Gameplay elements, evolution, and enjoyment not for everyone? [2019+] *spawn*

[Xbat]

Herein seems to lie the problem. From the sounds of it, some people feel having the player do something, even if minimal, is gameplay. To others, if there's no decision or influence, it isn't gameplay. I side with the latter

The winch in Uncharted 4 is definitely gameplay even more so in The lost legacy.
If I remember correctly it's a core gameplay aspect in The Lost Legacy.

I agree the slow walking parts with talking isn't gameplay I never said it was but I enjoy those moments of downtime.

Let's be clear here Iroboto said the winch in Uncharted isn't gameplay and even went so far as to say it looked painful to him. That right there is the problem with most discussions on forums, the absolute hyperbolic statements by people that's the real problem. You don't like something fine but don't make ridiculous comments like that.

To make matters worse from what I gather he hasn't played it either.

 

[Shifty Geezer]

It's not important how unrealistic this is, what matters it is, realism IS our goal with games.

Very debatable. Some are concerned with telling stories, and when it comes to telling stories, people are very happy for completely unrealistic given Hollywood standards. Some are also just concerned with fun. I don't think a reality engine is necessary to create games that are fun or tell stories. Heck, the kind of games some of us are complaining about seem loved by millions of people, so it's not like there's clear evidence they need changing.
We can't work towards evolving stories built from within the engine, so we need to find other ways to tell stories in games based on the limitations the technology presents. That's what games have been about since the Atari 2600 days - best we can do is assemble some blocks to look vaguely like a person, and in so doing, we convey the sense of adventure for the player of traversing a jungle. The interactive elements of these games are trying to create a sense of presence. Uncharted 1's use of waggle for the balancing on the narrow bridges is an example. We can see the devs took the idea of balancing in real life and translated that to the new option of having to balance the controller. Then the communication of the need to balance couldn't come from our internal sense, so they used an on-screen prompt. The end result was an attempt that just plain failed, but it was all part of that working around limitations of technology that video games are about.

 

[JoeJ]

I am not sure if you understood what I said. You are suggesting the ability of the engine to create it's own "story" which is a very complex mechanism

At first i suggest we start with dynamic alternation of the story. Example:
Cutscene alike conversation between the player, the quest giver and a third NPC. For some reason the player decides to shoot the quest giver during the talk.
As it is now, this is impossible becasue there so no recorded speech and animation that could deal with this situation.
To solve it, we can either prerecord any option of what might happen, but this takes too much efford.
But if a game character has the ability of reflexes and intelligent locomotion (which is doable), the animation problem is solved.
For speech, the only options i see is prerecorded panic reactions or something, likely that's nacassary anyways in a game about combat.
So we had the option to adapt prerecorded cutscenes to the players actions, getting rid of the current illusion breaking way to pause the interactive experience and switch to movies (which is wrong, but all we can do for now).

I think we have seen things like this already, but lack examples.

I do not dream about full reality simulation to become a thing in short time - i only say this is what we are heading to, which implies the concept of static cutscenes will become more dynamic with the years, until in the end we found our own way of story telling that differs from movies or books.
Plus, as said before, we do not need procedural storuies at all. In real life stories result from inteteraction between real people, and with multiplayer we have this option too. But this can't work if all they can do is getting quests, with limited options to solve them. With better simulation of reality their options become rich, ans so interesting relationships and stories will happen automatically. If we are able to design the game wolrld and rules in a way this is encouraged.

To talk realistically, i have to leave the cuscene topic. The real need we have is game charcters than can control their body on their own, so they no longer depend on animation data. This lifts ancient restrictions like 4-way movement, inability to interact with the environmant, poor physical representation of game worlds although looking realistic, and much more.
We need this for many reasons. The additional options in regard of story telling are just some extra bonus we might not even want to utilize, but it's worth to mention.

I do understand what you say. But are you willing tu understand my point as well? Saying it is unrealistic lacks looking at the larger picture and vision i try to submit. I had proper simulation of walking, selfbalancing ragdolls many years ago - it's posiible and not a performance problem.

 

[JoeJ]

so we need to find other ways to tell stories in games based on the limitations the technology presents

IMO, game development is about lifting technological limitations on the long run, and then utilizing some of those available technoligies to design games that ar fun on the short run.
Why is it always so hard to convince people about the need to keep technological progress going on? It's not that achieving my proposed goals would take anything away - it only adds.

so it's not like there's clear evidence they need changing.

