Non-player Criticisms of The Last Guardian *pile*

[bgroovy]

I do find the idea that the AI needs to be more directly responsive totally wrongheaded. Even professional game critics have a very bad habit of assuming incompetence on the part of developers when they find something frustrating. Sometimes that frustration is the only thing that makes the system work. Without the frustration there's no sense of progress. If everything is perfectly predictable suddenly all you see is a mechanism. In the case of this game that is totally antithetical to the goal of the experience.

 

[rabidrabbit]

I've only had one occasion when I was getting frustrated with Trico and I think it was a bug
I must have spent an hour trying to get Trico away from that hanging blue thing he was so interested. He (she?) did obey occasionally, but always went back to that thing before I cold use Trico as a platform to get where I needed.
Restarting from a checkpoint soled it, after restart Trico was already up there with me where we needed to be.
The problem was that I was not sure if it was a bug or just normal behaviour, as Trico's behaviour made sense in the context of the situation. That's why I spent an hour trying to make Trico do what I wanted. I think I should have saved that barrell and used it to lure Trico.

 

[goonergaz]

I agree that the response has been strangely emotional.

I guess the bond is much like someone dissing your pet for no good reason. For myself I have waited many years for this, I know it won't be perfect but I know it will be a unique magical experience - however it's for a niche group of gamers. As such people dissing Trico are clearly not in this niche and it'd be something like me going into online FPS threads and complaining endlessly because everyone has a better gun than me - or knows how to strategically play FPSs online (etc).

 

[Sigfried1977]

I had no issues with the Trico A.I. whatsoever. Like I said it the other topic: just reluctant enough to maintain the illusion of free will and therefor a life of its own, but never enough to cause frustration. If you have zero patience and/or exclusively subscribe to the Valve/Blizzard school of game design which aims for the removal of any and all friction, then this could become a problem. If you didn't grow up with pets or don't appreciate them in the first place, you might wanna steer clear of TLG as well.

I also think it's rather typical of someone like Jim Sterling to blame any and all of his failings on the game at hand exclusively. Never ever on himself though. Just watch what happens right at the end of his video: he overjumps his target by a country mile, only to rant about Trico for not having caught his rotund ass in mid air. Pro tip: next time, aim your jump more carefully. This is not Uncharted. It's not gonna take over and magically guide your trajectory the moment you press a button.

If you hate something because you're shit at it that is perfectly fine I think. In fact I think it's a valuable perspective. Heck, I wished there was at least some people who didn't try and review the shambolic likes of Street Fighter V through the excusatory prism of high-end, competitive tournament play and instead laid the smack down accordingly. Because for the average player, the game offered almost nothing in its sorry-ass launch state. Doesn't mean you should accuse everyone else of being an ass-kissing shill for seeing things differently.

 

[OCASM]

Having trained a few dogs and cats, both. Unless both you and the animal are very well trained, it will not always do what you want it too or what it has been trained to do. It's a living, wilful creature. Some days it'll be excited, pissed off or stressed and at those times it's behaviour will be more erratic.

'Never work with children or animals' - that maxim exists because no animal can be relied upon 100% to do what it's supposed to (which also applies to humans!).

So it's behavior is non-deterministic, just like I said.

'Yes if that's the intention. How about a game where your troops can get scared and not do what you want them to (XCom) leading to the end of the mission and failure? It becomes a mechanic you have to adapt to. That relationship with the animal including its independence is part of the game design, just as owning a pet and having it not do what it's told 100% of the time is part of the experience of owning a pet and why people don't get pet robots.

I doubt Ueda's intention was for players to become frustrated with Trico.

' I agree that the response has been strangely emotional.

There's always people who get angry when you don't praise their favorite things.

'We have exact quotes from players to the contrary in this thread. Why are you dismissing them? People in this thread, playing the game, are getting along fine with Trico and enjoying the AI and the game design surrounding that. Ergo it is not bad game design. There's zero way to argue it's bad game design when the purpose of game design is to entertain people and people are getting entertained by TLG!

Maybe you missed this post where I linked to other players agreeing with my assessment.

You guys know this will not stop, right?

How about spinning off the discussions that are general preconceptions of what game design should/shouldn't be, and leave the people currently playing or at least interested, discuss their impressions of this specific game?

No criticism allowed.

It's not possible to assess what works and doesn't work in TLG without playing it and get far enough in the game. It's pushing the boundaries enough that it has no comparisons. It's also difficult to argue against misconceptions without getting into spoilers.

What boundaries is it pushing?

This was my point about Agro. The player had direct control of the reins but not of the horse. It would be like a game where you had direct control of the sails but not of the ship.

I find the investigation of that relationship very interesting.

Well you control the horse directly through the reins just like you control the player characters through a game controller. In the case of Trico you can only suggest it what to do and the AI will decide whether it listens to you or not.

