Deviously Cursed Loot LE 9.0 (2021-03-09)

[Kimy]

Skyrim random number genrator is awfully bad.

For any mod you can put 1% of chances and get that 10 times in a row.

 

Yeah, the SKSE team wanted to include a replacement RNG one day. I truly can't wait...

[jaam]

They "scientifically" proved the vanilla one was working fine, so you can wait a long time

 

[Kimy]

They "scientifically" proved the vanilla one was working fine, so you can wait a long time

 

Scientifically, eh? I'd be curious what method they used.

[jaam]

This one : http://forums.bethsoft.com/topic/1501746-wipz-skyrim-script-extender-skse/page-7?do=findComment&comment=23721890

[thingwhatsqueaks]

 

Hmm, sometimes I do ask, why I keep this goddamned mod xD Started a new game (because of some other tweaks) and only did Bleak Falls Barrow ...

cursed-loot-result.png

 

BTW, is it normal that there only appear Yoke Keys besides of Restraints and Chastity ones? So, keys for special devices only appear if those devices are worn? Just to confirm that. And what is now, if a follower wears a slut collar? Can I as PC then find slut collar keys to help that poor girl (for a reward )?

 

Have you enabled key drops in MCM? If that's disabled, only the special/custom keys will drop.

And custom items should not get equipped on followers, except the steel shackles. Their AI wouldn't know how to handle them.

 

And yes, the person who wrote Cursed Loot is truly evil. When I started my last game, I was wearing shock boots, a parasite harness, a yoke and several other restraints including a gag by the time I reached Riverwood. Somebody should really tie her up so she can't code more evil things

Oh...wait...

 

Yeah, I imagine it would be rather difficult to implement bug fixes while restrained.

[Kimy]

This one : http://forums.bethsoft.com/topic/1501746-wipz-skyrim-script-extender-skse/page-7?do=findComment&comment=23721890

 

Well, seems slangster wrote a script that churned out 1k numbers generated by the RNG and had a look at whether or not every number was represented more or less equally (which it should). That's all fine and dandy, but doesn't tell us anything about the problem that people have when repeatedly initializing the RNG with the same values. e.g. from loading the same game over and over again. Skyrim's RNG does seem to deliver the correct distribution over time, but to me it seems to be -very- deterministic (which would indicate a serious seeding issue), which a good RNG shouldn't really be.  Anyway, seems we're stuck with it, then.

[Verstort]

Does Kimy keep a public repo for Cursed Loot?

 

I was curious if changes made were available/viewable as unofficial commits, for testing ect.

[EnragedRabisu]

Does Kimy keep a public repo for Cursed Loot?

 

I was curious if changes made were available/viewable as unofficial commits, for testing ect.

The source scripts are included in the download, though I can't speak for a public repo.

[Verstort]

Does Kimy keep a public repo for Cursed Loot?

 

I was curious if changes made were available/viewable as unofficial commits, for testing ect.

The source scripts are included in the download, though I can't speak for a public repo.

 

 

I know the source comes with the releases, I was more interested in the bug fixes that haven't been released yet actually (just impatient).

[Slorm]

On a point of curiosity, how are the weightings changed when looting a body rather than a container.

 

The reason I ask is that a Chaurus or Flashbang event doesn't trigger when looting a body so are the relative weightings for the other events recalculated, or does nothing happen if the RNG picks Chaurus/Flashbang when looting a body?

[Verstort]

On a point of curiosity, how are the weightings changed when looting a body rather than a container.

 

The reason I ask is that a Chaurus or Flashbang event doesn't trigger when looting a body so are the relative weightings for the other events recalculated, or does nothing happen if the RNG picks Chaurus/Flashbang when looting a body?

 

It's a weight system right? Doesn't that mean she can create the percentages on the fly even if she removes variables without leaving empty events?

 

Leaving events empty doesn't seem necessary, although that would explain why I'm seeing more events in my logs than I'm seeing in game...

 

speculation:

 

I thought a weight system behaved something like follows:

 

4 possible events, weights 10, 20, 30, 40: 100 total weight, percentages are the same as weights since 10/100 = 10%, 20%, 30%, 40%

 

remove one event:

3 possible events, weights 10,       30, 40:   80 total weight, percentages are 10/80, 30/80, 40/80,                        = 12.5%, 37.5%, 50%

 

The removed events changed the percentages but did not leave the possibility of a None or empty event.

 

 

[Slorm]

 

On a point of curiosity, how are the weightings changed when looting a body rather than a container.

 

The reason I ask is that a Chaurus or Flashbang event doesn't trigger when looting a body so are the relative weightings for the other events recalculated, or does nothing happen if the RNG picks Chaurus/Flashbang when looting a body?