All those perfect games you have in mind with this, classics that deserve a remake, all of them would not have been possible without progress in hardware and sotware technology.

If you want more such classics to appear in the future, then be open minded to new technology. Otherwise what happened to rock music will happen to games. Technical progress is free fuel, and we should be lucky our industry has it while others do not.

 

[Nesh]

I do understand what you say. But are you willing tu understand my point as well? Saying it is unrealistic lacks looking at the larger picture and vision i try to submit. I had proper simulation of walking, selfbalancing ragdolls many years ago - it's posiible and not a performance problem.

I didn say it is unrealistic. I understand your point and I can see great implementations if one day becomes a reality.
What I said is that the idea will find it's application in variable degrees depending on the game the developer wants to make even if the engine becomes fully capable. For example lets say a developer wants to tell a specific story and an NPC character is important from the beginning until the end both story wise and how it behaves in the game during gameplay, the director will demand a restriction in your ability to hurt him even if the technology allows more. I.e lets say we are playing The Last of Us 3 and the script writer decided Ellie and Joel will stay alive and do A and Z things until Chapter 9 killing any of them will be a NO. Some elements of the game will be pre-scripted to tell the story how the writer wants it to unfold, some gameplay elements will be in a certain "bounding box", whereas other elements that do not interfere with the story or a specific gameplayu experience (i.e secondary characters or some other aspects of the characters) will receive the "simulation treatment" to allow some story and gameplay branching.
Another developer on the other hand might consider a different experience and take more or fully advantage of the simulation.

It is the same thing with current technologies. Some games use advanced and contextual animation blending, some decide not to. Some games have complex traffic simulation like GTA, some decide not to. Some games have rugdoll physics, some decide not to. Some games decide to have procedural environments, some decide not to. Some games decide to have fully interactive vegetation some decide not to. Looking into the future one game will decide that burning the vegetation of the jungle is part of the gameplay, hence simulation is important, the other one decides that the lavish vegetation of the jungle should remain dense for gameplay or story purposes hence it will impose a certain "bounding box".

We are still going to see some of the "old" practices being implemented along with the new.

 

[Shifty Geezer]

IMO, game development is about lifting technological limitations on the long run, and then utilizing some of those available technoligies to design games that ar fun on the short run.
Why is it always so hard to convince people about the need to keep technological progress going on? It's not that achieving my proposed goals would take anything away - it only adds.

Huh? I wasn't in any way saying we shouldn't progress technology. If you want to discuss actual, implementable game changes, you need to talk about what devs can do. Just as it was pointless talking about what RTRT could bring to gaming in discussions held in 1986, there's no point talking about dynamic stories created in-engine in 2019. Instead, pick targets that are reachable and can be worked towards.

But if a game character has the ability of reflexes and intelligent locomotion (which is doable), the animation problem is solved.

I think that's an oversimplification. We've had behavioural animation tech since PS360. It hasn't appeared mainstream. In fact, I think it's barely been implemented at all. So you have to ask yourself why? Is the technology too weak? Or does it affect the game in some way? If it's really possible to simulate human movements, why do we bother with motion capture? Why has that been a constant area of work for the past 20 years instead of getting the computers to do all the acting for us?

I don't think computer-acting is as progressed as you think it is, which is why ND are using hundreds of prerecorded animation clips to assemble together their highly-acclaimed movements. That said, I don't know why computer acting isn't very progressed and don't know what opportunities next-gen may present.

 

[JoeJ]

For example lets say a developer wants to tell a specific story and an NPC character is important from the beginning until the end both story wise and how it behaves in the game during gameplay, the director will demand a restriction in your ability to hurt him even if the technology allows more.

I would deal with this like with the supermarket example: If the player kills an important NPC, the game kills the player quickly and he has to start over. I don't say this is how any game should be, but freedom of choice is desireable and here we could give the player the illusion to be free.
But this alone would not be worth the efford ofc. It's always hard to predict the outcome of new technology - we will see...

there's no point talking about dynamic stories created in-engine in 2019.

Ahhm... there is already a game out with procedural dynamic stories. But can't remember the name - looked like some JRPG.

If it's really possible to simulate human movements, why do we bother with motion capture?

Because existing tech barley works and is difficult to use. Requires more work.

