Coming Soon: Discussion Thread (2021)

[Ewe]

Re-rolls for Slippery Mind, Follow the Guide, and Indomitable Soul should properly reroll even if the first roll is a 1 per d20 rules. Note that these feats don't stack with each other.

[Most Horrible]

This is just a request for clarification:

As per feat description: Steadfast Determination should only prevent fortitude save failure upon the roll of one, but it applied to also Reflex and Will saves, so I presume that this has been now fixed.

Moreover, because of the above behaviour, hardly anyone ever took the Epic Resilience feat, which is supposed to prevent automatic save failure upon rolling a one, will this feat still work as described in its description? Or will this feat be broken after this change?

[DaloLorn]

This is just a request for clarification:

As per feat description: Steadfast Determination should only prevent fortitude save failure upon the roll of one, but it applied to also Reflex and Will saves, so I presume that this has been now fixed.

Moreover, because of the above behaviour, hardly anyone ever took the Epic Resilience feat, which is supposed to prevent automatic save failure upon rolling a one, will this feat still work as described in its description? Or will this feat be broken after this change?

Actually, Steadfast Determination worked as intended, preventing autofailed Fortitude (and only Fortitude) saves.

The problem was that even without any feats, you would never be able to autofail Reflex and Will saves. You will now be able to do so, unless you have Epic Resilience.

Personally, I'm frustrated that it was decided not to offer RCRs to PCs with SD, but... whatever. It's not like I play the affected characters a lot anymore, anyway, and RCRing 2-3 characters in the 15-25 range would have been a pain in the neck.

[EasternCheesE]

Hidden: show

This is just a request for clarification:

As per feat description: Steadfast Determination should only prevent fortitude save failure upon the roll of one, but it applied to also Reflex and Will saves, so I presume that this has been now fixed.

Moreover, because of the above behaviour, hardly anyone ever took the Epic Resilience feat, which is supposed to prevent automatic save failure upon rolling a one, will this feat still work as described in its description? Or will this feat be broken after this change?

Actually, Steadfast Determination worked as intended, preventing autofailed Fortitude (and only Fortitude) saves.

The problem was that even without any feats, you would never be able to autofail Reflex and Will saves. You will now be able to do so, unless you have Epic Resilience.

Personally, I'm frustrated that it was decided not to offer RCRs to PCs with SD, but... whatever. It's not like I play the affected characters a lot anymore, anyway, and RCRing 2-3 characters in the 15-25 range would have been a pain in the neck.

Sorry, but what's the problem about SD and why should they RcR because of this change?

[Hoihe]

Hidden: show

This is just a request for clarification:

As per feat description: Steadfast Determination should only prevent fortitude save failure upon the roll of one, but it applied to also Reflex and Will saves, so I presume that this has been now fixed.

Moreover, because of the above behaviour, hardly anyone ever took the Epic Resilience feat, which is supposed to prevent automatic save failure upon rolling a one, will this feat still work as described in its description? Or will this feat be broken after this change?

Actually, Steadfast Determination worked as intended, preventing autofailed Fortitude (and only Fortitude) saves.

The problem was that even without any feats, you would never be able to autofail Reflex and Will saves. You will now be able to do so, unless you have Epic Resilience.

Personally, I'm frustrated that it was decided not to offer RCRs to PCs with SD, but... whatever. It's not like I play the affected characters a lot anymore, anyway, and RCRing 2-3 characters in the 15-25 range would have been a pain in the neck.

Sorry, but what's the problem about SD and why should they RcR because of this change?

People were told taking Steadfast Determination gives immunity on 1s for fort and will.

They did not realize it was a bug.

That#s my read.

[DaloLorn]

Hidden: show

This is just a request for clarification:

As per feat description: Steadfast Determination should only prevent fortitude save failure upon the roll of one, but it applied to also Reflex and Will saves, so I presume that this has been now fixed.

Moreover, because of the above behaviour, hardly anyone ever took the Epic Resilience feat, which is supposed to prevent automatic save failure upon rolling a one, will this feat still work as described in its description? Or will this feat be broken after this change?

Actually, Steadfast Determination worked as intended, preventing autofailed Fortitude (and only Fortitude) saves.