 

It's a weight system right? Doesn't that mean she can create the percentages on the fly even if she removes variables without leaving empty events?

 

Leaving events empty doesn't seem necessary, although that would explain why I'm seeing more events in my logs than I'm seeing in game...

 

 

That's the reason I was asking. I was doing some testing on some bodies and noticed that having EC and Flashbang both at 100 and HS at 1 then HS should in fact become 100% but it didn't seem to always fire

 

[Verstort]

 

 

On a point of curiosity, how are the weightings changed when looting a body rather than a container.

 

The reason I ask is that a Chaurus or Flashbang event doesn't trigger when looting a body so are the relative weightings for the other events recalculated, or does nothing happen if the RNG picks Chaurus/Flashbang when looting a body?

 

It's a weight system right? Doesn't that mean she can create the percentages on the fly even if she removes variables without leaving empty events?

 

Leaving events empty doesn't seem necessary, although that would explain why I'm seeing more events in my logs than I'm seeing in game...

 

 

That's the reason I was asking. I was doing some testing on some bodies and noticed that having EC and Flashbang both at 100 and HS at 1 then HS should in fact become 100% but it didn't seem to always fire

 

 

 

I tried with only one equip event and with Flash bang + Estrus set high and got the same thing, log is full of cursed loot events but nothing happening in game.

 

However, when I turned Flash bang and Estrus to zero, I still get nothing, 8 bandits with 80% base chance and I see lots of event fired in log but nothing happens in game. Tried this with high sec and slave collar, nothing. Trying with new game now...

 

Edit: Something screwy here...

 

"[DCUR] Selecting event failed: Falling back to regular event." what are the conditions for getting this message?

 

Do we even have "Regular" and "Special" events anymore? with this new system, aren't regular events just a weight in "events"

 

Leaving FB + Estrus at high and 2-3 events all < 2% gives you a ~1/7 shot of getting a event, but the events are firing in debug more often then that, and almost everyone of them has the selecting event failed debug.

[Kimy]

On a point of curiosity, how are the weightings changed when looting a body rather than a container.

 

The reason I ask is that a Chaurus or Flashbang event doesn't trigger when looting a body so are the relative weightings for the other events recalculated, or does nothing happen if the RNG picks Chaurus/Flashbang when looting a body?

 

Some events cannot trigger from certain container types, indeed. Other events will be chosen, effectively neglecting the weights of any ineligible events.

[Coopervane]

Well, seems slangster wrote a script that churned out 1k numbers generated by the RNG and had a look at whether or not every number was represented more or less equally (which it should). That's all fine and dandy, but doesn't tell us anything about the problem that people have when repeatedly initializing the RNG with the same values. e.g. from loading the same game over and over again. Skyrim's RNG does seem to deliver the correct distribution over time, but to me it seems to be -very- deterministic (which would indicate a serious seeding issue), which a good RNG shouldn't really be.  Anyway, seems we're stuck with it, then.

 

To me it seems the RNG script can break under heavy game load.

 

I've had the same problems described by others, you go into a dungeon, and even though events should have a fairly tiny chance of occuring, every second or third container i touch is cursed. And not just that, but even though the amount of items you get should be random, it always gives you the maximum amount, and it's not just Dcur, the RNG is fiering like crazy so it's all RNG dependant functions going off constantly. When this is happening i can also forget all about using weapons, because DH will make me drop it constantly, and Spells are out too if i got a soulgem plug, because DDi will fire up the plug on every spell cast.

 

 

Here's the thing though: If i then fully quit the game and re-launch it, there's a good chance the problem goes away, and the RNG fires at normal rates once again, even in that same dungeon that had just caused me all those problems.

[Slorm]

I can certainly endorse Coopervane's comments. I've noticed this a few times where events constantly fire which a quit and relaunch fixes. From what Verstort is saying as well it looks as if there's a real problem with the RNG

[Coopervane]

Not quite Slorm, what it's suggesting is that the RNG is not the problem, but rather that Papyrus is.

 

The fact that a game re-launch can fix the issue suggests it's not the RNG script itself that's broken, but that it's script got suspended/broken in the previous play session due to script-lag or whatever else caused it. Sadly, this also means it would probably still happen even if someone tried to code a better RNG. If the actual problem is that Papyrus is a mess (which it is) and the game suspends the RNG if subjected to heavy load, then it would also suspend a different RNG script just the same.

 

 

However, SKSE does actually provide the option to load in C++ based scripts instead of Papyrus based ones, DDi uses that for a few functions, i belive SL does for a few things aswell, so maybe that could be a thing? Maybe an RNG coded in C++ and loaded via SKSE would perform more stable than the build in Papyrus based one?

 

 

Well, maybe..

[unmog]

To me it seems the RNG script can break under heavy game load.

