Install options discussion

[Mike1072]

EDITED by Demi: it all started here because of me, but it's growing out of control

 

Long Story: ...

By bundling the armor resistance changes with other changes, we would limit player choice and compatibility. Everyone who might want the armor resistance changes would be forced to take the other penalties, and everyone who might want the other penalties would be forced to take the armor resistance.

Let's start from the beginning:

- vanilla game was so aware that heavy armors (full plate in particular) were OP that it did not granted players highly enchanted specimens, while getting +5 or even +6 specimens of light and medium armors was stupidly easy (because that was the only way to make them even remotely appealing)

- IR's main component assumes some sort of penalties are added to heavy armors to even the odds and grants players highly enchanted heavy armors

--> if no penalty is installed IR is actually responsible of breaking the game balance more than vanilla did

--> If the user ignores the readme that states how components are balanced, then the user is responsible for "breaking" the game balance.

 

So long as it's clear that IR's main component is balanced according to having certain global changes installed, anyone who is looking to IR for balance will use those components or accept the consequences. I don't see why we need to ensure that no component could ever individually be used (or skipped) to upset the balance - that is policing how players use the mod. All we need to do is advise what players should install if they want to make the game more balanced.

[Salk]

That being said, if you agree with Mike you are actually agreeing on allowing IR to potentially make the game extremely easier when certain options are installed. As I explained above, even without physical resistance DEX penalties does not make the game harder, while no penalty options will make the party even more OP.

 

I would not choose options that would make the game extremely easier but I wouldn't want the default, recommended options to make the game any harder. From your own admission, the tendency of IR is to make the game slightly harder for the player and in my own experience it's been even too zealous in that. IR tries very much to nerf down those items that in the vanilla game make the player have an edge over the AI much more than trying to make weaker items become more powerful.

 

This happens because you (Demi and Arda) and the ordinary IR player (giving feedback and suggestions) are BG(2) veterans/experts, knowing every nook and cranny about each enemy's strategy, powers and weaknesses, often installing this modification together with SCS and playing on difficulty higher than Core (which is my reference).

 

The result is that IR makes in the end the game quite a bit more challenging when instead I feel it should be more neutral than this. In the next quote I am going to address my main issue.

 

Just because Charname is a Child of Bhall doesn't mean he has the strenght of an ogre, dexterity of a cat and is built like a rhino.

 

This is exactly my point. Veteran players accept, or rather encourage, a balance shift heavily tipping the odds towards the AI so it sounds simply ludicrous to think of Charname having the same powers of, for example, Sarevok or the other hostile children of Bhaal. It's okay instead to face specific enemies that have instant casting or that have "the strength of an ogre, dexterity of a cat and is built like a rhino". The god-like powers/items in the end are only expected to be used against you. As soon as a vanilla item would take the player anywhere close to it then we scream "it's practically cheating!" But it's no cheating since it is something the developers put in the game for the player to use. It was their way to counter the odds which are already at Core (vanilla) against the survival of the player. But what if my goal as player is to survive and not just to finish the game?

 

Not to mention that the use of some specific, admittedly overpowered items (which, just to make clear, I also like to see nerfed), are usually (too) effective against one type of enemy (ex. shield of Balduran) or in very specific contexts ("Item A is so overpowered because a character belonging to class B using it together with item C would just make the game such a walk in the park!") while in all other cases are only moderately good. As player, I am openly and strongly against any gaming design which includes metagaming.

 

All this rant to mean, let the players customize their challenge as much and as freely as they like.

 

Do it through .ini modifications like SCS does, if you like, but please let it happen, because - and I speak for myself here - without customization I'd not personally install some specific components (ex. IR's shields or SCS's arcane/divine AI enhancements) because I don't agree with their default behavior. I am sure I represent a minority but I still think you care about other people's gaming style and try to accommodate different perspectives, if valid.

[Jarno Mikkola]

I don't know anything about the specific issue, but I've confronted the general problem a lot with SCS: people want particular options but indulging all the requests clutters the install process and can lead casual users to make bad installs.

 

My eventual solution was the ini file: fine-tuning and not-author-recommended options end up there. Casual users don't see it but people who want advanced options can make appropriate choices there.

