One difficulty level? for Cyberpunk2077,Witcher 3 and other CDPR games

[andr01d]

Yes, but massive testing of AI have to be done anyway. But this way you have to test it against living players and against other AIs. And I would prefer if AI programmer would focus on programming AI with a human in mind rather than other AIs.

Why do you prefer AI programmed with humans in mind? I mean, AI would always require to react to certain conditions with certain (randomized) actions. Where's the difference between checking if the conditions are met by a human or the enemies AI?

Even more work to be done. As I said. It would be way more expensive than making more difficulty levels.

I didn't say that this way would be easier. I would only want it to be that way because it has less chances to be immersion breaking than just tweaked AI stats (because for 'really easy mode' or 'really hard mode' you may have to change the stats beyond their lore borders).

 

[freakie1one]

@AnDr01d:

Sure, if you take my suggestions to the extreme they wouldn't work. But I never suggested that CDPR should (and I would hope they would have the common sense not to).

When it comes to reaction times they simply need to find a reasonable number that works (say in the range of 25%-50% slower reactions). Instead of having an attacker react in 1 second he'll now take 1.5 seconds. This might not seem like a lot but gameplay-wise this would be enough to give people without as much twitch reflexes a chance to react.

Your logic behind tweaking an enemies damage modifiers is also flawed. Of course it would still take into account your armor and all your other stats. You simply decrease an attacker's base damage by a percentage before you calculate any stat modifiers (like armor, potions, etc.). So if an enemy would normally do 25 base damage, now he hits you for 20 instead. An example would be like this: attacker's base damage + stat modifiers + armor modifiers = total damage taken.

You also criticize my idea of having different (key word being different) enemies using better AI late in the game to keep the game challenging. A nekker would always use the nekker AI, no matter when you encountered one. Meaning that yes, if you encountered a nekker late in the game you'd have no problem killing one (which gives you a sense of actually becoming more powerful). But this doesn't mean they can't have other types of enemies later in the game that use more advanced AI.

You can twist my words and take everything to the extreme to try and make it look silly. If I would have known this was your intention I would have given very specific examples to prevent this kind of nonsense. Anyways, I hope I've made my suggestions more clear now.

 

[Aver]

Why do you prefer AI programmed with humans in mind? I mean, AI would always require to react to certain conditions with certain (randomized) actions. Where's the difference between checking if the conditions are met by a human or the enemies AI?

It's impossible to add random factor to every part of AI's code. If it would be then every game developer would do so - it would close a gap between human and AI. But it's impossible because of amount of work and testing that it would require. Even Crytek known for their devotion towards programming realistic AI didn't do it and yet their AI program was so complex, that when they were making Crysis 1 for consoles they had to program new AI, because consoles CPU were unable to handle old one that was used in PC version - it required too much of computing power.

I didn't say that this way would be easier. I would only want it to be that way because it has less chances to be immersion breaking than just tweaked AI stats (because for 'really easy mode' or 'really hard mode' you may have to change the stats beyond their lore borders).

If GM are able to make easy and hard campaigns in PnP Cyberpunk then it should be possible to make it also in its adaptation.

 

[dragonbird]

AnDr01d - as others have said, if the rationale for having only one difficulty level is to save costs, your proposed solution doesn't really address that.

Nor, I think, does it meet the needs of those saying that they want a difficulty level that is less challenging than the hardest possible. There's a big difference between "not wanting to play on insane" and "wanting guided play", and I don't think that anybody who has posted here criticising Costin's comments would be happy with guided play as the only alternative, so you'd be pleasing no-one.

DA:O has a tactics pseudo-scripting system, and has mods that allow you both to extend the scripting and to automate the controlled character as well as the companions. I've spent a LOT of time setting up tactics so that I could reach the point where I could sit back and watch during combat. The challenge was in designing the tactics, and seeing them work. If the game hadn't just played my tactics, but had designed them as well, I would have had absolutely no interest in it, and I would definitely not have considered doing such a thing on a first playthrough.

 

[andr01d]

Sure, if you take my suggestions to the extreme they wouldn't work. But I never suggested that CDPR should (and I would hope they would have the common sense not to).