 

I've had the same problems described by others, you go into a dungeon, and even though events should have a fairly tiny chance of occuring, every second or third container i touch is cursed. And not just that, but even though the amount of items you get should be random, it always gives you the maximum amount, and it's not just Dcur, the RNG is fiering like crazy so it's all RNG dependant functions going off constantly. When this is happening i can also forget all about using weapons, because DH will make me drop it constantly, and Spells are out too if i got a soulgem plug, because DDi will fire up the plug on every spell cast.

 

 

Here's the thing though: If i then fully quit the game and re-launch it, there's a good chance the problem goes away, and the RNG fires at normal rates once again, even in that same dungeon that had just caused me all those problems.

 

Heh, I never really considered that a problem, but now that I think about it, Ive probably experienced the same issues then. I might go a bunch of dungeons without getting hit once [i tried putting the odds to something realistically lower] then all of a sudden get destroyed by one, randomly. I figured that was just how random went though.

 

However, SKSE does actually provide the option to load in C++ based scripts instead of Papyrus based ones, DDi uses that for a few functions, i belive SL does for a few things aswell, so maybe that could be a thing? Maybe an RNG coded in C++ and loaded via SKSE would perform more stable than the build in Papyrus based one?

 

Thats just crazy enough to work! I admit though, coding is over my head. Its all the same to me, even though I know its not really the same at all. Whys C++ better, and whys papyrus bad? It isnt the first time Ive heard complaints about Papyrus and problems related to it though. Still, if most of these mods require SKSE anyway would it really hurt if some of the codes used one and some used the other? Or do they all need to use the same language to effectively communicate to each other? If it doesnt matter, then maybe more people could try and see if it helps or hurts anything. If it DOES matter... how hard might it be to get people all to use the same one? >.>

[Coopervane]

I'm not a programmer either, so i can't give you an expert description of the issues, but here is what i can offer:

 

 

Why is Papyrus bad? Because it's a script language Bethesda created for Skyrim, but never really finished. It's only got the most basic features implimented and working, only just what Bethesda had to do to make the vanilla game work, nothing more, and it's not very optimized. This means that when modding the game, trying to expand on it's features and create something new, it's actually really hard to do, and often requires jumping through fiery hoops to get it working that you woulden't have to with a more mature and fully featured script language. Looking at code written in Papyrus, it's often many times longer, with many more lines of code than a similar script written in Lua-Script, C++, uScript and other such mature and fully featured languages, because the Papyrus code has to constantly insert workarounds for missing functions. More lines of code equals worse performance.

 

Why is C++ good? Because it's a mature and fully featured script language, it's optimized, it's clean and it works.

 

 

But the script language itself is not the only factor here. Plainly C++ is far superiour to Papyrus, nobody could ever deny that, but at the end of the day the scripts alone are not the only factor, there's also the game engine and how it interacts with them. How much faster are C++ scripts really when the game engine first has to interprit them via SKSE?

The tests done by Ashal and Min suggest there are noticable gains, even having to pipe things through SKSE first, they were seeing noticable cutdowns on latency when processing C++ instead of Papyrus scripts. So that seems promising, right?

 

 

But is it enough to make the RNG behave itself? Well that's the thing, the RNG itself might not be the problem, seeing the RNG fail might simply be the only obvious symptom of something else, something even earlier on the chain of engine functions failing first. And if that's the case, an improved C++ RNG script woulden't make any difference, because the problem is elsewhere and happening even before the RNG script gets processed.

[Rorahusky]

I have a suggestion for a new feature for this mod. Namely increasing the chance a chest or door is boobytrapped with Devious Items based in relation to if it's locked or not, and how difficult the lock might be to pick. The easy and average locks for example would not only be easier to unlock, but also to disarm to keep from being ensnared, while the harder locks are more likely to end up with you in bondage, and that having the Novice/Adept/Etc lockpicking feats further decreases this chance for their respective locks.

[RandomOne]

It would be awesome if the drug effects from skooma whore were added. like , say, you open a chest and you get a msg about a needle pricking your hand as you reach in or a msg that a gas cloud released and then you get double vision or crazy colors like the drugs from skooma whore. being drugged while getting devices would certainly be a new challenge in the middle of a draugr burrow . Just a thought Oh... love the mod by the way

[Kimy]

But is it enough to make the RNG behave itself? Well that's the thing, the RNG itself might not be the problem, seeing the RNG fail might simply be the only obvious symptom of something else, something even earlier on the chain of engine functions failing first. And if that's the case, an improved C++ RNG script woulden't make any difference, because the problem is elsewhere and happening even before the RNG script gets processed.

 

Adding a C++ RNG to SKSE was exactly what the SKSE devs originally planned to do if I am not mistaken. But apparently that idea got scrapped after they did a (from what I can tell what they did, very incomplete) test on the vanilla RNG which showed it to be working fine in ONE of the criteria of a good RNG, which is distribution. The RNG's main issue seems to be seeding though, and that they didn't test.

