[RELz/WIPz] Unofficial Skyrim Patch

[Arthmoor]

Is this about the enchantment potion fix that went in? If that change is going to result in an exploit of similar magnitude to the one we already unintentionally fixed then that stuff is going to have to come out. Cause, yeah, that would look kind of dumb on our parts.

[Screwball]

Hm interesting. I would have guessed you could at least get a few percent extra with the new change. So just to ckarify: with the new 2.0.1 beta, the values did not change compared to december?

 

 Unknown - yet. Those were the max values I could achieve at the time (USKP 2.0.0, last release). I'll have a play today (with USKP 2.0.1 Beta, latest) starting with that savegame, and see what difference, if any, it makes.

Considering I replied at 1AM I wasn't in a position (or condition) to start testing immediately

[Screwball]

Starting point as above - with USKP 2.0.0, wearing 4 * 28% Fortify Alchemy items, Fortify Enchanting potions would be 32% and items of Fortify Alchemy (using those potions) would be 28%.

 

With USKP 2.0.1 Beta (latest), results are as follows, using the most recently created potions and items each time.

 

Iteration 1.

Fortify Enchanting Potion - 32%. No Change.

Items of Fortify Alchemy - 36%. + 8%.

 

Iteration 2

Fortify Enchanting Potion - 37%. + 5%.

Items of Fortify Alchemy - 37%. + 1%.

 

Iteration 3

Fortify Enchanting Potion - 37%. No Change.

Items of Fortify Alchemy - 37%. No Change.

 

So yes, diminishing returns even with 2.0.1 Beta, however, the effect of this on other items is considerable.

After the 3rd iteration above, Fortify Smithing (potion) rises from 127% to 149%, enchanted items from 29% to 37%.

 

Improving Glass Gauntlets (base armor rating of 22) with 2.0.0 rises to 115, but with 2.0.1 Beta (using the 3rd iteration potion/items) to 133.

Enchanting items of Fortify Archery rises from 47% to 61%. Given that you can wear 4 of them, the overall effect rises from 188 to 244 - a very noticeable improvement.

 

With 2.0.0 a Daedric Bow, improved has a Damage of 195. With 4 items of 47% Fortify Archery that rises to 562.

With 2.0.1 Beta 227, + 4 items @ 61%, 782. That's a lot...

[VaultDuke]

well, nice to hear it at least has diminishing returns.

i bet the armor rating on the gauntlet was without taking improved armor skill gear into account. a rise from 115 to 133 is quite okay, while the rise for the dame you showed is pretty insane.

i personally feel this little change has too wide implications on the balance the way it looks. it will probably be balanced for most enchantments like resistances and stats,w ho do not benefit from any multiplicative stacking. but the effect will be very profound on weapons and armor, where the weapon damage enchantments benefit from this as well as the tempering. in both cases from an improved potion and apparel. and this is on top of the already improved potion application formula.

[Screwball]

Does seem to be working as expected re. diminishing returns.

The armor values are indeed without any fortify armor enchantments or potions.

Much as I like powerful weapons, even I think that much of an increase is a bit much (though there are times when it would come in handy ).

[Jumps-Down-Stairs]

It's not that the USKP avoids making changes that will affect the balance of the game -- if they're clearly not working as intended, then they are appropriate for the project. It just doesn't make change solely for the sake of game balance. The USKP has fixed exploits before, to the occasional loud complaint that someone's favorite toy was taken away. Equally, if something is not providing its intended benefit, then it has and should be fixed.

 

Fortify Enchanting potions had minimal effect before, unless you used the Enchanting/Alchemy loop to achieve very high values. Abusing the game mechanics already results in extreme weapon damage -- I'm not sure the increase from enchanting potions actually working as described will even be noticeable at the top end. The toughest enemies in the game have only a few thousand health, and it doesn't matter whether you're hitting them with 562 damage or 782, they'll still go down in 3 or 4 hits. It's a Bethesda game -- If you set out to break it, they didn't do much to try to stop you. The change primarily benefits characters with low Enchanting skill, for whom Fortify Enchanting potions previously had virtually no effect.

[Arthmoor]

Threw down another beta update.

 

More critter script and weapon rack script updates. At this point those are probably the primary things left to poke at some more. Everything else appears to be doing well.