But it's the extremes that the suggestions would have to hold up to because else you'll have to do such things like TW2 insane, where the only added difficulty stems from not being able to correct mistakes that led to your death. (This is what keeps me from playing insane, I'd like to have another chance when making a bad mistake without having to replay the whole game.)

When it comes to reaction times they simply need to find a reasonable number that works (say in the range of 25%-50% slower reactions). Instead of having an attacker react in 1 second he'll now take 1.5 seconds. This might not seem like a lot but gameplay-wise this would be enough to give people without as much twitch reflexes a chance to react.

Yeah, sure it has to be in certain bounds. But this makes it only part of the difficulty solution. As just this little more reaction time alone will never make the game really easier or harder.

Your logic behind modifying enemies damage modifiers is also flawed. Of course it would still take into account your armor and all your other stats. You simply decrease all damage by a percentage before you calculate any stat modifiers (like armor, potions, etc.). So if an enemy would normally do 25 base damage, now he hits you for 20 instead.

All right so the basic damage is lowered by x% by your armor and stats. But then it's raised at the same time by y% by the raised difficulty multiplier. So in case of x=y you'll just have the same damage with better armor due to more damaging enemies.

You also criticize my idea behind having different (key word being different) enemies using better AI late in the game to keep the game challenging. A nekker would always use the nekker AI, no matter when you encountered one. Meaning that yes, if you encountered a nekker late in the game you'd have no problem killing one (which gives you a sense of actually becoming more powerful). But this doesn't mean they can't have other types of enemies later in the game that use more advanced AI.

So for a smoother difficulty curve you'll have to have many different enemies in the game (much coding and modeling to do) and additionally have to explain why this enemy appears there suddenly and weaker enemies vanish (or the areas get as overcrowded with some strong and many weak but annoying enemies like the swamps in TW1).
It can be done, but it has to be handled with intensive care as to not break immersion by just switching/removing weaker enemies.

You can twist my words and take everything to the extreme to try and make it look silly. If I would have known this was your intention I would have given very specific examples to prevent this kind of nonsense. Anyways, I hope I made my suggestions more clear now.

It was never about making your ideas look silly, it's just that I always like to discuss almost all possibilities (and that includes possible problems and how to solve them) of any idea.
Good for you, if you have specific solutions for specific problems but unless you list those I am not willing to assume you got a solution to the problem as I have to believe that you can't have a solution for every problem.
This is about some asking the game to be made really hard while others demand the game to be nearly as easy as watching a movie. That's why we'll have to check if the extremes are a viable option because we need them to meet the expectations.

It's impossible to add random factor to every part of AI's code. If it would be then every game developer would do so - it would close a gap between human and AI. But it's impossible because of amount of work and testing that it would require. Even Crytek known for their devotion towards programming realistic AI didn't do it and yet their AI program was so complex, that when they were making Crysis 1 for consoles they had to program new AI, because consoles CPU were unable to handle old one that was used in PC version - it required to much of computing power.

If GM are able to make easy and hard campaigns in PnP Cyberpunk then it should be possible to make it also in its adaptation.

All right so no completely random AI actions. I think the AI doesn't need to behave 100% like a human but it would still leave room for improvement from the current state of the art.
And I still don't see the problem in implementing rules for AI facing AI since both just have a certain set of possible options, thus limiting the code needed.—In fact you'd need those unlimited possibilities only for AI facing humans since you can never think of all possible actions a human would take...

AnDr01d - as others have said, if the rationale for having only one difficulty level is to save costs, your proposed solution doesn't really address that.

Nor, I think, does it meet the needs of those saying that they want a difficulty level that is less challenging than the hardest possible. There's a big difference between "not wanting to play on insane" and "wanting guided play", and I don't think that anybody who has posted here criticising Costin's comments would be happy with guided play as the only alternative, so you'd be pleasing no-one.

DA:O has a tactics pseudo-scripting system, and has mods that allow you both to extend the scripting and to automate the controlled character as well as the companions. I've spent a LOT of time setting up tactics so that I could reach the point where I could sit back and watch during combat. The challenge was in designing the tactics, and seeing them work. If the game hadn't just played my tactics, but had designed them as well, I would have had absolutely no interest in it, and I would definitely not have considered doing such a thing on a first playthrough.

