992 réponses à ce sujet

[In Exile]

The Gentle Ben wrote...
I'm not particularly advocating a return to the Origins List, why not a resizable box (you can even keep the intent icon in place), I don't really care, just something with a greater degree of flexibility might be less constraining both functionally and to the writers' approach.

I'm not sure what you mean by flexibility. With regard to the writers approach, I think the point is that it isn't the wheel that constraints them, but rather the 3 option wheel works well for their purpose because of other constraints.

'm fairly certain you're correct, which is why my objection was more conceptual/theoretical than immediate/practical in its outlook. My main concern is that the developers not become locked in to the interface (with its limitations) per say, and evaluate the feasability of increased responses on a project by project basis.

That becomes a financial decision, no? And I don't mean something so simple as "VO" is expensive, but rather expansive dialogue as a whole is expensive The more options and kinds of expression you give the player, the more overall zots you are investing per conversaton. At some point there will be diminishing returns.

Take your 4th aggressive option example - they could have added a 4th option... but that option would have meant unique reactions from your NPCs (plausibly) and that is a non-zero cost. Multiple that per conversation tree, and maybe even have some options offer new paths (if they would so logically entail them)... and suddenly your cost of production rises significantly.

I don't think we should ever lose sight of the fact that a game, in the end, has to sacrifice vision for feasibility.

Nilbog79 wrote...
I believe investigative options, no matter
how many there are, have to be neutral and do not move a conversation
forward. They are just an optional request for further information.

Absolutely; but most DA:O options were like this.

The Gentle Ben wrote...
I must admit I fail to see the problem
as you describe it. Write options as they come logically, assign them
on the scale where they fall. Does it matter fundamentally, say, if you
have two aggressive responses (with different areas of focus) and 1
diplomatic, as opposed to the need to fill the charming/humorous spoke.

I'm talking about how Bioware has handled their dialogue system. Right now, they have three options: aggressive, charming and diplomatic. You mentioned before that you prefered something that was more continous and had more than these three "core" personalities. That means that, on the implementation side, there has to be a more complicated sort of tracking taking place.

You also mentioned hybrid personalities - that would also involve more complicated tracking. How would these personalities hybridize? Would there be unique content for a hybrid personality? What would a hybrid personality be?

In general, I assumed you were talking at the implementation level in general, i.e. we are talking about ways to design a game versus minor changes we could make to the existing DA2 system.

The problem with "just one more" is that you could very well never stop. Which is to say, you could have a sarcastic demeaning reply "Well, at least we have one less mouth to feed?" or an emotional outburst "By the marker, no! Bethany/Carver (crying)!"

I'll
even use this opportunity to combine my two points. We'll use the scene
following Carver's death as an example. Even though it's been examined
exhaustively by this point, it seems a worthy illustration. Your mother
is crying over his body, our existing choices are:
1. He risked
himself to save us. (Diplomatic)
2. He's with father now (Charming)
3.
We're wasting time (Aggressive)

Why not add option 4. Vengeance:
"These souless bastards will pay" (Aggressive) It was the option, I
myself and (apparently) others were looking for. If the reason for its
ommission were either 1.) we've filled the three spokes, or 2.) we
already have an aggressive option, then I view that as a mark against
the dialogue interface as implemented.

Why should this be a 4th option? Why not "He was an idiot, he should have grabbed mother and run!" or the aforementioned crying option? There may well have been people who wanted to say those lines instead of the 3 options they had.

As I said: I think it comes down to cost and the problem of where enough is enough. There might also be complaints from players if Hawke is ignored. 

Take your exmaple: "These souless bastards will pay." Well, now you're not a mind to run away from the darkspawn, are you? So there may need to be unique content telling you that's suicidal. That costs zots. That's the problem.

Modifié par In Exile, 27 février 2011 - 10:24 .

[AlanC9]

Morroian wrote...

Nilbog79 wrote...

It seems The Witcher (another rpg) has a system where you have a voiced protagonist but the dialogue choices list exactly what the protagonist will say. So if one of the options is "We have to go", you click on it and the main hero will say "We have to go". It seems like such a system would at least solve the paraphrasing problem, while still keeping the voiced main hero.

