What has happened to the hacking system?

[Kahuna]

This is my first hack and Hacking Points are positive.
Why has hacking been removed from the game?
Also. AIs' tech level is II and the planet is Mark I.. yet the AI ships are Mark IV. Why?

[Kahuna]

There goes the motivation to play AI War once again. One thing gets fixed and other gets broken. hug that.
Maybe the AI's hacking response needed a buff but that doesn't mean you should go all Jay Wilson and double it. "More!" isn't always the way to go.

[Zincat]

Uhm, I actually noticed the huge increase in hacking response as well, compared to a few months ago. Nowadays I usually do not hack unless I have unconventional ships (spire... champion... the like), or a well established beachhead.

I do think that hacking response was too easy before, but now it's excessive in my opinion. At the very least, the game should factor in whether you have unconventional ships or not a bit, how high is the aip etc.

[TechSY730]

Two things.

What difficulty level were you playing against?
Do you have logs for this and/or a save game?

This is clearly more than the amount of buff stated in the patch notes; something is wrong. (unless you are playing vs difficulty 10, in which case, well, it is supposed to be unfair and stupid)

Anyways, the current AI tech level and the planet's tech level have no bearing on the tech level of the hacking responce ships. If they did, then the optimal thing to do would be to wait until you get a map with the important to hack stuff all on low tech level planets and only hack when AIP is very low, both of which are rather degenerate given the design goals of hacking.

[TechSY730]

There goes the motivation to play AI War once again. One thing gets fixed and other gets broken. hug that.

Ah, the life of playing on the betas.

If you want more stability and predictability, there is always the option of sticking with the official stable versions.

Maybe the AI's hacking response needed a buff but that doesn't mean you should go all Jay Wilson and double it. "More!" isn't always the way to go.

This is clearly greater than the advertised buff in the release notes, aka, a bug, not an intended buff.

That said, even though the "binary search" approach to balancing that Keith uses does get feedback quickly, it does lead to some short term pain and annoyance for players.

[keith.lamothe]

I'm happy to make further adjustments, I'm just not sure what's actually happening there because the last changes to hacking magnitude were around June 7th (over a month ago) and that was actually a nerf overall.  The last buffs (aside from some superterminal stuff) were around May 14th, or almost 2 months ago.  So my confusion is why I'm hearing about this now rather than before.  Not that all bugs are instantly found, but "The AI has an uncontrollable flood of MkIV ships" usually gets reported fairly quickly.

On the MkIV thing, that's from it combining its spawn strength so instead of hitting you with cpu-crippling numbers of MkI ships, etc...

But in general it seems pretty excessive for the first hack.  If you can post a save (with the hacking logs if you have them) I can figure out what part of the math decided to rewrite itself as

if %me% == SuperCat
  set /A me =


And on the overall issue of "agh I come back and just want to enjoy this game and some bug comes out of nowhere to smash things", I quite understand.  That's why I'm aiming to make the official versions more serious about their job of being stable versions, so someone who wants to play the game without also being part of the development process (or at least the testing process) can do so.  Most folks really enjoy the "being part of the process" aspect of the betas, but when someone doesn't have a lot of time, or has already spent thousands of hours helping us find bugs and refine the game... maybe they don't want to deal with it, and they shouldn't have to.

That said, if it takes 2 months to find something like this then maybe it will take a while to make those officials significantly more stable   We'll see.

[Kahuna]

What difficulty level were you playing against?

if %me%==SuperCat (
   set /A diff=10
)

Do you have logs for this and/or a save game?

No but I have beer and wine.
Ok ok I'll get new ones.

unless you are playing vs difficulty 10, in which case, well, it is supposed to be unfair and stupid

Still hacking is not supposed to be unfair and stupid compared to the rest of the game.

Not that all bugs are instantly found, but "The AI has an uncontrollable flood of MkIV ships" usually gets reported fairly quickly.

Hm. From what I've understood most people play on difficulties lower than 9. Maybe it's still manageable on that level. Maybe I could manage it on 10 diff if I drank a ciople of more beers. On the other hand it's stupidly overpowered cmpared to the rest of the game on any diffucilty.

what part of the math decided to rewrite itself as

if %me% == SuperCat
  set /A me =

mmMM-bwahaha

has already spent thousands of hours helping us find bugs and refine the game

Oh. I'm not TechSY730.

I'm aiming to make the official versions more serious about their job of being stable versions, so someone who wants to play the game without also being part of the development process (or at least the testing process) can do so.

k k

quote author=keith.lamothe link=topic=15974.msg176873#msg176873 date=1405177036]
That said, if it takes 2 months to find something like this then maybe it will take a while to make those officials significantly more stable   We'll see.
[/quote]
Arf

[keith.lamothe]

unless you are playing vs difficulty 10, in which case, well, it is supposed to be unfair and stupid

Still hacking is not supposed to be unfair and stupid compared to the rest of the game.

Exactly, we want equal-opportunity unfairness and stupidity.  If all the Diff 10 challenge mechanics were doing this it'd probably be fine, but if the rest of them are basically letting you through and then on your very first hack it says "NOPE!" then the stupidity-gradient needs to be smoothed out a bit.