There are multiple ways to do it, if we simulate physics the very first problem to solve is balancing. For bipeds that's very hard, because the center of mass is very high but the support area from the feet is small. Typical game physics engines can't do the jonts and motors accurately enough, so you may need custom solvers to get started.
After that, the control problem is very hard too. Balncing an inverted pendulum efficiently gives a problem of position dependend acceleration leading to complex equations, and converting the results back to a complex human body requires IK solver that preserves the solution.
Approximations may not be good enough, because they reduce efficiency. The balancing human body constantly acts at the edge of what's is physically possible, even to get simple things done in short time.

It's difficult, otherwise there would be more companies with biped robots walking under unstable balance. Actually Boston Dynamics is the only one i'm aware of. (I guess i would need 2-3 years of work to catch up)

An alternative is procedural animation, and recently there is lots of work doing this with machine learning on real world data (so motion capture still required). We may see this first in games i guess. This is not the real thing, but looks very realistic and it's much easier i guess. *)

Both can be mixed. In the end, i'm convinced physics simulation is best suited to solve basic balancing problems like walking, running, carrying objects, reflexes etc., and ML can be used to add everything else.

So, when i say 'it's possible' i mean performance wise, talking about myself. I don't know what Natural Motion can do exactly, but without doupts Boston Dynamics could offer dynamic ragdoll library middleware to games in very short time if they wanted.

Maybe the biggest reason we do not have this yet in games is this:

Why is it always so hard to convince people about the need to keep technological progress going on? It's not that achieving my proposed goals would take anything away - it only adds.

hehe

One thing for sure: After making your characters walk realisitcally, adding body language and facial expression on top to play back cutscene data is very easy.


* first random search result on YT, does not look so different form actual games. TLoU 2 maybe already looks better but also quite dynamic. So this is happening and not far fetched.

edit: fixed AI <-> IK typo

 

[PSman1700]

Some people are more wrapped up in the visuals so as long as a game looks exciting, that’s all that matters to them.

And that's what matters for sales, it does for much better marketing, 'back of the box screenshots' and finally reviews. It's all economics here.

What? This makes no sense, I shouldn't be on forums because I don't want to have an uneducated opinion on a topic I haven't experienced.

You can't expect people to praise what you like, on a forum we have discussions, also, this topic isn't based on uneducated opinions because it's a fact certain games take a very cinematic approach, some ar for it, some are against, that opens up for discussions. If we can't have those discussions, it wouldn't be a proper forum. You're free to challange peoples views but calling them uneducated just because it doesn't suite you're thinking isn't the way to go.

I guess that's why Sony's recent games are selling better than ever before.

Exactly, that was my point. The cinematic approach, filmic and very nice looking graphics, and excellent marketing, all which result in raving reviews is what sells. Do you really think the avarage GoW/death stranding buyer made a request or demand for 6 to 7 hours of cutscenes, and jeep-wrench/boat ride sequences?

In fact I almost enjoy the downtime

You almost do enjoy it, that's fine, but understand that some don't. Gaming is about being in control, that's why i like it.

To point how meaningless it is?

If it would be meaningless, i assume the discussion wouldn't have gone on for so long, due to both by interest and moderator action. Furthermore, discussions arenät meaningless if they are based on something, which this one certainly is.

I find it pretty intrusive in the ND games. I haven't played or watched GoW, HZD or Spider-Man for comparison.

They are largely the in the same direction, one more so then the other. HZD probably the least.

That right there is the problem with most discussions on forums, the absolute hyperbolic statements by people that's the real problem. You don't like something fine but don't make ridiculous comments like that.

That's not true, he's having very good arguments in his posts, not only in this discussion. You're attacking people personally over a discussion about gaming, and then calling others comments uneducated..... :S

 

[iroboto]

Let's be clear here Iroboto said the winch in Uncharted isn't gameplay and even went so far as to say it looked painful to him.

This is where I'm positive there are some comprehension issues with my post. I didn't say the winch was a mechanic that was painful and it should be outright removed. I said when the mechanic is being used to hide a loading screen is painful.

Let me be clear. This is par for the course for all games, and it doesn't apply to UC4 only.
Everyone does it.
https://www.gamesradar.com/the-secr...and-why-they-wont-be-going-away-anytime-soon/

In other words, American Wasteland did have loading screens, but it tricked the player into thinking they weren’t there by funneling them into mandatory “loading tunnels” that could render new zones as they were travelling (very slowly) towards them. Game developers deploy these optical illusions all the time to hide their loading process without cutting to an inanimate screen, tricking the player into thinking they’re just enjoying a slightly sedated, more linear part of the experience.