This is seen as redundant and the majority of people would fast forward through the actual spoken scene. However I do think to appease those who want it it should have been added, I doubt it would have taken much effort.

It might be difficult for the consoles -- you can't just go to a WoT there because it won't be legible. Depends on how long-winded the PC is to some extent -- Geralt's pertty laconic, so his lines work fine. A sizable fraction of Shepard's lines would cause problems. I don't have a big enough sample size for JHawke to judge.

Of course, the writers could just write dialog around that constraint if they really thought it was important

[Sylvius the Mad]

dbankier wrote...

Those aren't "good/evil" pictures, they're personality options. When you're making a decision those icons aren't even used.

That in itself is a problem.  This prevents the character from making decisions at what the writers think are personality hubs rather than action hubs.

[The Gentle Ben]

Demo-Mike wrote...
No, this is not good system...

Give options

Heh, I've been trying, but I'll give it one more go before I have to depart the virtual battlefield (out to dinner).

1.) Create an opt-in mechanic for a full-text display on mouse-over of the paraphrase, or alternately, merely display the subsequent subtitle on screen.

2.) Allow for an expandable version of the dialogue wheel in order to support more dialogue progression options as needed, or alternatively actually use all five available slots without worrying about covering the three-bases exclusively. Further, don't write to tone, but assign tones to writing, and finally, don't feel the need to implement all tones for every situation.

3.) Use a scaling mechanic as opposed to a counter for dominant tone (and ideally have 2 hybrid tones in-between the existing 3 (Diplomatic, Charming, Aggressive)).

Also, no worries AlanC9. I merely dislike the whole, you've lost move-on mentality that some seem to have adopted to critics of the paraphrase system. Their concerns and arguments have a right to be expressed.

[Sylvius the Mad]

AlanC9 wrote...

Yep, except that Bio's focus group testing showed that people preferred paraphrases to reading the exact line they're going to hear the character say next.

Actually, no.  BioWare's focus group testing showed that people, when given the full text in advance, are more likely to skip the voiced lines.

Now, BioWare doesn't want to produce content that the players don't see, but their solution here is to force the players to see it rather than only producing content they want to see.

I think BioWare misinterpreted the focus group data.

[In Exile]

Sylvius the Mad wrote...
Actually, no.  BioWare's focus group testing showed that people, when given the full text in advance, are more likely to skip the voiced lines.

Now, BioWare doesn't want to produce content that the players don't see, but their solution here is to force the players to see it rather than only producing content they want to see.

I think BioWare misinterpreted the focus group data.

That was the result, but a focus group is more than just empirical data. A significant part of it is to ask people why they choose a particular feature. Bioware could easily determine what the situation is via two simple questions:

Why did you skip the dialogue?
Would you prefer a description different from the dialogue to the full dialogue as an indicator for your choice? (or something to that effect - I can't word this question right off the top of my head).

Still, there's no prima facie reason to assume Bioware is incompetent.

[Sylvius the Mad]

AlanC9 wrote...

And they've made that calculation.

I notice that the people who are violently opposed to VO don't try to say they're in the majority anymore. Usually they'll say something like the "core fanbase."

"The really important fans dislike this feature, and Bio should have listened to us rather than all these other idiots."

I'm not sure what the "core fanbase" would be.

I still think the VO reduces or eliminates roleplaying as I understand the term (as RP requires total control over character expression), and I'm happy to call roleplaying a core mechanic, or even a critical component, of RPGs.

[In Exile]

The Gentle Ben wrote...
1.) Create an opt-in mechanic for a full-text display on mouse-over of the paraphrase, or alternately, merely display the subsequent subtitle on screen.

The problem is how to properly implement this across platforms. Deus Ex will try a system, but it's actually not the full line as people are saying. Rather, it is only the first part of the sentence that the character will speak. There will actually be more dialogue than is just written down that is being said.

A mouse tooltip won't work on console. I think the feasible cross platform solution is a subtitle toggle... but I don't know how easy that would be to implement.

2.) Allow for an expandable version of the dialogue wheel in order to support more dialogue progression options as needed, or alternatively actually use all five available slots without worrying about covering the three-bases exclusively. Further, don't write to tone, but assign tones to writing, and finally, don't feel the need to implement all tones for every situation.