OK, so it's not the all-in-one solution I thought it to be in the first place.
But expanding the idea this system would allow for a broader range of difficulties. If you combine the standard way of changing difficulties via AI modifiers with my idea you'd have let's say 3 difficulty levels with the unguided play and again 3 different levels with the guided system. That should make it possible to satisfy almost everyone, wouldn't you say?
You see, I was just searching for a way to make games insanely hard for those that enjoy this and still keeping an option for making it ridiculously easy for those that merely seek entertainment.

 

[Aver]

Android, would you like to play a game where AI take a control over your character in every action section of the game? Because so far nobody said that they would want something like this and it's hard to imagine for me why would anyone want this unless he is partially disabled.

It's boring to watch when someone else silently play a game, especially when it is not even human. It would be a real waste of resources to make a such big feature, even tho almost no one would use it.

 

[andr01d]

Android, would you like to play a game where every AI take over control over your character in every action section of the game? Because so far nobody said that they would want something like this and it's hard to imagine for me why would anyone want this unless he is partially disabled. It would be a real waste of resources to make a such big feature, even tho almost no one would use it.

I don't understand...
Wasn't the fighting system in DA or TW1 similar? You didn't need real skill to win the fight but mostly the right stats and clever tactics.
You demand your character to do a certain action and it executes this completely guided by AI (or a random factor).
And if you do nothing your character would do all that is possible to not die.—That way you'd be able to combine TW2 more action-based fighting system with the older system from TW1 for those that prefer it this way.
Needless to say that the older system would (in most cases) require less skill to keep you alive in a fight than the real time action in the higher difficulty settings.
Would you want to play that?

And still the option of a fully guided cinematic mode remains, where the AI completely takes over for those that merely want to follow the story.

That's my idea. Everyone can get what they want...

Thought I can already see, that this would be nearly impossible to come true due to being too complicated to code. Ah, but one's allowed to dream and imagine... sigh

 

[freakie1one]

But this makes it only part of the difficulty solution. As just this little more reaction time alone will never make the game really easier or harder.

So you don't think having enough time to react, aka enough time to either dodge, move, counterattack or whatever other action you might choose, would make a huge difference in combat? I beg to differ. This is basically what makes a game "twitch-based": how long you have to react to what is occurring in the game. This mechanic is already in place in TW2. Opponents don't immediately start attacking you again after finishing an attack. There is usually a pause between attacks. Making the game less "twitchy" would make it considerably easier. You'd be able to land and dodge more attacks.

All right so the basic damage is lowered by x% by your armor and stats. But then it's raised at the same time by y% by the raised difficulty multiplier. So in case of x=y you'll just have the same damage with better armor due to more damaging enemies.

Again, I'm not sure how you are confused about how reducing the percentage of damage that all enemies do would equal easier combat. It's pretty straightforward. Here's an example:

Easy mode: Nekker hits for 20 base damage + -5 damage (armor bonus) + -2 (stat bonus) = Nekker hit you for 13 total damage.

Hard mode: Nekker hits for 25 base damage + -5 damage (armor bonus) + -2 (stat bonus) = Nekker hit you for 18 total damage.

End result: on easy mode all monsters do less damage which results in you being able to absorb more hits before dying.

So for a smoother difficulty curve you'll have to have many different enemies in the game (much coding and modeling to do) and additionally have to explain why this enemy appears there suddenly and weaker enemies vanish (or the areas get as overcrowded with some strong and many weak but annoying enemies like the swamps in TW1).It can be done, but it has to be handled with intensive care as to not break immersion by just switching/removing weaker enemies.

Populating different areas of games with more difficult monsters is done successfully in many games. This isn't exactly a new concept. It's especially easy to do in The Witcher series since the games are divided into chapters. This makes it easy to populate each chapter with new monsters since each chapter takes place in a new location.

 

[Randomdrowner2015]

A recap of the main arguments? Someone? Please. I am to lazy to plough through 7 pages of back and fourth.