 

One other thing in this, we've taken out the edits to Heimskr's preaching scenes since our fixes were not correcting the problem. Looking like we may just have to live with this one. I'm sure everyone is just sad to know he won't always be preaching

[Screwball]

Just applied the latest Beta.

[02/02/2014 - 11:59:20AM] Reverting game...
[02/02/2014 - 11:59:22AM] Loading game...
[02/02/2014 - 11:59:22AM] warning: Variable ::DialogueWhiterunHeimskrPreachScene_var on script uskpretroactive201script loaded from save not found within the actual object. This variable will be skipped.
[02/02/2014 - 11:59:22AM] Cannot open store for class "USKP_AniseAliasScript", missing file?
[02/02/2014 - 11:59:22AM] warning: Unable to get type USKP_AniseAliasScript referenced by the save game. Objects of this type will not be loaded.
[02/02/2014 - 11:59:22AM] warning: Could not find type USKP_AniseAliasScript in the type table in save
[02/02/2014 - 11:59:22AM] warning: Function Critter.KickOffOnStart.OnUpdate in stack frame 0 in stack 218732 differs from the in-game resource files - using version from save
[02/02/2014 - 11:59:22AM] warning: Function Critter.KickOffOnStart.OnUpdate in stack frame 0 in stack 218731 differs from the in-game resource files - using version from save
[02/02/2014 - 11:59:22AM] warning: Function Critter.KickOffOnStart.OnUpdate in stack frame 0 in stack 218734 differs from the in-game resource files - using version from save
[02/02/2014 - 11:59:22AM] warning: Function Critter.KickOffOnStart.OnUpdate in stack frame 0 in stack 195067 differs from the in-game resource files - using version from save
[02/02/2014 - 11:59:22AM] warning: Function Critter.KickOffOnStart.OnUpdate in stack frame 0 in stack 195055 differs from the in-game resource files - using version from save
[02/02/2014 - 11:59:22AM] warning: Function Critter.KickOffOnStart.OnUpdate in stack frame 0 in stack 195068 differs from the in-game resource files - using version from save
[02/02/2014 - 11:59:22AM] warning: Function Critter.KickOffOnStart.OnUpdate in stack frame 0 in stack 195056 differs from the in-game resource files - using version from save
[02/02/2014 - 11:59:22AM] warning: Function Critter.KickOffOnStart.OnUpdate in stack frame 0 in stack 218735 differs from the in-game resource files - using version from save
[02/02/2014 - 11:59:22AM] VM is thawing...

None of the above warnings/errors were in previous logs.

Lots of critterfish spam ([None].ObjectReference.Is3DLoaded()) and similar, though these existed prior to applying the update. 400,000+ lines of it in the last log
 

Note. Everything above, except for Heimskr, appears in later logs too.

[BlackPete]

Just applied the latest Beta.

 

< ... >

 

None of the above warnings/errors were in previous logs.

Lots of critterfish spam ([None].ObjectReference.Is3DLoaded()) and similar, though these existed prior to applying the update. 400,000+ lines of it in the last log

 

Perhaps the "Cannot open store for class "USKP_AniseAliasScript", missing file?" error has to with one of these issues (14162, 14193, or 14697).

[Arthmoor]

USKP_AniseAliasScript was intended to advance her location quest to stage 40 on death. Turned out the script wasn't necessary because her reference has it attached directly. So just ignore that one as it won't be in the live release.

 

The Heimskr thing is also expected since we backed out his changes. That guy can do whatever he wants now

[Screwball]

And the critter.kickoffonstart.onupdate? Those didn't appear until I applied the update you posted yesterday. Prior to that there were critter problems, but not those specific ones.

[Arthmoor]

Those are Papyrus telling you the obvious - that you had critters stuck using the old scripts and that it's going to ignore all logical processes and use the old broken shit instead of the new fixed shit.

 

Welcome to baked in data. The biggest clusterfuck of all time.

 

Also, 400K logs seems excessive. How long of a test were you running?

[Screwball]

Ok - that's a bit of a bugger then, because I really don't want to have to redo 2 full days of gameplay

This was supposed to be a test playthough for the USKP 2.0.1 Beta, but I've been beset by so many log errors I can't see the wood for the trees

 

Not 400KB of log - 400,000+ LINES (mostly critter/none errors) within the log.