Remember when Gears of War made Marcus slow down to walking pace and talk to his buddies over the radio coms? Or the infamous elevator trip in Mass Effect (not so subtle, that one)? Or, to venture into even more recent territory, how about the Tree of Life in God of War? What, you thought activating its realm-hopping machinery every time the story required Kratos to venture to distant lands was just a conveniently plodding plot device?

Lift back the curtain on each of these moments, and you’ll find the game’s code scuttling around like stage managers arranging props on a theatre set, desperately working to put everything in place before players even begin to notice they’re being held up. Jak & Daxter was one of the earliest games to adopt this seamless loading technique, and Naughty Dog’s co-founder Andy Gavin describes the techniques employed by the studio while working on the iconic platformer as a Producer.

When it's hiding a loading screen is when I have a problem with it.
I timestamped all the winch sequences I could find in UC4 in the video above, and to avoid getting my post edited, I refrained from posting the gif of each time. But it's pretty clear to me that it's there to help with loading. I found 1 time it was used for an off-side quest puzzle. The rest of the time you needed to use the winch to move to the next area. And every time you used the winch it made sure you couldn't go back. The winch that shortly after you pull on the leg you had moments to go back before the jeep started going up the elevator. The winch that throws your car over the cliff.
Most of the winch sequences i saw were gates, and each one took at least 2 minutes to get through. With some exception.
The winch isn't bad, but hiding load sequences with it is where I have the problem.

Gears of War does the exact same thing. To open a door always requires the second NPC to help open it. That's a loading screen. Jack is a NPC character, sure like the jeep, and has abilities. But every once in a while, you get to a room where only jack can access it to unlock the doors. And you have to find the spot and send jack, and you wait there for.. 1-2 minutes until jack gets the door open and you go.

As I see it, as hardware gets better, loading screens go away, but using this mechanics to hide loading screens means you are forever bound to have to do it, regardless of how many play throughs or how powerful your hardware is. And _that_ is what is painful about it.

To make matters worse from what I gather he hasn't played it either.

We're all adults here. We make millions of decisions on things we haven't tried to know we don't want or to want it. As humans we are more than capable of transferring our learning from our past to our present. I don't need to play UC4 to have experienced the interactive loading sequence. We've seen a fair number of examples of NPCS helping each other up, or squeezing through corridors. Having to lift one thing up helping 1 NPC pass and then in turn they help your character through.

We've seen these interactive cutscenes hide background loading all of this generation. And if you love it, that's great. But next generation when they have the hardware to actually get around this, I don't expect to see this type of thing anymore at all. It may be, as you say, these mini loading sequences actually become a core mechanic to greater gameplay.

 

[milk]

I am not sure if you understood what I said. You are suggesting the ability of the engine to create it's own "story" which is a very complex mechanism that lets the engine decide based on random actions. Even if that ever becomes feasible the director or developer, depending on the game will want more or less control of what is happening either in terms of story or how the game is played to provide a certain experience. Unless you are also suggesting that the engine works like some super AI that has superhuman creative, artistic and game design skills that decides on the fly for any 3D game in the same manner a director, developer or game designer plans for months or years. In other words even if that ever happens, developers will opt in or out in variable degrees depending on the game they want to make.

Dwarf Fortress does that, through text. What he is saying, is when procedural speech and animation hit a certain sophistication level, we might start seeing AAA cinematic games aproach the kind of flexibility and responsiveness of Dwarf Fortress, and less that of Night Trap.

 

[Nesh]

I would deal with this like with the supermarket example: If the player kills an important NPC, the game kills the player quickly and he has to start over. I don't say this is how any game should be, but freedom of choice is desireable and here we could give the player the illusion to be free.
But this alone would not be worth the efford ofc. It's always hard to predict the outcome of new technology - we will see...

Yes thats something we already got with some games, where killing your ally ends the game. You see that's the thing. Freedom of choice is desirable either more or less based on context. It is not a panacea.

Dwarf Fortress does that, through text. What he is saying, is when procedural speech and animation hit a certain sophistication level, we might start seeing AAA cinematic games aproach the kind of flexibility and responsiveness of Dwarf Fortress, and less that of Night Trap.

