong in the current system that we have to use hard-coded max range?
Can't we just replace all max ranges with "how much the projectile veers off course from where the weapon was facing" dispersion value in an inverse fashion?
I.e. low max ranges get high dispersion, high max ranges get low?


Offline Alec White

  • NPC
  • **
  • Posts: 205
  • I'm shy
    • View Profile
At this peace, I see 0.D release for 2020.

More or less at the same time for the next Dwarf Fortress version.

Offline Tamior

  • Zombie Food
  • *
  • Posts: 78
    • View Profile
At this peace, I see 0.D release for 2020.

More or less at the same time for the next Dwarf Fortress version.
I say we should start aiming to merge the two in 0.E
Then merge with DayZ in 0.F

Offline zombiechow

  • Zombie Food
  • *
  • Posts: 23
    • View Profile
I personally think the handgun max range in particular is laughable, you can throw a rock farther than you can shoot a pistol currently..

EDIT: Just for reference, the men's shot put record (not even throwing the object to try to damage anything) is 23 m. USP 9mm effective range is supposed to be at least 50 m (going off of wikipedia). There is a huge disparity here with ranges vs. real life.
« Last Edit: July 14, 2017, 04:10:44 PM by zombiechow »

Offline RedPine

  • Zombie Food
  • *
  • Posts: 71
    • View Profile
I personally think the handgun max range in particular is laughable, you can throw a rock farther than you can shoot a pistol currently..

EDIT: Just for reference, the men's shot put record (not even throwing the object to try to damage anything) is 23 m. USP 9mm effective range is supposed to be at least 50 m (going off of wikipedia). There is a huge disparity here with ranges vs. real life.

Of course, it's okay if things don't scale the same in CDDA as real life, but I'd expect at least a similar balance to IRL.  A thrown rock definitely should not have MORE range than a pistol, and if it does, only at a severe penalty to damage and accuracy.  Also, IRL, it takes a LOT longer to throw a hefty rock (or even a throwing knife) than it does to pull a trigger.

As for max range, I'm fine with having a max range for weapons since it's easier to balance a fixed value than a weird formula.  I'm sure the formulas are very fancy and hard to make, but at the end of the day the only feedback a UI can give to a player is a fixed value and NOT the formula itself.  It makes sense to feed the game the same information that is fed to the player to allow more transparency.  Engines are another example of something that is hard to balance because it uses a complicated and realistic formula instead of fixed values.

Offline Tamior

  • Zombie Food
  • *
  • Posts: 78
    • View Profile
How exactly are engines "hard to balance"?
What engine is … unbalanced?

Offline Coolthulhu

  • Contributor
  • Survivor
  • ****
  • Posts: 3748
    • View Profile
Engines are another example of something that is hard to balance because it uses a complicated and realistic formula instead of fixed values.

Currently they aren't. They were hard to balance at one point, but that wasn't ever finished and got reverted.
Nowadays, for combustion engines, it's just "bigger=better".

Offline RedPine

  • Zombie Food
  • *
  • Posts: 71
    • View Profile
Engines are another example of something that is hard to balance because it uses a complicated and realistic formula instead of fixed values.

Currently they aren't. They were hard to balance at one point, but that wasn't ever finished and got reverted.
Nowadays, for combustion engines, it's just "bigger=better".

I see, I haven't played in quite awhile.  I've got the CDDA itch again, but I'm hesitant to jump in due to the gun controversy (prior, I was hesitant due to the engine controversy).  Then again, it's not like I use guns anyway due to limited ammo and noise.  I suppose making a melee/throwing character would completely sidestep the guns issue until things settle down.

Offline zombiechow

  • Zombie Food
  • *
  • Posts: 23
    • View Profile
I understand the need for max range as a gameplay balancing thing. 1shotting everything the moment it enters your reality bubble would probably make a sniper playstyle way too good.

Why not have Focus come into play? Kind of like a 'stamina' for aiming ranged weapons, the longing you are aiming and scoring hits outside of the weapon's 'preferred range' the quicker your Focus would drain lowering your accuracy and speed of aiming, while shooting inside that range could have a greatly reduced penalty. Maybe with a high Focus you can make a difficult shot near the edge of your LoS and easily pick off the occasional straggler but a hoard could be another thing entirely.
« Last Edit: July 15, 2017, 09:02:05 AM by zombiechow »

Offline Tamior

  • Zombie Food
  • *
  • Posts: 78
    • View Profile
I understand the need for max range as a gameplay balancing thing. 1shotting everything the moment it enters your reality bubble would probably make a sniper playstyle way too good.
1) Melee play-style is already overpowered over "sniper". Not in the least because it's way more common for sniper to be forced into melee (and be killed) than for a melee to be forced into ranged combat.
2) You will NOT be "1shotting everything the moment it enters your reality bubble" if we drop hard-coded max ranges and replace them with accuracy fall-off any more than you do now. Because right now there is already stuff like (heavy) rail rifle with ranges higher than range of sight, the limiting factor on them is ammo, if anything.
« Last Edit: July 15, 2017, 01:57:14 PM by Tamior »