^This

I am really unhappy with that as it comes to the fact that the player needs to be able to edit the .ini file before they install the mod... if it was done so that the choices would matter after the install, I would be fine, but the way the weidu.exe work it won't... So I vote for the Mike's solution, where you get to choose from the default or from the custom... you can have two defaults or more... but that doesn't really matter for me, as I will never take an option in which heavy armor will reduce thieving skills for the multi&dual classes, simply because it will only add to my in game clicking, it will not actually do a thing more.

[Ardanis]

I'd like to clarify that the discussion is not whether there should be an option to customize the component. It will be possible simply because of its modular structure controlled by a variable statement.

///////////////////////////////////////////////////////////////////////////
// Movement Speed, Dexterity and Speed Factor Penalties

BEGIN @25
DESIGNATED 1040
SUBCOMPONENT @9
REQUIRE_PREDICATE (ENGINE_IS ~soa tob totsc~) @9002

OUTER_SET move = 1
OUTER_SET dex  = 1
OUTER_SET sf   = 1
INCLUDE ~item_rev/components/heavy_armor_encumbrance.tpa~

 

 

The question is how that customization should be presented to the player. There're two things to consider:

1) There're experienced long time players who want "this, that and don't you tell me how I should play Improved Anvil your mod". Given the tools we're using for distribution and installation, that's to be expected.

2) And there're novice players who often answer "[Y]es" to the "Install All" prompt and then wonder what's wrong with their game. This happens everywhere every time too.

 

By providing a couple of default options we ensure a new player won't get a not recommended installation. This is no different from a game itself, you get what designers saw fit to give you - for reasons I don't have to explain (I hope ). You don't see a ton of flags in the options menu to control the game on the content level, instead you get a moddable architecture to customize manually with selected mods.

Demi, David and I are suggesting exactly the same approach, only applied to a large mod rather than the underlying base game.

[DavidW]

I am really unhappy with that as it comes to the fact that the player needs to be able to edit the .ini file before they install the mod...

 

The player needs to be able to edit the .ini file before they install the mod if they want to do advanced customisation. If they're happy to choose from the range of installs that the mod author recommends, they don't need to alter the ini. If they want to make an advanced selection of options and they've studied the readme before installing, they know what to do. If they want to make advanced and not-author-recommended selection of options but they haven't read the readme, then (at least as far as SCS is concerned) tough.

[Jarno Mikkola]

///////////////////////////////////////////////////////////////////////////
// Movement Speed, Dexterity and Speed Factor Penalties

BEGIN @25
DESIGNATED 1040
SUBCOMPONENT @9
REQUIRE_PREDICATE (ENGINE_IS ~soa tob totsc~) @9002

OUTER_SET move = 1
OUTER_SET dex  = 1
OUTER_SET sf   = 1
INCLUDE ~item_rev/components/heavy_armor_encumbrance.tpa~

The how I would approach the same thing would be pretty much the exact same thing, and the install all would install the default, dah... just adding a different sub component to edit the 3 outer_set options if it's chosen... there's no need to include those in a .ini file as there no advantages from them, as they can be defined during the install anyways if the custom install option is chosen.

[kreso]

I guess you could do as DavidW does - put even components such as "only physical resistance, no penalties" under regular install options and write something as ("not reccomended!") in brackets - this way everybody will see it, even if they don't feel comfortable editing the .ini file and do as they please (I do install SCS "All mages get HLA's in SoA" which isn't reccomended install for example, but it makes me feel good ).

This way, you'd ensure easy compatibility with BWS as well. However, instaling IR on BWS can (and does) lead to disasters such as a bunch of Gnolls wielding stunning halberds and getting +2 crippling throwing daggers and similar stuff in BG1 .

[DavidW]

there's no need to include those in a .ini file as there no advantages from them, as they can be defined during the install anyways if the custom install option is chosen.

The advantages are (i) it makes the install order less cluttered; (ii) I think there is some reasonable expectation among installers of a mod that components which they are offered are in some sense approved by the mod author as functional and playable; (iii) relatedly, it is helpful to make sure that options that you shouldn't install unless you've read the instructions aren't available unless you've read the instructions. I am speaking from fairly extensive experience with SCS, where the install framework was starting to get unmanageable. Adding extra install options to the main WEIDU installer is not an unalloyed good in a large complex mod.