Yes we might see AAA titles aproaching that flexibility, but who dictates that this will be the norm (not due to technological limitations but due to developer choice)?

If it would be meaningless, i assume the discussion wouldn't have gone on for so long, due to both by interest and moderator action. Furthermore, discussions arenät meaningless if they are based on something, which this one certainly is.

But it has gone on for so long, and people were pointing out that the problem you think is a problem is not really a problem, just a particular game design that doesnt fit your personal taste but serves it's purpose well for the experience they targeted.

 

[DuckThor Evil]

When it's hiding a loading screen is when I have a problem with it.

I don't understand this, surely hiding a load screen is the one instance such mechanic has a place? Looking at a static 1-2 min loading screen is the thing that really sucks and until that can be completely removed, hiding it is the best you can do.

 

[Xbat]

We're all adults here. We make millions of decisions on things we haven't tried to know we don't want or to want it.

I never said you should play it or do anything you don't want to, just that for me sharing ones opinion on a game should be backed up by actually experiencing said game.

Ok the loading screen makes more sense but even that I don't get, how do you prefer loading screen overs the clever ways they hide it? Now we getting to personal preference I know and whatever floats whoever's boat is perfectly fine but for me I loathe loading screens it's the most jarring thing that can happen when I'm getting into the story of a game so much so that I tend not to even use fast travel in open world games very often.

 

[JoeJ]

Yes we might see AAA titles aproaching that flexibility, but who dictates that this will be the norm (not due to technological limitations but due to developer choice)?

Admit we can't know.
What sounds realistic to me is a scenario where a Keanu Reeves still appears in AAA games, but less dramatic charatcters like a minor quest giver are purely artificial - if only to save some money.
Likely there will be some bluring where AI learns Keanus voice, body language and facial expressions, and so can become dynamic and adaptive. Static movies will appear old school in the modern game of the future i guess.
It's just one step more beyond scanning and motion capture, and usually if something is technically possible, it does become the norm.

Freedom of choice is desirable either more or less based on context. It is not a panacea.

Yes, and freedom of choice is also boring and not what i want. What i want is a consistent experience of being 'in the game', without interruptions. Taking away control from the player, or even just changing the controls is such an interruption, and making selected characters resistent to gunfire is another. (I guess there would be better examples.) Those things constantly remind us we are fooled, and normally you don't want this when designing games.

Like most people i have no desire for computer generated stories or synthesized speech. I assume this could not compete manual work. But i (we) might be wrong.
For example, i always said ML could never learn to compose earwig melodies, because even humans that can do this do not know how it works, why some melodies stick in our heads but most do note.
Then i found this:

https://soundcloud.com/id%3Dyichao-zhou-555747812%252Fsample-32-measures%3Btrack_id%3D348935584

And this is really very good, but genearted by ML. It proofes me wrong. Also the Mitsuku chatbot is able to hold an intelligent conversation over multiple sentences, if you are just a bit forgiving.
At this point dynamic / adaptive cutscenes no longer seem so far away. All those idle Tensor cores need some work

Edit: But i do not think it's a next gen thing - give it a decade.

 

[JoeJ]

Dwarf Fortress does that

Ha, did not know this. It's not the game i can't rmember, so there may be more than 2 games already.

 

[iroboto]

I don't understand this, surely hiding a load screen is the one instance such mechanic has a place? Looking at a static 1-2 min loading screen is the thing that really sucks and until that can be completely removed, hiding it is the best you can do.

Ok the loading screen makes more sense but even that I don't get, how do you prefer loading screen overs the clever ways they hide it? Now we getting to personal preference I know and whatever floats whoever's boat is perfectly fine but for me I loathe loading screens it's the most jarring thing that can happen when I'm getting into the story of a game so much so that I tend not to even use fast travel in open world games very often.

It just comes down to understanding the limitation behind it.
If this next gen hardware works the way we expect it to work; you don’t need to design and spend production costs to make these interactive cutscenes just to do background loading.
The up swing is that the first time through the title, you’re generally okay with it. Additional play through or having stronger hardware ie PS5; the PS5 cannot eliminate those sequences now even if the loading is near instant. If there was a XB2 or a PS5 enhanced edition; you’re still going to spend an hour on interactive loading sequences.

