Cast spell (146) at least with Type=1 is either inaccurate or incorrect + Is IESDP starting to lag behind latest community knowledge?

[Guest Spellcaster]

Hello,

Cutting the post into digestible bits:

Cast spell (146):

In the section on Cast spell (146), IESDP only lists two values for Type where Near Infinity has three (which it chose to call Mode) : Cast normally (0), Cast instantly (ignore level) (1) and Cast instantly (at level) (2). Moreover, behaviour noticed in game[1] with Limited Wish's party-wide Minor Globe of Invulnerability, where Type=1 and Caster level=10 show that the IESDP description for Type=1 is either incorrect or incomplete: The Globe lasts for a duration equal to the target's mage level, meaning 42s for non-mages and up to 120s for level 20+ mages, rather than 60s for everybody. [2]

I don't know if the cause is that this wish uses Cast Spell (146) with Target=Party (3) and that such a target would happen to be an edge case not covered by IESDP, or if IESDP really did get the behaviour of Type=1 incorrect in the general case, a situation later only fixed in Near Infinity -- meaning the modding community is aware but IESDP is lagging behind.

IESDP issue:

If the former case is true, the purpose of this post is to have a note hanging around saying that there is something to look into to make IESDP's description on Cast spell and perhaps other effects more correct.

If it is the latter case that's true, the purpose of this post is:

• To ask why IESDP is lagging behind latest community knowledge on this particular issue where Near Infinity is not: Couldn't both communicate and remain in sync?
• To incite people curious enough to evaluate how many similar issues there could be, to learn if IESDP has an actual problem or not
• To invite considerations on which organisational tweaks would be best in either case

As a bit of an outsider (I am focused on the Wiki for now), I can't be of much use pondering on this, but my blind trust in IESDP is being shaken a bit, having me resort to time wasting in-game tests in both editions of the game to double check things written there. All the while with a feeling that someone else somewhere already knows the answers I am seeking, but it is buried in their head or in a post lost to time in a forum, rather than IESDP or the BG Wiki.

Some user feedback:

In case the current organisation is not good enough and IESDP does have an actual problem -- something I don't have a wide enough visual range to assess, I'll leave around the main reason I won't easily contribute to IESDP. It's not because it's over my head or because I'm not invested enough in Baldur's Gate, it's only a contribution-process issue: I would be drawn in a lot more easily with a lower barrier to contribution stemming from the possibility to make many small incremental changes over time with after-the-fact community-wide vetting, rather than being required to submit final, perfect pull requests to before-the-fact vetting by a few admins based on their idea on community consensus. I don't want to bother people with a thousand incremental pulls. I don't want them to slow my work down either. I don't think it is efficient that all knowledge has to pass through the same couple of heads in order to become true. I don't think a gated approach is good for knowledge repositories in general, in short. It's good for software development. I know moving IESDP over to a Wiki platform has been considered and rejected. Maybe a Github configuration exists that fits my low barrier to contribution criteria and maybe IESDP even uses it already and I've been clueless about it, and therefore this feedback doesn't apply?

 

PS: The IESDP is great work so please don't take this as a negative stance towards it or its major contributors.

Cheers!

 

 

 

[1]: With both BG2-ToB 2.5.26498 and BG2EE 2.5.16.6

[2]: Possibly Cast spell at point (148) and other similar effects suffer from having two Types/Modes instead of three and an inaccurate description of #1. I didn't check.

[kjeron]

AFAIK mode 1 has always worked that way, unable to specify a casting level, for both op146 and op148.

Op148 has only 2 casting modes - that third one is exclusive to op146.

The IESDP hasn't been abandoned.

 

[lynx]

IESDP, or any research for that matter, should never be trusted blindly. Similarly, I still find forum archives very useful, even if a lot of old research is vague, wrong and/or misleading. Nowadays people almost exclusively only test EEs and more shalowly, so there's a gap.

But all the research, the tool updates and IESDP are a result of volunteer work. So any discrepancies are just that, discrepancies from different priorities (and the far past when there was less oversight on quality). There's nobody actively hunting for info and making sure all users have it (the closest thing is @Luke ). This subforum used to be the focal point for research, but the few years in between, when IESDP wasn't updated, hurt the practice.

As far as contribution review goes: it's fully open. Sure, there's  only a few people that can merge on github, but that's a technicality. Everyone is free to help with review, just like when stuff is shared on the forum. It's unfortunate that more people don't participate in this, but it's sort of boring, so it's understandable. It also makes it harder to grow the pool of people with commit access.

I know you don't like git for some reason, but you're projecting a lot. Pull requests can even be marked as draft if you want to explicitly signal that they're not ready yet or that you just want to have a structured discussion before preparing a final proposal. And small incremental changes are just as welcome as big ones — as long as they're incremental improvements, not all around the place. And finally you can make as many changes in a single PR as you want, have it open for months if needed.

[Magus]

1.  I've submitted smallest pulls correcting typos in IESDP and they were accepted. That is not a problem.
2. I also submitted large ones, with a lot of discussion and back-and-forth.
3. You can easily maintain your own fork and work in it (which is what I do, incidentally).
4. The real reason IESDP is missing stuff is simply because no one cares enough.
5. The way open source works, once you've bugged admins with good pulls one too many times, they'll give commit access to you =).

 

On 11/24/2020 at 3:57 AM, Guest Spellcaster said:

To ask why IESDP is lagging behind latest community knowledge on this particular issue where Near Infinity is not: Couldn't both communicate and remain in sync?

This one I find very important. I would love to have the efforts centralized, and the logical place for that is IESDP. Also, various differences in naming between IESDP and NI always throw me off when I get back to modding.

I've done some work on converting the data into machine readable format, which enabled all the nice tooltips in MLS. I think that overall it would be beneficial if NI also used that, but I think @argent77 seems to be sceptical of the idea.

 

[Sam.]

11 hours ago, qwerty1234567 said:

This one I find very important. I would love to have the efforts centralized, and the logical place for that is IESDP. Also, various differences in naming between IESDP and NI always throw me off when I get back to modding.

I've done some work on converting the data into machine readable format, which enabled all the nice tooltips in MLS. I think that overall it would be beneficial if NI also used that, but I think @argent77 seems to be sceptical of the idea.

Not that my opinion counts for anything, but in general I prefer NI's jargon over the language used in the IESDP, especially the names of Effects.  I'm sure both are technically correct, but NI uses language I find more natural.  I acknowledge this is purely a personal preference.

[Magus]

11 hours ago, Sam. said:

Not that my opinion counts for anything, but in general I prefer NI's jargon over the language used in the IESDP, especially the names of Effects.  I'm sure both are technically correct, but NI uses language I find more natural.  I acknowledge this is purely a personal preference.

Me personally, I wouldn't care much which specific word is used in this or that case (would've probably some kind of combination of both), as long as terminolgy is consistent.