I do find the idea that the AI needs to be more directly responsive totally wrongheaded. Even professional game critics have a very bad habit of assuming incompetence on the part of developers when they find something frustrating. Sometimes that frustration is the only thing that makes the system work. Without the frustration there's no sense of progress. If everything is perfectly predictable suddenly all you see is a mechanism. In the case of this game that is totally antithetical to the goal of the experience.

The sense of progress comes from the problem managing to solve a problem or completing a task. I don't see how an unresponsive AI contributes to that.

I guess the bond is much like someone dissing your pet for no good reason. For myself I have waited many years for this, I know it won't be perfect but I know it will be a unique magical experience - however it's for a niche group of gamers. As such people dissing Trico are clearly not in this niche and it'd be something like me going into online FPS threads and complaining endlessly because everyone has a better gun than me - or knows how to strategically play FPSs online (etc).

That's just a cop out. It's like defenders of modern art: if you don't think it's perfect it's because you don't understand it. Nah.

I also think it's rather typical of someone like Jim Sterling to blame any and all of his failings on the game at hand exclusively. Never ever on himself though. Just watch what happens right at the end of his video: he overjumps his target by a country mile, only to rant about Trico for not having caught his rotund ass in mid air. Pro tip: next time, aim your jump more carefully. This is not Uncharted. It's not gonna take over and magically guide your trajectory the moment you press a button.

Slow motion without Trico reacting is just misleading and annoying. Just let the player die so he/she can restart faster.

 

[MrFox]

I doubt Ueda's intention was for players to become frustrated with Trico.

Trico can be quite disobedient. Do you worry players might find that frustrating?

Fumito Ueda: It's something that's deliberate in the game. There is a worry that it might stress out some people out there. But this game isn't continual action, it's not fast-paced. Whether pepole stress out about that is down to personal preference. If you take music, some people like hard rock music that's a bit more fast-paced. Some people might like slower music. Their personal preference might stress them out.

 

[Nesh]

Seems like you want to have your cake and eat it too. Either the AI behaves like a realistic wild beast that constantly disobeys the boy's commands OR it behaves like a tamed animal that never gets in the way of the player progressing.

Define constantly. If the game has rules which the player needs to understand in order to avoid "disobedience" then it is not constantly

Realistic animal AI = good game design?

Realistic animal AI=Bad design?

The AI always obeys the boy's commands?

Does the AI always disobey the boy's commands?

Just stating a fact. The framerate does suck.

And your goal is?

Except I didn't. They jumped my throat simply for saying that Trico's mechanics are fundamentally different than Agro's. What a crime. Why have a civil discussion when you can simply throw ad hominems at the dissenter?

You are trying to tell people that the game design is bad which means the game sucks
You are trying to convince people who ENJOY the game and do NOT find it frustrating, that the game is frustrating. Do you understand that?

Seems like people like Trico's AI by itself, not it's relation to the game design:

Its part of the whole experience. Why should it be dismissed completely?

 

[Shifty Geezer]

So it's behavior is non-deterministic, just like I said.

I don't disagree with that. That doesn't make for bad game design though. It's the very basis for good AI in many games, like shooters for example.

I doubt Ueda's intention was for players to become frustrated with Trico.

You can't please all gamers. What some players get frustrated with (lacking patience perhaps), others will love.

There's always people who get angry when you don't praise their favorite things.

To be fair on your detractors, you argument is a bit all over the place and you are implying anyone who likes the game is an idiot.

Maybe you missed this post where I linked to other players agreeing with my assessment.

Missing the argument. As above, Ueda isn't trying to please everyone on the planet. His choices for game design are for his audience. The fact players (not all of them) are loving Trico means his choices were correct for them. He could have changed the design so Trico was an obedient as a robot and others would like it more then, but those who like the present design would dislike it. Some gamers lie fast-paced twitch shooters. Others playing the same game would say they'd rather it was slower and more tactical. If the designers were to change the game to make it slower and more tactical, they gamers who like the fast, twitch shooter would complain. Pick your audience and make the best game for them (in Ueda's case, make the game he wants to make and the audience will be whoever likes it!).

No criticism allowed.

Criticism perfectly allowed. The suggestion was to spin off this discussion to keep the game discussion focussed and the particulars of AI and game design here. Comments like these only encourage a less constructive response and discussion.

Well you control the horse directly through the reins just like you control the player characters through a game controller.

Very wrong. When you control a player character, they do exactly what you input, one to one. Agro instead takes a suggestion from the reins but may do his own thing, like a real horse.

In the case of Trico you can only suggest it what to do and the AI will decide whether it listens to you or not.

Like a real animal which is what Ueda was trying to simulate. Also like Agro - He misbehaves..

