Author Topic: Bugs & Crash Reports  (Read 1434426 times)

xcomfan

  • Guest
Re: Bugs & Crash Reports
« Reply #3690 on: September 09, 2020, 07:37:26 pm »
Don't know but maybe is: in L3 please someone might check the power stun value for dart pistol and haroppon gun? usin on the ground the pistol seems to be far more effective thna the harpoon when it comes to stunning, also i killed some people usin the pistol's sleep dart clip :O when i hit someone the target sprinkles with some purple stuff :O

Offline legionof1

  • Moderator
  • Commander
  • ***
  • Posts: 1899
  • Bullets go that way. Money comes this way.
    • View Profile
Re: Bugs & Crash Reports
« Reply #3691 on: September 09, 2020, 09:09:25 pm »
Seems intended, the dart pistol Daze ammo is a 1 tier higher tech. So it should be compared to the Harpoon tranquilizer ammo you get from the same tech and not the stun bolts.


Offline Delian

  • Colonel
  • ****
  • Posts: 240
    • View Profile
Re: Bugs & Crash Reports
« Reply #3692 on: September 09, 2020, 09:22:46 pm »
If I wear Ganger armor (85 CUTTING) and Shepherd's Staff (Says 15% less Cutting damage), you would expect your CUTTING to be 70, right? Well, it's not. It's 77?
Huh? It says 15% reduction but it only decreases it by 8%? Clearly there's a bug here.

The problem is in Yankes_Scripts.rul where it incorrectly calculates the reduction. The formula it uses is:
Final resist = Armor resist - (1 - Item resist) * (Armor resist - Item minimum resist)

For the above example, this makes it: 0.85 - (1 - 0.85) * (0.85 - 0.30) => 0.85 - 0.15 * 0.55 => 0.85 - 0.0825 => 0.7675 => 0.77

The forumula it should be using instead should be:
Final resist = Max(Armor resist - (1 - Item resist), Min(Item minimum resist, Armor resist))

This would make it: Max(0.85 - (1 - 0.85), Min(0.30, 0.85)) => Max(0.85 - 0.15, 0.30) => Max(0.7, 0.3) => 0.7
« Last Edit: September 10, 2020, 12:46:23 am by Delian »

Offline legionof1

  • Moderator
  • Commander
  • ***
  • Posts: 1899
  • Bullets go that way. Money comes this way.
    • View Profile
Re: Bugs & Crash Reports
« Reply #3693 on: September 09, 2020, 10:05:36 pm »
Not a bug. This is the function that all such resist items where designed around. Diminishing returns with a hard floor. Otherwise you can very easily reduce incoming  damage to below armor and therefore become immune to harm.

Offline Delian

  • Colonel
  • ****
  • Posts: 240
    • View Profile
Re: Bugs & Crash Reports
« Reply #3694 on: September 09, 2020, 10:31:36 pm »
Not a bug. This is the function that all such resist items where designed around. Diminishing returns with a hard floor. Otherwise you can very easily reduce incoming  damage to below armor and therefore become immune to harm.

Diminishing returns? What are you talking about? There's no "Diminishing returns" in that function. Also, it's not a hard floor at all because this part of the function is bugged also. As it stands, this function is just badly written and confusing.
« Last Edit: September 09, 2020, 10:44:29 pm by Delian »

Offline Meridian

  • Global Moderator
  • Commander
  • *****
  • Posts: 8597
    • View Profile
Re: Bugs & Crash Reports
« Reply #3695 on: September 09, 2020, 10:53:43 pm »
Diminishing returns? What are you talking about? There's no "Diminishing returns" in that function. Also, it's not a hard floor at all because this part of the function is bugged also. As it stands, this function is just badly written and confusing.

Diminishing returns are in the component, which you identified as "(Armor resist - Item minimum resist)".

Correctly, it should be "(Current resist - Item minimum resist)", where "Current resist" is equal to "Armor resist" only at the beginning of the iterative calculation and gradually gets smaller and smaller with each item... diminishing returns.

The function and the whole script (and this whole feature concept!) is confusing, I give you that... that doesn't mean it's wrong.
« Last Edit: September 09, 2020, 10:56:05 pm by Meridian »

Offline Delian

  • Colonel
  • ****
  • Posts: 240
    • View Profile
Re: Bugs & Crash Reports
« Reply #3696 on: September 09, 2020, 11:19:41 pm »
Well, the main problem is in the description an item of what it says it provides. If I have only one item, without any stacking or diminishing returns going on, the item saying it gives 15%, but actually gives only 8%... this can only be described as a bug, even if a visual one.

A player doesn't care about what goes on under the hood of the game. If the game tells them that they have 100% chance to hit something, and they miss, then they're going to get mad. And if such discrepancies keep happening, then eventually the player will ragequit and delete the game. That's why it's important to provide the players with correct information, which in this case, it is not. And I'd like to ask you to fix it. Either fix the function, or fix the description.

Offline Meridian

  • Global Moderator
  • Commander
  • *****
  • Posts: 8597
    • View Profile
Re: Bugs & Crash Reports
« Reply #3697 on: September 09, 2020, 11:38:25 pm »
Well, the main problem is in the description an item of what it says it provides. If I have only one item, without any stacking or diminishing returns going on, the item saying it gives 15%, but actually gives only 8%... this can only be described as a bug, even if a visual one.

Completely agreed.
I understood what you wanted to say in the first post, and I reacted specifically only to the second post.

A player doesn't care about what goes on under the hood of the game. If the game tells them that they have 100% chance to hit something, and they miss, then they're going to get mad. And if such discrepancies keep happening, then eventually the player will ragequit and delete the game. That's why it's important to provide the players with correct information, which in this case, it is not.