 

I guess you could do as DavidW does - put even components such as "only physical resistance, no penalties" under regular install options and write something as ("not reccomended!") in brackets - this way everybody will see it, even if they don't feel comfortable editing the .ini file and do as they please (I do install SCS "All mages get HLA's in SoA" which isn't reccomended install for example, but it makes me feel good ).

 

On the current version of SCS, SOA and TOB HLAs are combined into a single component, so that only one of about five options is the "not really recommended" one. Prior to v22, there was a separate component for SoA HLAs. In retrospect that was a serious mistake: even if I said "not recommended" plenty of people just do install-all and then complain. And BWP makes it quite a lot worse. I have learned to have options that I don't recommend hidden away so that you don't get to see them unless you've read the instructions and know what you are doing (cf my reply to Jarno.)

[Jarno Mikkola]

The advantages are (i) it makes the install order less cluttered;

I) Yeah, and making the SCS v25 in to a one single component install also makes it less cluttered, after all you can make all the component choices in a .ini file... I dare you to implement that.

II) It would make your recommendation the primary and the almost singular vision of the mod... and how it should be installed... close to that of the Improved Anvil.

III) NO BODY READS teh README files ! If they don't have to... so you can state what the component does in the .tra file during it's install...

 

Example of this: my mods .tra files first few lines to a component:

@1 = ~Arcane Spell Refreshment~

 

@2 = ~Which type of function would you like the timer to be for the arcane spells ?

1) One timer for all levels

2) Timers depend linearly from the spell level

3) Timers depend exponentially from the spell level

4) Timers are randomized~

These two are displayed at the same time

@3 = ~So in:

1) timer = constant

2) timer = constant + spell level * level variable

3) timer = constant + spell level ^ (exponent/root)

4) timer = ~

 

@4 = ~Set *constant* for all levels~

 

@5 = ~We'll try to keep it on the level dude.

So set the darn *constant* to a number, dude. OK ?~ This is displayed if the previous function cave me an error as the player gave a wrong insert.

 

@6 = ~Set the *level variable* for the levels~

 

@7 -@11 various option

 

@12 = ~The constant is randomized for all levels~ To tell what the effect of the option 4 does.

 

@13 = ~The timers for the Arcane Spell restorations will be as of following:

Level 1 spells: %timer1% real seconds

Level 2 spells: %timer2% real seconds

Level 3 spells: %timer3% real seconds

Level 4 spells: %timer4% real seconds

Level 5 spells: %timer5% real seconds

Level 6 spells: %timer6% real seconds

Level 7 spells: %timer7% real seconds

Level 8 spells: %timer8% real seconds

Level 9 spells: %timer9% real seconds~ And this is displayed so the player can make a choice to return, or go through with the changes. With the choice in @14.

 

@14 = ~Are you sure you wish to use those ? [Y/N]~

And this is what it shown:

[kreso]

On the current version of SCS, SOA and TOB HLAs are combined into a single component, so that only one of about five options is the "not really recommended" one. Prior to v22, there was a separate component for SoA HLAs. In retrospect that was a serious mistake: even if I said "not recommended" plenty of people just do install-all and then complain. And BWP makes it quite a lot worse. I have learned to have options that I don't recommend hidden away so that you don't get to see them unless you've read the instructions and know what you are doing (cf my reply to Jarno.)

The 2 components I do miss from v21 are "Improved Irenicus in Hell" (Tactics version) and "Mages don't use PI and Simulacrum" for the sole reason I don't use them either.

 

III) NO BODY READS teh README files

I do, and I'm sure there are quite a few players which do as well. And if a mod doesn't have a fairly well Readme file, I tend to skip it - I want to know what (and how) exactly I'm doing to my install. Wisp's aTweaks, RR, Revisions, SCS, COI mods are examples of a good Readme file. A lot of mods, even newer ones, unfortunately, suffer from lack of it. If my .exe gets patched, I really want to know that in advance, since it may cause install failure on a consequent .exe patch. That's just an example, but there is other stuff such as installing items mod and getting several non-documented quests etc. For me, such stuff is a complete deal-breaker.

