FO2 Mechanics Overhaul Mod

Sduibek said:
On that case, for that you could just consider 4= dumb, 5= no modifiers, 6= smart :shrug:

Heh, yeah, but the 4-6 values don't seem to have a lot of logic behind them either, just one of those cases of "they have to have some value".

On the other thing, how do you get around save/load issue? I agree that if player knows it's a roll, they'll just keep trying. Those without meta knowledge are safe though. But see Fallout doesn't maintain variables until saved again so I don't see a way around it.

You get around it by removing it wherever you can. I really wish I could change the way skills like doctor and steal work for instance, just remove randomization altogether. Ideally, after hundreds of years of Timeslip working on sfall, I'd restrict all randomization to unsaveable combat (and then only to hit and crit rolls).

Of course, a game like FO with a luck stat should have some type of randomization in it, but then I would prefer it to be an extremely early roll, so it can't be reloaded. For instance, some big luck rolls (zeta scan, vault city test, etc.) all get done at the very start of the game, and saved in variables. Try savescumming that, you bastards! :p

Speaking of which, I hate how Fallout does skill rolls. No modifiers in most cases, just roll 1 to 100 vs player's skill. Really?? The extreme randomness of that is fucking retarded. Imagine if D&D used a D100 for everything...

Yeah, Fallout 2 solved a lot of that by ommission, but ideally they'd still have GURPS's bell curves (3d6 rolls mean that 3 and 18 are very rare).
 
That exact roll is in the Caravan Encounters script files in Fallout 1.... so it begs the question, do you think most people would believe that 3d6 is "objectively superior" function to 1d18?

Luck only at beginning of game
Well to my mind, Luck should always matter. For example if you have a check against Science to hack a terminal, the calculation might look something like this:

if ( Skill_SCIENCE > [50 - Luck] ) then success

That gives me interesting ideas though, maybe someday FIXT will have a "prevent savescumming" install option :lol:

Do keep in mind though, that unless your mod is for Hardcore/IronMan -type players, things like savescumming should not be 100% impossible to do. Some people just like playing the game to have fun, and if save-load is fun and fine to them, why not let them? I know I do it, and it doesn't make me enjoy the games any less. Sure it makes it easier or whatever, but it also lets me get certain playthroughs that I want instead of having to start the whole damn game over.

A bad analogy is Diablo II vs Diablo III. Diablo II, if you pick the "wrong" skills, once in Hell mode you're fucked and your character is forever worthless for anything but farming. Diablo III lets you change whenever, so you can always have things be fun and interesting. Some would say Diablo III is for weaklings or whatever, but personally I feel the "fuck you, you fail lol start over" approach of Diablo II is just totally unnecessary.

^ So, same goes for if you prevent any and all things like savescumming. It makes the hardcore players happy but will make everyone else frustrated.
 
I suppose you could do something like that (personally I hate the Luck stat, but from most people I gather that they really like that about FO so I'm in the minority there I guess).

In general, a FIXT version that removes all skill rolls would be awesome in my opinion, no matter which way it turns out.

it begs the question, do you think most people would believe that 3d6 is "objectively superior" function to 1d18?

If you're going to do skill rolls in the first place, then yeah because it seriously cuts down on randomization and thus savescumming. Like you already hinted at, the 1-100 against player skill rolls are probably the single biggest disaster to have come out of the engine rewrite. Still pretty damn impressive that Cain managed that in a single week of course.
 
I would appreciate such a probability distribution for many skill rolls. Maybe 2 or 3 averaged random numbers.

For the save scumming i have to say: Maybe better leave it alone. The zeta scan is a good example. Many players regard this, as the create their character. Btw this is also exploitable, by travelling early to NCR. So fixing these rolls, you only make it more timeconsuming for achieving the right outcome.
Like this charsi thing and socketing in Diablo2.

And for my testing... can you provide a fix for the ammo issue? It's a bit disturbing, because this makes buying ammo sometimes to expensive. If the vendor has only one stack, that can't "magically" merged with other stacks before you complete the deal, the price is another ~5 times higher.
 
Panzerkampfwagen said:
Issue with 100% THC happened usually after creature critically missed and lost its next turn. At least for me. I fought this wasn't bug. :P
I can confirm this now.