That's just a cop out. It's like defenders of modern art: if you don't think it's perfect it's because you don't understand it. Nah.

That's arrogant, that anyone who disagrees with you is a blind fanboy. Why can you not accept that Trico is a perfectly legitimate game design, just not one that appeals to you nor everyone? Why is everyone who says they like Trico a liar or delusional because they have different tastes to you?

At first I felt the others were being hard on you, but now I'm with them and think you suck too, for your very sucky attitude and narrowmindedness.

 

[Deleted member 11852]

So it's behavior is non-deterministic, just like I said.

Except that isn't what I said. You can train an animal but the training isn't absolute.

 

[OCASM]

...

He says that Trico's disobedience is deliberate, not causing player's frustration. Just like musicians who produce slow music don't do it with the intention of stressing those who like hard rock, lol.

Define constantly. If the game has rules which the player needs to understand in order to avoid "disobedience" then it is not constantly

If such rules exist I'd like to see them in action.

Realistic animal AI=Bad design?

No.

Does the AI always disobey the boy's commands?

No.

And your goal is?

Express my opinions / observations.

You are trying to tell people that the game design is bad which means the game sucks

I think the game has it's flaws, not that it sucks. Extremism is bad

You are trying to convince people who ENJOY the game and do NOT find it frustrating, that the game is frustrating. Do you understand that?

Not at all. Although the notion of universal praise by its players is factually wrong as exemplified by the videos I posted.

Its part of the whole experience. Why should it be dismissed completely?

Should it be dismissed completely?

Except that isn't what I said. You can train an animal but the training isn't absolute.

I'm talking about Trico's AI, not real life animals.

I don't disagree with that. That doesn't make for bad game design though. It's the very basis for good AI in many games, like shooters for example.

You can't please all gamers. What some players get frustrated with (lacking patience perhaps), others will love.

If it gets in the way of the player's enjoyment and progression, it is.

To be fair on your detractors, you argument is a bit all over the place and you are implying anyone who likes the game is an idiot.

When did I imply such a thing? My argument isn't all over the place. It's very clear: forcing the player to rely on a non-deterministic AI to progress is bad design since it replaces player agency with randomness.

Missing the argument. As above, Ueda isn't trying to please everyone on the planet. His choices for game design are for his audience. The fact players (not all of them) are loving Trico means his choices were correct for them. He could have changed the design so Trico was an obedient as a robot and others would like it more then, but those who like the present design would dislike it. Some gamers lie fast-paced twitch shooters. Others playing the same game would say they'd rather it was slower and more tactical. If the designers were to change the game to make it slower and more tactical, they gamers who like the fast, twitch shooter would complain. Pick your audience and make the best game for them (in Ueda's case, make the game he wants to make and the audience will be whoever likes it!).

That's just a cop out. By that logic no bad games exist, since if anybody dislikes them it simply means that they're not the intended audience.

Criticism perfectly allowed. The suggestion was to spin off this discussion to keep the game discussion focussed and the particulars of AI and game design here. Comments like these only encourage a less constructive response and discussion.

Well his argument there was to create another thread to discuss the generalities of game design, not the particulars of this game.

Very wrong. When you control a player character, they do exactly what you input, one to one. Agro instead takes a suggestion from the reins but may do his own thing, like a real horse. Like a real animal which is what Ueda was trying to simulate. Also like Agro - He misbehaves..

Agro's behavior is completely deterministic. Its reactions to environmental obstacles are 100% predictable. Wander "misbehaves" as well. Try to make him run off a cliff and it will disobey you.

That's arrogant, that anyone who disagrees with you is a blind fanboy. Why can you not accept that Trico is a perfectly legitimate game design, just not one that appeals to you nor everyone? Why is everyone who says they like Trico a liar or delusional because they have different tastes to you?

I disagree with the position: "it's perfect, you just don't get it". How did you interpret that as me calling everybody who disagrees with me "a blind fanboy", "liar", "delusional" is beyond me.

At first I felt the others were being hard on you, but now I'm with them and think you suck too, for your very sucky attitude and narrowmindedness.

"I think this game has some flaws".
"You suck!"

Very nice.

And since everybody is hellbent in hating me might as well throw another stick into the fire :

HDR post-processing received a massive downgrade from it's initial 2009 trailer (not the leaked one). I remember being impressed with it backed then since only Halo 3 was anywhere near close to that level of realism:

Tone mapping was more realistic and the bloom had a longer tail. Now it looks clipped:

 

[Deleted member 11852]

I'm talking about Trico's AI, not real life animals.

And yet you introduced the discussion on comparing Trico's AI to a "a realistic wild beast".

 

[Karamazov]

i've met a few Tricos in my life and they behave exactly like in the game, it's pretty realistic.
the game is far from perfect, but the magic is elsewhere.