This is where you start to have implementation problems. If the tones are not equivalent in # then the mathematics for a dominat tone (if you want to keep that feature) becomes complicated. What if you have significantly more aggresive tones per conversation than other tones? That could easily bias players toward one personality type.

Again - the obstacle to more dialogue is not the paraphrase, but rather the complication of new content in a conversation. If Hawke can express very different opinions, NPCs have to react (otherwise the whole situations contrived). This means exclusive content. Lots of exclusive content gets expensive, and that's only looking at exclusive content within a conversation.

3.) Use a scaling mechanic as opposed to a counter for dominant tone (and ideally have 2 hybrid tones in-between the existing 3 (Diplomatic, Charming, Aggressive)).

What would a "hybrid" tone be? Angry sarcasm? Would this mean unique content? Unique reactions?

I think your solution is very zot intensive and that's a problem from the implementation side.

[Marbazoid]

The one thing that bothers me about the the three "tones" is that in order to rp a consistent character, your mostly going to choose the same spot on the wheel every-time, rather than choose a response on a case by case basis.

In Mass Effect it boiled down to exhausting investigate options, then selecting upper right/lower right until conversation concludes.

[AlanC9]

Sure, Sylvius, and you also don't bother with that kind of rhetoric in the first place. You care about being right, not about being in the majority.

[The Gentle Ben]

In Exile wrote...
I'm not sure what you mean by flexibility. With regard to the writers approach.

I'm refering to their need to support all three tones and only those tones, instead of say, supporting 2 aggressive and 1 diplomatic.

In Exile wrote...
That becomes a financial decision, no? And I don't mean something so simple as "VO" is expensive, but rather expansive dialogue as a whole is expensive The more options and kinds of expression you give the player, the more overall zots you are investing per conversaton. At some point there will be diminishing returns.

Take your 4th aggressive option example - they could have added a 4th option... but that option would have meant unique reactions from your NPCs (plausibly) and that is a non-zero cost. Multiple that per conversation tree, and maybe even have some options offer new paths (if they would so logically entail them)... and suddenly your cost of production rises significantly.

I don't think we should ever lose sight of the fact that a game, in the end, has to sacrifice vision for feasibility.

I agree, however, the investigate options supported are also a part of this calculation, so the idea that the writers are truly beholden to only three options per dialogue as a financial restraint isn't really supported by the evidence. I fully acknowledge the finances in play, I'm merely talking about implementing the flexibility to offer additional options as deemed useful by the writers.

I'm talking about how Bioware has handled their dialogue system. Right now, they have three options: aggressive, charming and diplomatic. You mentioned before that you prefered something that was more continous and had more than these three "core" personalities. That means that, on the implementation side, there has to be a more complicated sort of tracking taking place.

Not really, just a sliding scale with personality ranges.

You also mentioned hybrid personalities - that would also involve more complicated tracking. How would these personalities hybridize? Would there be unique content for a hybrid personality? What would a hybrid personality be?

In general, I assumed you were talking at the implementation level in general, i.e. we are talking about ways to design a game versus minor changes we could make to the existing DA2 system.

Which is why I hesitated to mention it (and said ideally), because I recognize it as an added cost, but a "buffer" personality between the three extremes seems useful in order to ease transitions between them, even in the current implementation.

Why should this be a 4th option? Why not "He was an idiot, he should have grabbed mother and run!" or the aforementioned crying option? There may well have been people who wanted to say those lines instead of the 3 options they had.

The main point is to try and cover as wide a range of potential intents as possible. The crying option, for example would at least somewhat fall under the range of the existing diplomatic response, while the "he was an idiot" is less supported, it still nominally falls under the agressive (not caring) option. There is no comparable range for vengeance, but thix is getting bogged down in the details to an extent.

Plus I really have to go.

[Sylvius the Mad]

In Exile wrote...

That was the result, but a focus group is more than just empirical data. A significant part of it is to ask people why they choose a particular feature. Bioware could easily determine what the situation is via two simple questions:

Why did you skip the dialogue?
Would you prefer a description different from the dialogue to the full dialogue as an indicator for your choice? (or something to that effect - I can't word this question right off the top of my head).