[DavidW]

The advantages are (i) it makes the install order less cluttered;

I) Yeah, and making the SCS v25 in to a one single component install also makes it less cluttered, after all you can make all the component choices in a .ini file... I dare you to implement that.

Don't be ridiculous. There is a trade-off between install customisability and install clutter.

 

II) It would make your recommendation the primary and the almost singular vision of the mod... and how it should be installed... close to that of the Improved Anvil.

 

SCS v25 (on a BGT+ascension install) has 125 optional components almost all of which can be installed independently. The total number of permutations is somewhere around

2^100 * 3^10 * 4^5 * 5^5 *6^6 = 1.1 * 10^46,

or eleven billion trillion trillion trillion combinations. If the current human population of the Earth were duplicated in every solar system in the visible Universe, and if all of them played speed-runs through BGT taking about a day per run, they'd exhaust possible SCS combinations around the time at which the last stars started to die.

 

I am not excessively worried about the extent to which I am restricting player choice.

 

III) NO BODY READS teh README files ! If they don't have to... so you can state what the component does in the .tra file during it's install...

I have no interest in supporting people who want to use advanced options but don't read the instructions. (Except insofar as for their own good they should be kept away from options they haven't understood.)

 

[Salk]

Personally I always read ReadMe:s and I am rather displeased when any software comes without it. So much that I tend to distrust its source.

[DavidW]

On the current version of SCS, SOA and TOB HLAs are combined into a single component, so that only one of about five options is the "not really recommended" one. Prior to v22, there was a separate component for SoA HLAs. In retrospect that was a serious mistake: even if I said "not recommended" plenty of people just do install-all and then complain. And BWP makes it quite a lot worse. I have learned to have options that I don't recommend hidden away so that you don't get to see them unless you've read the instructions and know what you are doing (cf my reply to Jarno.)

The 2 components I do miss from v21 are "Improved Irenicus in Hell" (Tactics version) and "Mages don't use PI and Simulacrum" for the sole reason I don't use them either.

The first one of those is an interesting case study, actually - I got a bit fed up with people complaining that it wasn't in the style of SCS as a whole. Players who were familiar with Tactics weren't surprised, of course, but others were caught off guard. Having said which, you're not the only person to miss it. I may reenable it in v25 - but via the ini!

 

The second one was technical: it adds quite a bit to the complexity of the already-complex mage scripting, and was making maintenance hard to cope with. If you miss it, I'll see if there's a backdoor way to reenable it (don't get your hopes up).

[kreso]

I may reenable it in v25 - but via the ini!

That would be great. I find the present Irenicus a bit too easy, and Tactics version somehow always keeps me on my toes.

 

The second one was technical: it adds quite a bit to the complexity of the already-complex mage scripting, and was making maintenance hard to cope with. If you miss it, I'll see if there's a backdoor way to reenable it (don't get your hopes up).

It's not neccecary - I guess it just takes some time to get used to. These spells are far less overwhelming when used by AI and not nearly as exploitable as when player uses them - I installed that option just for consistency sake, and pretended that they don't exist in my game, since I never use them myself.

Anyhow, T-up on "Irenicus in Hell" if you decide to re-implement it. It's a relic, but a good one.

Back to the "Revised Armors".....

[Ardanis]

II) It would make your recommendation the primary and the almost singular vision of the mod... and how it should be installed... close to that of the Improved Anvil.

Are you implying there is anything wrong about this? In case you haven't noticed, the prime issue with IA has been its author's attitude...

 

That's how all the games are being produced, primary and singular vision of its creators. We're not talking about a bunch of unrelated tweaks, but a complex project - not as complex as the game itself, but no longer in the same class as a tweakpack.

 

III) NO BODY READS teh README files ! If they don't have to... so you can state what the component does in the .tra file during it's install...

All the more reason to describe not recommended options and how to use them in readme, so that they won't be installed by mistake.

 

Example of this: my mods .tra files first few lines to a component:

 

And this is what it shown:

So you suggest the user should read a WoT of equal size for each component they're installing?