For the session I just finished:-

(Line 1) [02/02/2014 - 09:10:55PM] Papyrus log opened (PC)
(Line 620,286) [02/02/2014 - 09:44:02PM] Log closed
620 thousand log lines in around 35 mins.

[Deathbydestiny]

Don't feel bad, I'm loosing out on a 2 week play through upgrading to the next USKP.

 

You boys just do what you do best though, it's always worth it.

[Arthmoor]

That's very strange nonetheless. I've got 4 different chars who have been migrating through several updates and are of varying ages and have never had a USKP update begin generating hundreds of thousands of lines like that.

 

You weren't using any other outside attempts to fix the critters, were you?

[DayDreamer]

And the critter.kickoffonstart.onupdate? Those didn't appear until I applied the update you posted yesterday. Prior to that there were critter problems, but not those specific ones.

I can explain that one. You saved so quickly after entering a new space that 8 critters were still spawning.

 

Those are 1 time warnings, telling you the old code was used instead of the new code. I moved common code out of the 5 critter scripts into the kickoff code. This was a combined effort with a similar change by STEVE40.

 

Since (as you can see by comparing and because I worry about backward compatibility and even kept the old indentation for easy unified diff) it is the same code, you will call into the OnStart() and do the same things. No harm no foul.

 

Why move the common tests? Because then they fail more quickly, not even calling the critter-specific code:

    ; STEVE40+USKP consolidate duplicated code for most critters
    if landingMarker && dummyMarker && CheckCellAttached(self) ;/&& Game.GetPlayer().GetDistance(self) <= fMaxPlayerDistance/;
        OnStart()
    else
;~        Debug.Trace(self + " Unable to place any markers.")
        ; Kill this critter
        DisableAndDelete()
    endif

 

Also, because birds were badly broken. As Sclero pointed out early in the critter topic, they could fail, be disabled and deleted, and then be re-enabled and add to bloat. Very bad code.

 

EDIT: Also, note that "<= fMaxPlayerDistance" commented out? You'd enter the exterior on 1 edge of a cell, and 3 cells away the critters would spawn, and immediately delete themselves because they were too far away. And then you'd wander the fields (or go by on a carriage) and they'd never respawn again until you came back the other way. Instead, now the code only checks to make sure they are in a valid cell. You're going to see a lot more critters!!!

 

Now, what does the repeated critter spam look like? I need to see the whole log in context. And the savegame, too.

[Arthmoor]

I can vouch for seeing a whole lot more critters, even with just what's in the beta. If there's still yet more code to come for the live, it ought to get interesting.

 

I can also say that aside from the cleanup of ages old critters, I haven't seen any signs of new errors at all. Sadly it does not appear to have had a corresponding reduction in the save size (yet). Maybe that just needs more time or perhaps my saves just have way to much garbage in them that can never be removed.

 

(now if Wet & Cold would stop spamming my logs.....)

[Screwball]

Ok, looks like I spoke too soon - the None errors did occur in the savegame immediately prior to installing the latest Beta update.

 

Save 70 - Riften, outside Honeyside. Honeyside not yet purchased, so could not be an exit to Skyrim followed by an immediate save.

Log 105 (exit game after making save 70) is a mere 12KB, no critter spam.

 

Save 71 - also Riften (log 106) has critter errors from the start. So loading save 70 (which had no critter errors in the log) causes immediate errors when reloaded. Log 106 has 546,000+ lines and is 31,676KB. This seems to be consistent here - load savegame 70 and the critter spam appears instantly

 

Latest Beta applied after saving game 72, so not the cause of the None errors as I first thought. Sorry about that.

(But see my Edit below).

 

@DayDreamer. Which save(s) and associated log(s) do you want so I can upload them?

 

@Arthmoor. No "outside attempts"? If by that you mean external tools modifying savegames then No (because I don't have any such tools).

Much earlier in the game (around savegame 17/18) I got rid of those existing at the time by loading an earlier game and using COC to avoid entering the areas which were causing the problems (mainly outside Lost Knife Hideout). So even that was not "fixing" existing issues - it was circumventing them so they couldn't happen in the first place.