 

[Shifty Geezer]

If it gets in the way of the player's enjoyment and progression, it is.

If it gets in the way of the majority of player's enjoyment, yes. If lots of players like the design, it isn't. Again, the definition of good game design is making a game that people enjoy playing.

When did I imply such a thing? My argument isn't all over the place. It's very clear: forcing the player to rely on a non-deterministic AI to progress is bad design s...

You've said everyone who likes Trico likes bad game design. "If you were smarter and appreciated good game design, you wouldn't like Trico" is the implication. The logical conclusion to your argument, should you be successful, is for everyone who likes Trico to come to realise they were wrong, that Trico was annoying and got in the way of their progress, and to dislike the experience with the simulated animal.

That's just a cop out. By that logic no bad games exist, since if anybody dislikes them it simply means that they're not the intended audience.

No. It's not as black and white as that. You can come up with an analysis of a game's design that'd find 'bad' game design in a game universally panned. You could also find elements of bad design in a game that's lauded. You can't argue that a feature of game that's liked by a sizeable part of its player base is bad design though.

I disagree with the position: "it's perfect, you just don't get it".

No, it's "we like it, therefore it can't be wrong as you say."

How did you interpret that as me calling everybody who disagrees with me "a blind fanboy", "liar", "delusional" is beyond me.

"If you were smarter and appreciated good game design, you wouldn't like Trico."

And since everybody is hellbent in hating me might as well throw another stick into the fire :

Wrong thread. Also wrong fire - analysis of the visuals ain't gonna have any impact on the love of the feathery little critter.

 

[Nesh]

If such rules exist I'd like to see them in action.

Have you played the game? No. So why are you commenting negatively when you cant back up your claims by a first hand experience when those disagreeing with you have first hand experience?
Also since you havent played the game, how can you have an opinion let alone define "constant disobedience" in the game?

No.

Then whats the problem?

No.

Then whats the problem?

Express my opinions / observations.

I dont get your opinions/observations. You are calling the game frustrating and that it has bad design when people love it. You even used remarks that imply that people that like the game dont know what they are talking about. Why are you trying so much to jump from one negative to the next without coherence?

I think the game has it's flaws, not that it sucks. Extremism is bad

But you seem to find flaws in what people loved about it. Which also happens to be the backbone of the game. And you said that the game design sucks. If the game fundamentals and therefore design sucks you imply that the game sucks as well.

Not at all. Although the notion of universal praise by its players is factually wrong as exemplified by the videos I posted.

How on earth can you dismiss that people factually loved it and enjoyed it?
No matter how much people loved the game you are trying to forced down their throat that they didnt. I dont understand you. You dont make any sense.

Should it be dismissed completely?

No. Then why are you dismissing that it actually adds positively to the experience and you call it bad design?

Also why on earth are you trying to needlessly "throw a stick" into the fire? What are you trying to accomplish?
It is a unique enjoyable experience, targeted for a niche, and the final product is beautiful to behold thanks to its art direction. If the game was downgraded (which isnt but lets assume it is for the sake of the discussion) this wont change that it is still a great game, aesthetically and in terms of gameplay and succeeds at what is trying to do.

 

[MrFox]

forcing the player to rely on a non-deterministic AI to progress is bad design since it replaces player agency with randomness.

It's not randomness.

People are now doing speed-runs in under 5 hours, easily. It's even one of the trophy. You think it's just luck?

 

[milk]

It's not randomness.

People are now doing speed-runs in under 5 hours, easily. It's even one of the trophy. You think it's just luck?

Now it's empirical. Case closed. Thanks god.

 

[Shifty Geezer]

Not closed. Whenever an argument is proven, we need to provide time for those who were on the other side to humbly acknowledge they were wrong, to notice what they misunderstood of the arguments, and to learn how to better reason and comprehend counter arguments.

 

[milk]

Not closed. Whenever an argument is proven, we need to provide time for those who were on the other side to humbly acknowledge they were wrong, to notice what they misunderstood of the arguments, and to learn how to better reason and comprehend counter arguments.

That one never happens publicly, unfortunately.

 

[Shifty Geezer]

Indeed. I always like those posts when someone's big enough to admit they're wrong though. And TBH don't understand why people don't. What is it about one's particular arrangement of neurons (thought processes) that makes them sacred such that, when wrong, we can't just acknowledge as much and replace them with updated thinking?

 

[milk]

Indeed. I always like those posts when someone's big enough to admit they're wrong though. And TBH don't understand why people don't. What is it about one's particular arrangement of neurons (thought processes) that makes them sacred such that, when wrong, we can't just acknowledge as much and replace them with updated thinking?

It's not the neurons so much, but the way they initially present their arguments. When they put their ideas forth in such arrogant way in which it is implied that everyone else is a worthless idiote, being wrong becomes a much harder option.