Still, there's no prima facie reason to assume Bioware is incompetent.

That approach only works if you think that the subjects are sufficiently introspective to understand the motives behind their own actions.  And I think our prior discussions have established that they are not.

People don't generally know why they what they do.  They don't plan ahead.  They don't give decisions any thought (particularly decisions they view as unimportant, like how they behave when playing a game).  They don't typically remember their decision-making process (or even have one that they follow).  When asked to explain why they've done something, most people will rationalise an answer that sounds plausible to them.  That doesn't make it true, though they'll insist that it is.

I don't think BioWare is incompetent.  I think mainstream market research is badly flawed, and BioWare is a victim as much as we are.

[The Gentle Ben]

NVM

Modifié par The Gentle Ben, 27 février 2011 - 10:46 .

[Sylvius the Mad]

Marbazoid wrote...

The one thing that bothers me about the the three "tones" is that in order to rp a consistent character, your mostly going to choose the same spot on the wheel every-time, rather than choose a response on a case by case basis.

That's only true if your measure of consistency is BioWare's categories.  I see no reason why one couldn't play a consistent character who was consistent by some other measure, and thus jumped back and forth on BioWare's scale.

The only time this could cause a problem is what the "dominant tone" system is being used for action choices, but I've voiced that specific concern before (I think BioWare should have offered us all of the possible options there each time, rather than limiting us to those it thinks are consistent with our prior behaviour).

[Nilbog79]

In Exile wrote...

The Gentle Ben wrote...

2.) Allow for an expandable version of the dialogue wheel in order to support more dialogue progression options as needed, or alternatively actually use all five available slots without worrying about covering the three-bases exclusively. Further, don't write to tone, but assign tones to writing, and finally, don't feel the need to implement all tones for every situation.

This is where you start to have implementation problems. If the tones are not equivalent in # then the mathematics for a dominat tone (if you want to keep that feature) becomes complicated. What if you have significantly more aggresive tones per conversation than other tones? That could easily bias players toward one personality type.

There is a number of ways to solve the problem of uneven distribution of tones. Responses could be weighted according to their frequency. Most likely though the number of responses in the entire dialogue would work out evenly anyway, even if some dialogues are biased towards one tone.

In Exile wrote...
Again - the obstacle to more dialogue is not the paraphrase, but rather
the complication of new content in a conversation. If Hawke can express
very different opinions, NPCs have to react (otherwise the whole
situations contrived). This means exclusive content. Lots of exclusive
content gets expensive, and that's only looking at exclusive content
within a conversation.

It was done in Origins. I don't know how different the budget for DA2 was from budget for DAO, of course.

Modifié par Nilbog79, 27 février 2011 - 10:53 .

[Marbazoid]

Sylvius the Mad wrote...

Marbazoid wrote...

The one thing that bothers me about the the three "tones" is that in order to rp a consistent character, your mostly going to choose the same spot on the wheel every-time, rather than choose a response on a case by case basis.

That's only true if your measure of consistency is BioWare's categories.  I see no reason why one couldn't play a consistent character who was consistent by some other measure, and thus jumped back and forth on BioWare's scale.

The only time this could cause a problem is what the "dominant tone" system is being used for action choices, but I've voiced that specific concern before (I think BioWare should have offered us all of the possible options there each time, rather than limiting us to those it thinks are consistent with our prior behaviour).

I see your point. I guess the paragon/renegade persuade options requiring so many "points" encourages my mindset. I have similar concerns over the 'dominant tone' system.

[tmp7704]

AlanC9 wrote...

I notice that the people who are violently opposed to VO don't try to say they're in the majority anymore.

Point out enough times there's no data to prove it and it's eventually going to stick. Doesn't mean anyone who tries to claim the opposite isn't goint to get [citation needed] just the same, though. It applies both ways.

[In Exile]

Sylvius the Mad wrote...
That approach only works if you think that the subjects are sufficiently introspective to understand the motives behind their own actions.  And I think our prior discussions have established that they are not.

Well, yes and no. I mean, I'd agree with you that people can't generally sufficiently introspect. But that's not the same as saying people are inaccurate in their personal narratives to the extent that there is nothing of value there.