My unarmed skill is now @ 102%. Attacking pig rats now leads to torsohitchance of 96 or 99%. Should these values exceed 95%?
 
And for my testing... can you provide a fix for the ammo issue? It's a bit disturbing, because this makes buying ammo sometimes to expensive. If the vendor has only one stack, that can't "magically" merged with other stacks before you complete the deal, the price is another ~5 times higher.

You can set ReducedAmmoStacks=0 in the .ini until I fix the problem, that'll just remove the feature. I'm too lazy to continuously put out piecemeal updates and am working on a big update that'll add a lot of new features and (hopefully) fix the issues everyone is dealing with.

QuFu said:
My unarmed skill is now @ 102%. Attacking pig rats now leads to torsohitchance of 96 or 99%. Should these values exceed 95%?

Yes, they should. In vanilla FO2, critical failures are tied into misses: high chance to miss is high chance of your gun exploding, etc. Now the two are separated: so if you have a 100% to hit, you'll almost certainly still have a small chance of critical failure. That's why the 95% cap is now gone. Though I have to admit that those values sound a bit high... might need to do some rebalancing of the formula again (I'll probably have to continuously do this of course).
 
Doesn't SOME kind of cap, even if it was 99%, make sense?

I mean even the top marksmen from Marines or Navy SEALs or Special Forces is NEVER going to "Never Miss."

That's just not humanly possible.
 
Sduibek said:
Doesn't SOME kind of cap, even if it was 99%, make sense?

I mean even the top marksmen from Marines or Navy SEALs or Special Forces is NEVER going to "Never Miss."

That's just not humanly possible.

Yeah, you're right, I think I'll set it at 99%. I remember reasoning that shooting a brahmin at point blank range as a marksman would be effectively 99.99999% unless hit gun misfires, but 100% probably seems odd to a lot of people.
 
Cool :ok:

See this is why threads like this and wikis for feedback are useful :) I'm glad I've gotten so much on mine.
 
There is another good reason for capping at 99%. The numbers in the window for aimed hits getting messed up, when exceeding 99% (there are only 2 digits).

And now for some other things: fleeing... ;)
Had the quest fight at the church in den (lara, tyler,...). Fleeing isn't an option in such a fight. I had to reload 4 to 5 times to get all enemies killed. The worst case was, when i ended up (stuck in combatmode) in front of the church alone and nobody gets killed. :roll:

Vic is also completely useless. I gave him a 10mm gun. But i didn't see a single fight, where he wasn't fleeing. Sometimes when he gets his distance, he starts shooting... But you know... at early levels he was never good at that and you regret that you gave him a gun. :D
Btw, Sulik seems to be fine.

Well... i think you can't heard about fleeing anymore. :) So finally i say: I like this mod. ;)
 
QuFu said:
There is another good reason for capping at 99%. The numbers in the window for aimed hits getting messed up, when exceeding 99% (there are only 2 digits).

Heh, forgot about this one, good call.

And now for some other things: fleeing... ;)
Had the quest fight at the church in den (lara, tyler,...). Fleeing isn't an option in such a fight. I had to reload 4 to 5 times to get all enemies killed. The worst case was, when i ended up (stuck in combatmode) in front of the church alone and nobody gets killed. :roll:

Hmm.. this is starting to make me think something might have gone wrong with your install, because I've used the Den gang fight for a lot of my testing, and it seemed to go fine without fleeing last time I tried it. Could you check in Patch000.dat/data/AI.txt and see if the min_to_hit values (they're all the same) are anything other than -80?
 
No, they are all at -80.

Otherwise, i think their fleeing was also caused by having to low HP. Someone starts fleeing, some other chase him, ending in fleeing too. I'm not completely sure of this. Are NPCs affected by morale and exhaustion?

I watched (encounter) some radscorpions killing some fleeing traders. Was a very long "fight". After the scorpions finished all the sitting ducks, they come after me. But my morale and exhaustion values were so bad, i get a hitchance of ~-150.