 

[Edit]

Maybe I spoke too soon again. Just reloaded saves 68 and 69 - and they are producing critterspam immediately. I've checked back in the associated logs and they are critterspam free. The only difference between now and when those saves/logs were created is the version of the Beta. My history of the game is as follows:-

USKP 2.0.1 Beta, SBBasement, no killcams from the start
Updated USKP Beta (Jan 30th) after save 57.
Updated USKP Beta (Feb 1st) after save 72.
[/Edit]

[DayDreamer]

I can vouch for seeing a whole lot more critters, even with just what's in the beta. If there's still yet more code to come for the live, it ought to get interesting.

The only code I know of for the live is adding an extra wait and check to try to make the log entry reported over on BS forum more rare. And the f variable fix. So this is hopefully very very close to done. That's why I need the Screwball logs.

 

I can also say that aside from the cleanup of ages old critters, I haven't seen any signs of new errors at all. Sadly it does not appear to have had a corresponding reduction in the save size (yet). Maybe that just needs more time or perhaps my saves just have way to much garbage in them that can never be removed.

I see 104K drop just walking from Riften into Honeyside. Huge cleanup for me. And no log entries from the old critters.

There are those 8 fireflies in the infinite loop for 3D, it tells me about them on every load. All I have to do is figure out where they are, and they should go away.

 

(now if Wet & Cold would stop spamming my logs.....)

yeah, I tried it testing with carriages because folks complained about my scripts seeming slow -- wow, W&C really slows everything! (So I don't use it for my play.) And I really like Splash of Rain, and intend to use it for my next run, but when it's raining and I go into Warmaidens, I can see the individual items rack one at a time as SoR cleans up....

 

@DayDreamer. Which save(s) and associated log(s) do you want so I can upload them?

Just pick one that seems to give a lot of spam. Post the log and the savegame. All I'm trying to do is figure out where your spam comes from, and whether it can be reproduced.

[Screwball]

Done. Upload is here. Contains 1 savegame and 2 logs.

 

The SaveGame (70) originally created log 105 (small). This was with the version of USKP Beta released on Jan 30th.

The next game (i.e. loading savegame 70) produced log 106 (big).

 

Loading that same savegame now (with the Beta release of Feb 1st)  produces similar logs to 106 - the longer you're in-game the larger they will be.

[DayDreamer]

Thanks. Unfortunately, it's not likely to be helpful.... Those are all [None].

 

You have 4 just spawned critters.

[02/03/2014 - 04:36:33PM] warning: Function Critter.KickOffOnStart.OnUpdate in stack frame 0 in stack 195067 differs from the in-game resource files - using version from save
[02/03/2014 - 04:36:33PM] warning: Function Critter.KickOffOnStart.OnUpdate in stack frame 0 in stack 195055 differs from the in-game resource files - using version from save
[02/03/2014 - 04:36:33PM] warning: Function Critter.KickOffOnStart.OnUpdate in stack frame 0 in stack 195068 differs from the in-game resource files - using version from save
[02/03/2014 - 04:36:33PM] warning: Function Critter.KickOffOnStart.OnUpdate in stack frame 0 in stack 195056 differs from the in-game resource files - using version from save

And that seems to track, although they are all reported as [None]

[02/01/2014 - 10:06:36PM] error: Unable to call Is3DLoaded - no native object bound to the script object, or object is of incorrect type
stack:
    [None].ObjectReference.Is3DLoaded() - "<native>" Line ?
    [None].critterdragonfly.OnUpdate() - "Critter.psc" Line 259
[02/01/2014 - 10:06:36PM] error: Unable to call Is3DLoaded - no native object bound to the script object, or object is of incorrect type
stack:
    [None].ObjectReference.Is3DLoaded() - "<native>" Line ?
    [None].critterdragonfly.OnUpdate() - "Critter.psc" Line 264
[02/01/2014 - 10:06:36PM] error: Unable to call Is3DLoaded - no native object bound to the script object, or object is of incorrect type
stack:
    [None].ObjectReference.Is3DLoaded() - "<native>" Line ?
    [None].critterFish.OnUpdate() - "Critter.psc" Line 264
[02/01/2014 - 10:06:36PM] error: Unable to call Is3DLoaded - no native object bound to the script object, or object is of incorrect type
stack:
    [None].ObjectReference.Is3DLoaded() - "<native>" Line ?
    [None].critterFish.OnUpdate() - "Critter.psc" Line 259

Lines 259 and 264 are the old lines I've already identified as an infinite loop on your Save 17: http://www.afkmods.com/index.php?/topic/3781-the-critter-thread/?p=149647

 

But instead of 3 fireflies, you now have 2 dragonflies and 2 fish.