The trick is in knowing how to ask people what they want in such a way that you're largely getting at a real preference. In the case of the paraphrase, it doesn't really matter why people prefer it... so long as you know they prefer it and can reliably test for what they like.

So long as you know a relative preference (paraphrase > full text) you're golden.

People don't generally know why they what they do.  They don't plan ahead.  They don't give decisions any thought (particularly decisions they view as unimportant, like how they behave when playing a game).  They don't typically remember their decision-making process (or even have one that they follow).  When asked to explain why they've done something, most people will rationalise an answer that sounds plausible to them.  That doesn't make it true, though they'll insist that it is.

This is all true. But self-report can still be reliable so long as you investigate carefully. So, I could ask you forced choice questions about what you would enjoy more, and then test to see if there are statistically significant differences across groups. That would give me confidence in my measures.

I don't think BioWare is incompetent.  I think mainstream market research is badly flawed, and BioWare is a victim as much as we are.

I've seen only academic marketing research, so my experience is limited, but it generally seems to apply relatively rigorous analysis.

[In Exile]

Nilbog79 wrote...
There is a number of ways to solve the problem of uneven distribution of tones. Responses could be weighted according to their frequency. Most likely though the number of responses in the entire dialogue would work out evenly anyway, even if some dialogues are biased towards one tone.

That becomes punishing. So I have to be aggresive 8 times to re-write my one charming choice? It's not so easy to come up with the ideal implementation.

It was done in Origins. I don't know how different the budget for DA2 was from budget for DAO, of course.

What? No it didn't. Origins often had 2 options be worded differently and give you the exact same outcome. Whenever you actually had a choice it was essentially nice or naughty and sometimes you had a third option, too. But quests with multiple solutions was not something Bioware did in DA:O.

[Sylvius the Mad]

In Exile wrote...

The trick is in knowing how to ask people what they want in such a way that you're largely getting at a real preference. In the case of the paraphrase, it doesn't really matter why people prefer it... so long as you know they prefer it and can reliably test for what they like.

I don't think people can know that they have a relative preference without knowing why.

[Veex]

Marbazoid wrote...

I see your point. I guess the paragon/renegade persuade options requiring so many "points" encourages my mindset. I have similar concerns over the 'dominant tone' system.

If you're really concerned about the tone icons, and you happen to be playing on a PC, you can use the number keys to select your conversation answer without seeing the icons.

4 (    )  1
5 (    )  2
6 (    )  3

It's a rough sketch, but you get the idea.

[In Exile]

Sylvius the Mad wrote...
I don't think people can know that they have a relative preference without knowing why.

Sure you can. In fact, priming studies are entirely about this sort of thing.

[Sylvius the Mad]

Marbazoid wrote...

I see your point. I guess the paragon/renegade persuade options requiring so many "points" encourages my mindset. I have similar concerns over the 'dominant tone' system.

ME was extremely poorly done in that regard.  I would encourage everyone to mod the game to give Shepard maximum Paragon and Renegade points right from the beginning (AngryPants gave me this idea).

[Sylvius the Mad]

In Exile wrote...

Sylvius the Mad wrote...
I don't think people can know that they have a relative preference without knowing why.

Sure you can. In fact, priming studies are entirely about this sort of thing.

Unless I'm thinking of the wrong thing, Priming Studies actually measure things that are beyond the scope of knowledge (implicit memories, or something like that).  I don't see the relevance.

[tmp7704]

In Exile wrote...

Sylvius the Mad wrote...

I don't think BioWare is incompetent.  I think mainstream market research is badly flawed, and BioWare is a victim as much as we are.

I've seen only academic marketing research, so my experience is limited, but it generally seems to apply relatively rigorous analysis.

The problem starts right from trying to decide the composition of the test groups, i'd think. Do you test your existing playerbase (and good luck figuring out good representative sample for that in the first place)? Or is that for some market group you're aiming for that may or may not be quite different from it? Or is it some mix of both? If so, what sort of a mix is that? Etc and so on, there's enough variety and issues in this alone to potentially render any results beyond some most universal trends highly meaningless.

Modifié par tmp7704, 27 février 2011 - 11:15 .