These hidden load screens are probably a symptom of these last two generations. But designing a game with them in mind vs a game that doesn’t; the gameplay flow is very different. Looking at RDR2, FH, and other open world games without hidden loading gates.

And then the preference comes down to actually having downtime. Loading screens is a time for me to reflect on the game; plan what to do next; bio breaks; stretching; checking the phone; typing messages etc. I can get around the load time but it’s loading while I can do my thing.

When I’m forced to do the interactive cutscene I can’t do my own thing so really to me there is never any break without a loading screen.

It’s weird. There are times in which I have seen serious faux pas with loading screens; like a 5 minute load and you die and have to do it all over again.

But these interactive load screens are also feeling pretty faux pas. It can really kill the flow of the game. When you just want to get on with the interesting bits.

 

[Shifty Geezer]

Then i found this:

https://soundcloud.com/id%3Dyichao-zhou-555747812%252Fsample-32-measures%3Btrack_id%3D348935584

And this is really very good, but genearted by ML. It proofes me wrong.

Sounds to me like random music from someone who doesn't know what they're doing. I have quite a lot of faith in ML, but I'm also starting to see patterns in its limitations. It's good at filling in broad strokes, but cannot do details like a human can. I'm sure ML could move and populate a crowd good enough, but I really doubt it'll be able to take on a significant role in games any time soon. The place we'll see it first is Hollywood and CGI. Virtual actors won't need paying, either as actors or as a fleet of CGI artists and animators, so creating virtual actors that can act, and virtual script writers that can churn out episodic content, is a Holy Grail for them. Until Hollywood is able to manage that with 'limitless' processing power, we shouldn't be expecting that from home consoles.

 

[PSman1700]

But it has gone on for so long, and people were pointing out that the problem you think is a problem is not really a problem, just a particular game design that doesnt fit your personal taste but serves it's purpose well for the experience they targeted.

Some like it, some don't. Yes they targetted a cinematic approach that doesn't fit my personal taste. The whole discussion started with Death Stranding, a game with 6 or 7 hours worth of cutscenes, the same as GoW. No idea if the gameplay itself is also cinematic in DS but it wouldn't surprise me, only how the MP is done then.

 

[JoeJ]

Sounds to me like random music from someone who doesn't know what they're doing.

Hehe - with this you admit it sounds human made. You should hear earlier approaches of computer generated music.
But notice this is just notation played back by a cheap midi virtual instrument, similar to music in early PC games if you remember. Better arrangement, better sounds, and you could make a good commercial jingle out of this, maybe even a radio hit. Very catchy.
(Otherwise i share your doubts about ML being useful for games - but i do not know much about it.)

The place we'll see it first is Hollywood and CGI.

Sure. Let them be first, as always. meh.

Virtual actors won't need paying, either as actors or as a fleet of CGI artists and animators, so creating virtual actors that can act, and virtual script writers that can churn out episodic content, is a Holy Grail for them.

No, it's the holy grail for us. In games uncanny valley is acceptable, in movies it's not. They already have camera equipment and actors - this is what they do. For games this is extra costs to save. In games this tech will increase the quality of animation, in movies it will only decrease it.
And most important: Movies are not dynamic, so they have no need for this at all, except for some special cases of super crowded scenes to save money (and where they already use it).

But i see taking cutscenes as an example was a very bad choice of mine. Personally i have only basic locomotion in mind, during gameplay - not Shakespeare during cutscenes. I admit the latter is far fetched and leave it at this.

 

[Nesh]

Admit we can't know.
What sounds realistic to me is a scenario where a Keanu Reeves still appears in AAA games, but less dramatic charatcters like a minor quest giver are purely artificial - if only to save some money.
Likely there will be some bluring where AI learns Keanus voice, body language and facial expressions, and so can become dynamic and adaptive. Static movies will appear old school in the modern game of the future i guess.
It's just one step more beyond scanning and motion capture, and usually if something is technically possible, it does become the norm.


Yes, and freedom of choice is also boring and not what i want. What i want is a consistent experience of being 'in the game', without interruptions. Taking away control from the player, or even just changing the controls is such an interruption, and making selected characters resistent to gunfire is another. (I guess there would be better examples.) Those things constantly remind us we are fooled, and normally you don't want this when designing games.

Well then I guess we are going to have procedurally generated cinematic games and cut scenes.
We cant just say it is wrong because this technology is not available. Under this logic all games are wrong.