The problem was that even without any feats, you would never be able to autofail Reflex and Will saves. You will now be able to do so, unless you have Epic Resilience.

Personally, I'm frustrated that it was decided not to offer RCRs to PCs with SD, but... whatever. It's not like I play the affected characters a lot anymore, anyway, and RCRing 2-3 characters in the 15-25 range would have been a pain in the neck.

Sorry, but what's the problem about SD and why should they RcR because of this change?

Steadfast Determination, not Shadowdancer. Some players might want to switch to Epic Resilience now (if they can spare the feats), and others might want to just drop Steadfast Determination in favor of something else.

[EasternCheesE]

Hidden: show

Hidden: show

Actually, Steadfast Determination worked as intended, preventing autofailed Fortitude (and only Fortitude) saves.

The problem was that even without any feats, you would never be able to autofail Reflex and Will saves. You will now be able to do so, unless you have Epic Resilience.

Personally, I'm frustrated that it was decided not to offer RCRs to PCs with SD, but... whatever. It's not like I play the affected characters a lot anymore, anyway, and RCRing 2-3 characters in the 15-25 range would have been a pain in the neck.

Sorry, but what's the problem about SD and why should they RcR because of this change?

People were told taking Steadfast Determination gives immunity on 1s for fort and will.

They did not realize it was a bug.

That#s my read.

Oh, my bad. I read SD as Shadowdancer so i wondered how was it even related except their slippery mind finally has some use

[Ewe]

Steadfast Determination will only protect against fort auto failures. Even with this feat combat log showed failure for will and reflex but due to an engine bug they were treated as successful with or without this feat.

Epic Resilience always properly protected and showed combat log correctly for all auto failures.

Take away is it was impossible to auto fail on will and reflex no matter what feats you had due to an engine bug.

[Most Horrible]

Well, if Epic Resilience and Steadfast Determination work as per their descriptions, then you get my thumbs up.

[Most Horrible]

• Cipher normalization. Lowercase is normal text, uppercase is uppercased text. None of this "lowercase is uppercase but uppercase is lowercase" or "uppercase isn't same as lowercase" nonsense that my last update exposed.

So in other words: the language translations will be boring lines of 'Ceasar ciphered' English instead of trying to make it seem that people are actually using a different language that sounds different. Every language will look and feel the same, and thus there isn't such an easy way to make the related languages 'sound/look' very similar while being pronounced differently as hinted by the use of lower and upper case letters. Thus when actually writing the translation cipher bits, you will have to manually adjust each cipher by replacing vowels, instead of just making lower and upper case letters appear. And generally speaking, the languages will be less unique... And it will be harder for players to detect that a particular language is spoken, because everything will just look like English ciphered into random gibberish.

Oh and speaking of which, did you know that the word barbarian comes from the fact that some ancient folk though the speech of barbarians sounded like 'bar-bar-bar' to them?

So this complain here is just about the nature of human linguistics, about making it so that people can actually identify a language used without actuall understanding it. You know, for example, someone speaking Spanish or French while walking on the streets, or German while sitting in the subway. You might not understand a word spoken, but you can still identify the language spoken.

[Rhifox]

Each language has a cipher of many words often specific to that language, rather than complete random gibberish. It'll still be plenty possible to identify what language someone is speaking (or in some cases, even translate it manually).

[DaloLorn]

• Cipher normalization. Lowercase is normal text, uppercase is uppercased text. None of this "lowercase is uppercase but uppercase is lowercase" or "uppercase isn't same as lowercase" nonsense that my last update exposed.

So in other words: the language translations will be boring lines of 'Ceasar ciphered' English instead of trying to make it seem that people are actually using a different language that sounds different. Every language will look and feel the same, and thus there isn't such an easy way to make the related languages 'sound/look' very similar while being pronounced differently as hinted by the use of lower and upper case letters. Thus when actually writing the translation cipher bits, you will have to manually adjust each cipher by replacing vowels, instead of just making lower and upper case letters appear. And generally speaking, the languages will be less unique... And it will be harder for players to detect that a particular language is spoken, because everything will just look like English ciphered into random gibberish.

I'm not sure I know what you're getting at here. And if I do, you're drastically misreading the situation.