I dont see any problem with a single difficulty as long as it reflects the story (if it is a story driven game).
As long as one can play it differently (if it contains RPG-elements). But I see no problems with different difficulty levels either, on the contrary, I think it is good that people of different skills are able to play and appreciate the game.

 

[andr01d]

So you don't think having enough time to react, aka enough time to either dodge, move, counterattack or whatever other action you might choose, would make a huge difference in combat? I beg to differ. This is basically what makes a game "twitch-based": how long you have to react to what is occurring in the game. This mechanic is already in place in TW2. Opponents don't immediately start attacking you again after finishing an attack. There is usually a pause between attacks. Making the game less "twitchy" would make it considerably easier. You'd be able to land and dodge more attacks.

I never said it wouldn't make the game easier (or harder) but I said that this difficulty gained by just being more twitchy is not enough to make the game hard enough for the die-hards. Especially since insanely fast enemy reactions can get pretty frustrating.
That's what I wanted to say: This modifier alone would not be enough. (Though it may be a good measure when combined with other modifiers.)

Again, I'm not sure how you are confused about how reducing the percentage of damage that all enemies do would equal easier combat. It's pretty straightforward. Here's an example:

Easy mode: Nekker hits for 20 base damage + -5 damage (armor bonus) + -2 (stat bonus) = Nekker hit you for 13 total damage.

Hard mode: Nekker hits for 25 base damage + -5 damage (armor bonus) + -2 (stat bonus) = Nekker hit you for 18 total damage.

End result: on easy mode all monsters do less damage which results in you being able to absorb more hits before dying.

Ah, so we misunderstood each other.
For me it was more like:
Early game Nekker hits for 20 base, early game armor protects for 5
Late game Nekker hits for 20 base, late game armor protects for 10, but late game difficulty modifier adds 5 to Nekker damage

I thought you meant it as a means to scale enemies throughout the game.

Populating different areas of games with more difficult monsters is done successfully in many games. This isn't exactly a new concept. It's especially easy to do in The Witcher series since the games are divided into chapters. This makes it easy to populate each chapter with new monsters since each chapter takes place in a new location.

Yeah, but we still have the problem with common enemies i.e. enemies that should be found everywhere. What would you do with them?
Make 'em stronger or smarter?—You'll have to explain that and it might conflict with lore.
Make 'em more numerous?—You should explain that and it might get a little annoying (TW1 swamps).
Leave 'em be but add stronger enemies at their side?—Might be the best way to go, but might also be annoying.

 

[dragonbird]

I don't understand...
Wasn't the fighting system in DA or TW1 similar? You didn't need real skill to win the fight but mostly the right stats and clever tactics.

I wouldn't put TW1 in that category, but yes for DA, and for traditional turn-based RPGs, but the one thing that they all have in common is that it's the PLAYER who decides the tactics, not the AI. So you still have a game of skill, but it's tactical skill rather than reflexes and co-ordination. (And of course, there are plenty of games that require both tactical skills and good reflexes).
A guided system removes the skill element, and leaves you with a game of chance. Dice-throwing.

 

[Randomdrowner2015]

I wouldn't put TW1 in that category, but yes for DA, and for traditional turn-based RPGs, but the one thing that they all have in common is that it's the PLAYER who decides the tactics, not the AI. So you still have a game of skill, but it's tactical skill rather than reflexes and co-ordination. (And of course, there are plenty of games that require both tactical skills and good reflexes).
A guided system removes the skill element, and leaves you with a game of chance. Dice-throwing.

I do not think a mandatory pure "guided" system has much of a place in modern computer RPG. It seems to me any RPG must integrate the skill of the player and the skill of the character into one functional and mutually supporting whole (the same goes for gameplay and lore of course). It makes no sense to, for example, "miss" a swing if you are facing a large and immobile target (unless your character trips over or something), but your character might not be that good of a fighter so he/she might not inflict much damage due to not being skilled, fast or strong enough.

 

[andr01d]

I wouldn't put TW1 in that category, but yes for DA, and for traditional turn-based RPGs, but the one thing that they all have in common is that it's the PLAYER who decides the tactics, not the AI. So you still have a game of skill, but it's tactical skill rather than reflexes and co-ordination. (And of course, there are plenty of games that require both tactical skills and good reflexes).
A guided system removes the skill element, and leaves you with a game of chance. Dice-throwing.