For my install: Took a fresh unpatched F2 install. Installed RP 2.2 with standard options. Then copied all files from your mod 1.48 with overwriting. File protection wasn't necessary. There where no older versions of your mod.
Did some graphic changes after that. Set only fixedhp=0 and didn't use your derivedstats. Anyway, i set combat difficulty to hard ingame (i always did that).
 
Ugh, that sounds a lot more terrible than I expected. Thanks for the info, I'll do a lot more testing and see if I can solve the exhaustion and morale issues.
 
Sduibek said:
How did you come to decide on -80 specifically as the value?

Just needed a very low value. The to hit script checks the "can't damage" and "probably will die" factors, and if they apply, the to hit chance gets set to -81 and they flee. Though obviously this still needs a lot of work.
 
Okay, thanks. Just makes me wonder if my "No-Flee" AI should have MinToHit set to -80 instead of 0 :shrug:

Currently if using that AI version, all critters just have this:
Code:
min_to_hit=0
min_hp=0
aggression=100
hurt_too_much=

The only exception being a few critters that couldn't have those numbers because it messed up their in-game function (such as fighting Garl one-on-one. He was supposed to yield when you got him hurt really bad, instead you had to kill him. My bad :mrgreen::roll:)
 
Don't think so, the problem the -80 is meant to address is that with my to hit script, critters with melee weapons think they're getting distance penalties with their weapons when calculating whether or not to flee, which results in really low values. I think.
 
But what about the critters with guns? I saw an unhurt Tyler (with a gun) fleeing. But this was at night. And lvl 1&2 Vic (10mm gun)? His chance to hit shouldn't be that bad (only annoying sometimtes). Beside, he has much more HP than any other critters in the starting area. It would be fine, seeing him ran from a deathclaw but not some tiny geckos.
 
Maybe this isn't important right now, but i did some thinking about your AP system and the related traits and perks.
While traits have benefits and drawbacks, perks should only have benefits.

So for the fast shooter trait i think it's ok that you get the aimed body hit chance for the snapshot but aren't able to do aimed shots anymore.

The increased ROF perk shoudn't reduce the base cost. Instead it should only reduce the costs for big guns, optional burst modes and aimed shots by 1. So, you don't need the the aiming penalty for this perk. This way it keeps the awfulness of it, while avoiding those OP 2AP for most weapons. Also it prevents the fastshooter trait from getting OP.
 
QuFu said:
But what about the critters with guns? I saw an unhurt Tyler (with a gun) fleeing. But this was at night. And lvl 1&2 Vic (10mm gun)? His chance to hit shouldn't be that bad (only annoying sometimtes). Beside, he has much more HP than any other critters in the starting area. It would be fine, seeing him ran from a deathclaw but not some tiny geckos.

That's interesting, if anything you'd think I'd made targets too easy to hit with my last changes, so lower than -80 sounds totally impossible. Must be something else that's causing it I guess. Thanks for the info, finally finished my new features (traps, assassinations and companion leveling) yesterday, so I can get to the bugs now.

As for the exhaustion problem you had (with the -150 to hit after a long fight), I think it was due to the amount of stuff you were carrying: if both you and other combatants are carrying the same, the disparity won't ever become bigger than a certain value (namely the difference in EN), but because the Radscorpions didn't carry anything, that totally threw it off.

Anyway, I'm going to remove the carry weight factor (it was just a stupid attempt at getting people to carry less junk and balance the economy anyway), because of this kind of nonsense. I'm also going to make a couple of other changes to make it work a bit better.

The increased ROF perk shoudn't reduce the base cost. Instead it should only reduce the costs for big guns, optional burst modes and aimed shots by 1. So, you don't need the the aiming penalty for this perk. This way it keeps the awfulness of it, while avoiding those OP 2AP for most weapons. Also it prevents the fastshooter trait from getting OP.

I get where you're coming from, and I didn't like the penalty for the perk either, but the problem I have with your alternative is that it creates this weird situation where increased rate of fire leads to players carefully aiming more and thus on the whole probably shooting slower, not faster because they stop using snapshots. That's why I plan on making the Sniper perk what you suggest in part and have it reduce aimed AP cost by 1 and scoped AP cost by 2.

The burst changes I like better, but what would you then do with the HtH attacks variant?
 
Back
Top