1. A lot of ciphers were internally inconsistent. Lowercase and uppercase letters often didn't map to the same thing - for instance, the current Gnoll cipher maps "a" and "A" to "Gr" and "gh", respectively. (That's tame compared to some of the others! Jotun does "a->Ugh" and "A->plop"! Why the hell is A "plop"?!)
   
   This was previously unknown, because the old translation script lowercased the whole string before translating it, and consequently ignored the uppercase cipher altogether. Incidentally, the old translation script also lowercased the translated string, so uppercase letters could never even appear in the output until I started poking around the script.
2. As the above example demonstrates, the ciphers don't necessarily map a character to another character. Heck, the old sign language ciphers mapped a character to a whole emote (or would have, if the author hadn't screwed up the 2DAs for it). Furthermore, there is no requirement that the cipher must be reversible, or that the result must still use the whole English alphabet.
3. Additionally, as Rhifox alluded to, many of the most commonly used languages have dictionaries of up to 5000 words (which is a huge contributor to the slowdown, because some of the dictionaries are broken too - Celestial being a particularly severe example - and I have to inspect all that too). Sometimes these dictionaries have some foundation in canon (Elven being a notable example). Sometimes they're pulled from RL languages. (Celestial is an odd mixture of English, Latin, and ciphered gibberish; I've also noticed French in Aglarondan, Finnish in Jotun, Italian in something, German in Illuskan, and I think some Middle Eastern language or another in Alzhedo.) Sometimes I don't have the faintest clue how they were populated.
   
   Those dictionaries aren't going away. If anything, by fixing them, I'm making the game aware of words that it currently doesn't know it should translate!
4. The authors of these languages are no longer on the team, if they ever were. I don't have documentation telling me what the intent was or how to fix the bugs without breaking the design. I only have a variety of anomalies to indicate that they were, in fact, bugs.
5. Speaking of anomalies: While the Jotun and Sylvan ciphers have a mixture of uppercase and lowercase letters, their respective dictionaries do nothing of the sort. Jotun doesn't translate "cause" to "aiHeUTtaa", it translates it to "aiheuttaa". This tells me that either the design is irreconcilably inconsistent with itself, or the implementation is terribly broken.

[Ewe]

This should be live now.

[Louvaine]

I absolutely love this system. I want to roll up so many clerics!

The magic domain receives Spell Mantle at 7th circle, just as any other cleric would. Wiki page for Spell Mantle says it'd be circle 6th for magic domain, which would make more sense. Greater Scrying, however, doesn't list magic domain as its source at all. Seems like maybe Spell Mantle was supposed to be assigned for 6th cirle and Greater Scrying for 7th.

[EasternCheesE]

• Cipher normalization. Lowercase is normal text, uppercase is uppercased text. None of this "lowercase is uppercase but uppercase is lowercase" or "uppercase isn't same as lowercase" nonsense that my last update exposed.

So in other words: the language translations will be boring lines of 'Ceasar ciphered' English instead of trying to make it seem that people are actually using a different language that sounds different. Every language will look and feel the same, and thus there isn't such an easy way to make the related languages 'sound/look' very similar while being pronounced differently as hinted by the use of lower and upper case letters. Thus when actually writing the translation cipher bits, you will have to manually adjust each cipher by replacing vowels, instead of just making lower and upper case letters appear. And generally speaking, the languages will be less unique... And it will be harder for players to detect that a particular language is spoken, because everything will just look like English ciphered into random gibberish.

Oh and speaking of which, did you know that the word barbarian comes from the fact that some ancient folk though the speech of barbarians sounded like 'bar-bar-bar' to them?

So this complain here is just about the nature of human linguistics, about making it so that people can actually identify a language used without actuall understanding it. You know, for example, someone speaking Spanish or French while walking on the streets, or German while sitting in the subway. You might not understand a word spoken, but you can still identify the language spoken.

Sorry, but inventing 30+ languages with just like 100-200 words provided as base (or even less for some exotic ones) would be a ton of unnecessary work that will require full staff being professional linguists.
Can you propose better way to automate language creation without too much efforts put into each one?
Just in advance, using RL languages won't fit as people can copy-paste others speaking from CE console and google translate is good at identifying the language.