A normal player will never notice this particular issue.
May I ask how you found out?
The game never communicates these percentages in any form... and the difference between 70% resistance and 77% resistance is practically indistinguishable from a slightly better or slightly worse RNG roll.

Edit: 70%, not 60%, 85-15=70

And I'd like to ask you to fix it. Either fix the function, or fix the description.

I can't do that, I'm not the script author nor the mod author.
Let's wait for their feedback.

But removing all numbers from descriptions of such items sounds like a good idea to me ;)
« Last Edit: September 10, 2020, 12:00:32 am by Meridian »

Offline Greep

  • Colonel
  • ****
  • Posts: 298
    • View Profile
Re: Bugs & Crash Reports
« Reply #3698 on: September 10, 2020, 12:15:53 am »
Cutting resistance is pretty noticeable since melee tends to be many small attacks that will kill you if they do any damage.  E.g. dogs won't hurt you at all in scale but can  kill you in a single turn in warrior.  (30 armor 80 cutting vs 30 armor 60 cutting)

Offline Delian

  • Colonel
  • ****
  • Posts: 240
    • View Profile
Re: Bugs & Crash Reports
« Reply #3699 on: September 10, 2020, 01:32:11 am »
May I ask how you found out?
The game never communicates these percentages in any form... and the difference between 70% resistance and 77% resistance is practically indistinguishable from a slightly better or slightly worse RNG roll.

Edit: 70%, not 60%, 85-15=70

But removing all numbers from descriptions of such items sounds like a good idea to me ;)
Math still fails me lol, yes, 70, not 60.
In the description of Shepherd's Staff it says "Wielding Shepherd's Staff as a weapon reduces Cutting damage taken by 15% (to a min res. of 30%)". This doesn't say much, so I tested it out by stunning an enemy and then giving it a cutting damage weapon. I tried stacking (it's funny that stacking the staves stacks the morale penalty, but not the resistances - another bug?), and seeing what the maximum damage taken was. Needless to say, I was very disappointed to find out that the damage reduction was not what I expected it to be. In most other games resistances stack in more logical ways. Well, later I found out that the script writes the actual resistance in the savegame file, so you can just read it from there. Anyway, 15%, 8%, the difference isn't much, but it's enough to turn a "possibly useful weapon" into "garbage".
Yeah, removing all the numbers, including damage, from the descriptions of all items would be the best, right? Players should choose their favorite weapons based on feelings ;)

Cutting resistance is pretty noticeable since melee tends to be many small attacks that will kill you if they do any damage.  E.g. dogs won't hurt you at all in scale but can  kill you in a single turn in warrior.  (30 armor 80 cutting vs 30 armor 60 cutting)
Does wielding Shepherd's Staff help against dogs? It doesn't. Because the reduction is too small. So it's currently a useless item for this purpose. The purpose being, a swappable resistance item.

Offline Greep

  • Colonel
  • ****
  • Posts: 298
    • View Profile
Re: Bugs & Crash Reports
« Reply #3700 on: September 10, 2020, 01:46:29 am »
Oh I was just replying to meridian.  It would make a difference if it gave the displayed value (e.g. warrior + 15% staff vs 8% staff)

Offline Meridian

  • Global Moderator
  • Commander
  • *****
  • Posts: 8597
    • View Profile
Re: Bugs & Crash Reports
« Reply #3701 on: September 10, 2020, 02:13:12 am »
Yeah, removing all the numbers, including damage, from the descriptions of all items would be the best, right? Players should choose their favorite weapons based on feelings ;)

OK, so what would be your counter-proposal... assuming the authors want to keep the algorithm the way it is.
What description would you write?

Oh I was just replying to meridian.  It would make a difference if it gave the displayed value (e.g. warrior + 15% staff vs 8% staff)

I shouldn't even be replying to this, but since you insist:
I just tried standing in a warrior armor next to 4 guard dogs for 4 turns in a humanist villa, they attacked me well over 60-70 times and did a grand total of 17 damage... big whoop.

Anyway, even if you find one specific combination of circumstances, attack type, armor, weapon, difficulty, equipped items and god knows what else to "prove me wrong"... in 99.8% of the combinations, the few percent resistance difference MAKES NO FU*KING DIFFERENCE. Your 0.2% doesn't prove me wrong, it proves me right.

Offline Greep

  • Colonel
  • ****
  • Posts: 298
    • View Profile
Re: Bugs & Crash Reports
« Reply #3702 on: September 10, 2020, 02:16:38 am »
Hmmm, does difficulty affect attack damage?  Because I have been dropped from full hp by a single dog in warrior.  Maybe it was just an insane fluke.

Offline Meridian

  • Global Moderator
  • Commander
  • *****
  • Posts: 8597
    • View Profile
Re: Bugs & Crash Reports
« Reply #3703 on: September 10, 2020, 02:27:47 am »
Hmmm, does difficulty affect attack damage?  Because I have been dropped from full hp by a single dog in warrior.  Maybe it was just an insane fluke.

Actually yes, for dogs it does depend on difficulty... yet another proof that any small resistance deviation is negligible compared to other factors.
For the record, I was testing on standard Veteran difficulty.

Offline Greep

  • Colonel
  • ****
  • Posts: 298
    • View Profile
Re: Bugs & Crash Reports
« Reply #3704 on: September 10, 2020, 02:32:05 am »
Well, I don't mean to get into a big argument (or any argument, I don't see a big reason to angry over something so trivial), so I'll just give my final 2 cents here.  If the difference between warrior and scale is just 20 resistance and is the difference between being invulnerable and killed in one turn, kinda seems to make a massive difference.

But yes, shepherd's staff and studded stockings might be  the only protective gear cases where it matter.