[Kahuna]

Ok I have the save and the logs.

[keith.lamothe]

Ok, thanks.

Here's the mainly-relevant lines from the first entry:

Code: [Select]

7/12/2014 7:40:57 PM (7.045) 1:42:20 Counter-Saboteur Spawn on LV-51315  (playerNumber:8)
spawnStrength = base value = 17.5
spawnStrength *= handicap = 17.5
spawnStrength *= Game.Instance.Options.GetDifficultyCausedSpecialDifficultyModifier = 157.5
multiplierForSabotageResponseSpawn = 1 = 1

BaseHackingAntagonism = Max(10,BaseHackingAntagonism) = 10

spawnStrength *= multiplierForSabotageResponseSpawn = 157.5

maximumNumberOfWildRolls = 1 + (totalHackingAntagonism/50) = 1
wild-roll:SwitchBackToNormalSpawn, spawnStrength = 157.5

spawnedType:SpireGravityDrainIV quantity:10 (strength-each:4)
spawnedType:ZenithBombardmentIV quantity:7 (strength-each:12)
spawnedType:BomberIV quantity:5 (strength-each:2)
spawnedType:FighterIV quantity:6 (strength-each:2)
spawnedType:CruiserIV quantity:6 (strength-each:2)Note: BaseHackingAntagonism  only feeds into the number of wild rolls here, but once it gets past 100 it feeds into multiplierForSabotageResponseSpawn as well.

Also, GetDifficultyCausedSpecialDifficultyModifier() is:
- take the higher of the two AI difficulties
- if < 7, then diff * 0.15
- if 7, then 1
- if 8, then 2.250
- if 9, then 4.5
- if 10, then 9
(if it's an inbetween difficulty it interpolates linearly, so 9.3 gets 4.5 + (4.5 * 0.3), or 5.85)

This modifier is also used for exo sizes, champ nemesis stuff, heroic AI spawns, preservation warden spawns, hybrid pop-cap, special forces, and strategic reserves.

A different multiplier is used for waves; it's not being used here at all but since the concern isn't so much over the size of the hacking response but its relation to other parts of the diff 10 challenge:
- take the difficulty of the AI launching the wave
- if <= 3, then 1
- if 4, then 1.5
- if 5, then 1.75
- if 6, then 2
- if 7, then 2.25
- if 7.3, then 2.5
- if 7.6, then 2.75
- if 8, then 3
- if 8.3, then 3.75
- if 8.6, then 4.5
- if 9, then 6
- if 9.3, then 7.8
- if 9.6, then 9.6
- if 9.8, then 10.8
- if 10, then 12

So the 7-to-8 relationship is gentler, but 8-to-9 and 9-to-10 are doublings in both cases.  Though the minimum size of waves does tend to obscure the differences in the very early game.


Anyway, 1 cap of mark 1 ships is basically 200 strength (192 to be exact, but it's not that exact).  So the 158 strength spawn here should be somewhat less than that.  And the ships there do add up to 158, with the 7 bombards composing the bulk of it (84).

These checks run every 10 seconds in general.  Is 3/4 of a mkI cap (or a bit more) every 10 seconds excessive on Diff 10?


That's with it rolling "nicely" on the wild-rolls though, and basically getting nothing for it.

On the next check, 10 seconds later, it rolled QueueAnotherSpawnAfterThisOne, which just makes it do it all over again (though the do-over isn't allowed to roll a do-over, iirc)

And on another one later it rolled AddStrength_PreRollStrengthTimesOne, which adds the base 157.5 in over again, so also a double result.


Anyway, is it mainly the rolls causing double spawns and whatnot that makes it seem out of control, or is it inherent in the base rate of basically 16 strength per second being applied directly to the forehead?

[Kahuna]

These checks run every 10 seconds in general.  Is 3/4 of a mkI cap (or a bit more) every 10 seconds excessive on Diff 10?

It's hard to say how many ships in x seconds is excessive on y difficulty level.

Anyway, is it mainly the rolls causing double spawns and whatnot that makes it seem out of control, or is it inherent in the base rate of basically 16 strength per second being applied directly to the forehead?

All I know is that the problem is the sheer number of the ships and the fact that they're all Mark IV. Hacking shouldn't be "unfair and stupid" compared to the rest of the game and the first hack shouldn't require the use of Warheads. Hacking is supposed to get you stuff while avoiding AIP increase right?

[Gemzo]

Is this really just hacking being too hard? I'm not saying hacking isn't too hard, but the situation seems awful specifically because of the bombards. If your turrets didn't have to eat the damage penalty of being under forcefields in order to avoid the antimatter bombs, things would probably be a lot different, no?

[Kahuna]

So apparently this is not a bug and the hacking response is working as intended?
Or will something be done to fix this?

[keith.lamothe]

So apparently this is not a bug and the hacking response is working as intended?
Or will something be done to fix this?

I was trying to determine what the actual problem was before changing something.  But yes, something will be done.