[Screwball]

And a log folder rapidly approaching 1GB.

 

Much as I hate to do it I may as well just abandon this game totally. There's no way I can look for odd errors with the logs so full of junk that won't go away.

 

Just for the hell of it I just loaded Save 3 - made just after exiting the tunnels under Helgen.

[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5630 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5627 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5617 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5628 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5625 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5626 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5623 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5618 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5624 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5629 differs from the in-game resource files - using version from save

How the heck can those be there? Haven't had chance to move yet and they're already having issues? Something is really not right with these damn critters... Or is that because of script changes not working (baked in issues) right from the start of the game?

[DayDreamer]

And a log folder rapidly approaching 1GB.

 

Much as I hate to do it I may as well just abandon this game totally. There's no way I can look for odd errors with the logs so full of junk that won't go away.

 

Just for the hell of it I just loaded Save 3 - made just after exiting the tunnels under Helgen.

[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5630 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5627 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5617 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5628 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5625 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5626 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5623 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5618 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5624 differs from the in-game resource files - using version from save
[02/03/2014 - 11:04:42PM] warning: Function critterspawn..OnCellAttach in stack frame 0 in stack 5629 differs from the in-game resource files - using version from save

How the heck can those be there? Haven't had chance to move yet and they're already having issues? Something is really not right with these damn critters... Or is that because of script changes not working (baked in issues) right from the start of the game?

Pretty simple, really. Those aren't critters, those are critter spawners. Yes, we fixed bugs in the spawners. It will sit in the old code until it spawns, or you leave the cell.

 

Those fireflies/moths/fish that are "cluttering your logs" -- they were sitting there and looping invisibly 10 times per second -- forever and ever. That's a USKP 1.3.3 bug that's now fixed. Abandoning that game makes sense.

 

But the only way those fairly rare bugs happen is some very odd things in the first place. Perhaps two fast travels back-to-back with no real time in between. Or a fast travel and then going through a door and then saving, again without enough time for the scripts to keep up....

 

I've been testing by going in and out and in and out and in and out through the same door. I've got the creation/deletion scripts going much faster now. I don't get any errors anymore.

 

But as I mentioned on another thread, no matter how fast I make these scripts, there are others that are very slow. For example, going in a door from outside in the rain, and I can see the weapon racks fill.

 

I had to up the CheckFor3D loop from 5 seconds to 8 seconds (latest code, not yet posted). There are just too many places that cannot load all the objects in the 5 by 5 group of cells in under 5 seconds!

 

Yet 99% of the time, they load in 0.01667 seconds. That's a huge range of behavior!

[Arthmoor]

How the heck can those be there? Haven't had chance to move yet and they're already having issues? Something is really not right with these damn critters... Or is that because of script changes not working (baked in issues) right from the start of the game?

Cart ride. I suspect there's quite a few things that somehow get pushed down on the priority list.

 

Also, your logs more or less prove something I've always suspected. Stack numbers build up from the start of the game. So I *DO* have real reasons to worry when a DPS command tells me stuff in my stacks have numbers into the 30 million range. If those are all real stacks with real issues, I ain't gettin my stuff fixed.

[DayDreamer]

Cart ride. I suspect there's quite a few things that somehow get pushed down on the priority list.

In his case, they are probably just spawners scattered in 5x5 around Helgen. They go away (and have always gone away) OnCellDetach, so those stacks will clean up as he walks either toward them or away from them.

Also, your logs more or less prove something I've always suspected. Stack numbers build up from the start of the game. So I *DO* have real reasons to worry when a DPS command tells me stuff in my stacks have numbers into the 30 million range. If those are all real stacks with real issues, I ain't gettin my stuff fixed.

Only if you have 30 million concurrent stacks. I haven't figured out the memory re-use policy, but it seems everything is renumbered on LoadGame. The holes get filled. They do seem to remain in the same relative order.