 

I had repeatedly situations when an event having a very low probability to happen triggered repeatedly in a row. Yes, math allows that to happen. But in Skyrim it happens enough to make me raise my eyebrows. As Coopervane indicates, we also don't really know other factors that might play a role. For example, it -could- be that the RNG gets assigned only very limited CPU time under heavy engine load (which a lot of us produce with our heavily modded installations), which might result in poor quality of numbers generated by the RNG. I really don't know if that's the case or not as that would require thorough testing.

[thetwo2012]

 

 

I've had the same problems described by others, you go into a dungeon, and even though events should have a fairly tiny chance of occuring, every second or third container i touch is cursed. And not just that, but even though the amount of items you get should be random, it always gives you the maximum amount, and it's not just Dcur, the RNG is fiering like crazy so it's all RNG dependant functions going off constantly. When this is happening i can also forget all about using weapons, because DH will make me drop it constantly, and Spells are out too if i got a soulgem plug, because DDi will fire up the plug on every spell cast.

 

Here's the thing though: If i then fully quit the game and re-launch it, there's a good chance the problem goes away, and the RNG fires at normal rates once again, even in that same dungeon that had just caused me all those problems.

 

 

Honestly, I'd guess that there are two problems. One is the clustering illusion - if you ask someone to make or evaluate a string of random numbers, they tend to *strongly* prefer ones with less clustering than you'd actually get from true randomness.

 

The second is that the default settings for Deviously Cursed Loot are hugely influenced by arousal. I mean, the "base" chance is only something like 5%, but if you have 100 arousal that gets boosted to something like 50%. And there is a setting that always forces the maximum amount of devices over a certain arousal as well, enabled by default. Which would totally explain why sometimes it feels like half the chests have cursed loot, and you always get the maximum amount of devices.

 

If you really feel like you're getting screwed by RNG, I'd suggest turning on the debug settings, which will show you the "actual" chance for getting loot on any particular item you open, after all modifiers. You might be surprised by the result.

[Slorm]

 

Honestly, I'd guess that there are two problems. One is the clustering illusion - if you ask someone to make or evaluate a string of random numbers, they tend to *strongly* prefer ones with less clustering than you'd actually get from true randomness.

 

The second is that the default settings for Deviously Cursed Loot are hugely influenced by arousal. I mean, the "base" chance is only something like 5%, but if you have 100 arousal that gets boosted to something like 50%. And there is a setting that always forces the maximum amount of devices over a certain arousal as well, enabled by default. Which would totally explain why sometimes it feels like half the chests have cursed loot, and you always get the maximum amount of devices.

 

If you really feel like you're getting screwed by RNG, I'd suggest turning on the debug settings, which will show you the "actual" chance for getting loot on any particular item you open, after all modifiers. You might be surprised by the result.

 

 

I don't have arousal in play (it's set to zero), I have the base chance set to 5% and have still seem unusual number distrbution. Indeed the debug shows the percentage rolled against and (oddly) on occassions DCUR events fire but there is no action in game.

 

I do accept that there will be clusters but nevertheless the results from Skyrim seem inordinately "lumpy" at times, to the point of repeatability. It's academic in any case without a replacement RNG to test against

 

[Coopervane]

Honestly, I'd guess that there are two problems. One is the clustering illusion - if you ask someone to make or evaluate a string of random numbers, they tend to *strongly* prefer ones with less clustering than you'd actually get from true randomness.

 

The second is that the default settings for Deviously Cursed Loot are hugely influenced by arousal. I mean, the "base" chance is only something like 5%, but if you have 100 arousal that gets boosted to something like 50%. And there is a setting that always forces the maximum amount of devices over a certain arousal as well, enabled by default. Which would totally explain why sometimes it feels like half the chests have cursed loot, and you always get the maximum amount of devices.

 

If you really feel like you're getting screwed by RNG, I'd suggest turning on the debug settings, which will show you the "actual" chance for getting loot on any particular item you open, after all modifiers. You might be surprised by the result.

 

It's not clustering illusion, and it's not arousal either as i have that feature disabled, aswell i'm running lower than default chances for everything else (i belive the default base chance is 5%, and i've got it at 3%, or is it 0.5% and i have it at 0.3%... eh, one of the two).

 

At such settings, getting hit by 3 to 5 Dcur events whilst going through the Bleakfalls dungeon would be a supricingly high number, but not an impossible one, 3 to 5 could be a legit cluster, it could happen. That is not what happened to me on this game start however, no, i got hit with somewhere between 30 and 50 Dcur events going through that dungeon. That's not hyperbole, this actually happened to me.