So why not offer both systems?
Turn-based CRPG (mild AI so to speak) for those who'd still want the game to require tactical skill.
The AI for this would be enemies AI modified to follow the players inputs (defends itself but only attacks when told to).
Guided mode (full AI takeover) with only (mildly) luck based gameplay for those that want a cinematic mode.
The AI for that should already be in the game in the form of enemies.
(Bear in mind that this will only be for the fights. All dialog and attribute choices could remain with the player.)

And on top of that you offer heavily skill based gameplay with real time combat like in TW2 with just the smarter AI that is used for the other two modes, too.

This system would even be fair in an MMO, since fighting the fully guided player is no different than fighting other AI and fighting mildly guided players is like fighting somewhat smarter, more flexible AI.

 

[freakie1one]

I never said it wouldn't make the game easier (or harder) but I said that this difficulty gained by just being more twitchy is not enough to make the game hard enough for the die-hards. Especially since insanely fast enemy reactions can get pretty frustrating.That's what I wanted to say: This modifier alone would not be enough. (Though it may be a good measure when combined with other modifiers.)

Yes, I agree that this by itself wouldn't be sufficient, which is why I also suggested the second part about changing enemies damage modifiers based upon which difficulty was selected. Combine the two ideas and I think you'd have acceptable difficulty levels for those who want a challenge and also for those who still want to partake in combat but not be frustrated by constantly dying and reloading. Very few people want a game that treats them like a complete idiot and these two methods wouldn't insult anyone's intelligence.

Yeah, but we still have the problem with common enemies i.e. enemies that should be found everywhere. What would you do with them?Make 'em stronger or smarter?—You'll have to explain that and it might conflict with lore.Make 'em more numerous?—You should explain that and it might get a little annoying (TW1 swamps).Leave 'em be but add stronger enemies at their side?—Might be the best way to go, but might also be annoying.

I don't think this would be a problem. Like I said previously, it would be a good thing to have some weaker enemies at later stages of the game. It makes you feel like you're actually stronger when you completely dominate something that gave you troubles at the start. Just make the weaker opponents not as common or add an element of fear into the game. Creatures that would once attack you on sight now fear you and run away. Or like you said, you could also mix them in with stronger opponents. Even if they only distracted you for a short while it might be enough time for a stronger opponent to land a hit on you.

My main argument is that the problem with most games is the lack of diverse AI. Instead of having enemies that do 2-3 different attacks I want enemies that can do 5-6 attacks and bosses that can do even more. Sure, when you start out the game it makes sense to fight enemies with limited abilities since you also have limited abilities. But as you progress in the game (and unlock more skills/feats) the enemies you face should also have more skills/feats. This is why TW2 was so easy when you started unlocking feats... you become more powerful while your opponents are still limited.

 

[manuelexar]

Personally I play for the story, I really don't care about challenges, I really prefer puzzles over challenge in combat.

 

[andr01d]

[...]
My main argument is that the problem with most games is the lack of diverse AI. Instead of having enemies that do 2-3 different attacks I want enemies that can do 5-6 attacks and bosses that can do even more. Sure, when you start out the game it makes sense to fight enemies with limited abilities since you also have limited abilities. But as you progress in the game (and unlock more skills/feats) the enemies you face should also have more skills/feats. This is why TW2 was so easy when you started unlocking feats... you become more powerful while your opponents are still limited.

Now I can fully agree with your ideas.
And it's also why I've already implemented this idea of a better and smarter AI into my idea.

Maybe this problem of the late game being to easy could be changed by not getting/unlocking better/stronger powers and more protective gear. Instead you could simply offer more diversity.
Let's take an imagined TW3 as an example. Since you've regained all your powerfull abilities over the last two games

Spoiler

by finally having beaten your amnesia

you can now from the beginning of the game on do everything quite good.
So instead of getting even better the game offers you to specialize in certain abilities (which makes you less proficient in other fields). That way the overall strength of each of the enemies needn't rise with the progression of the game as you don't get more powerful but rather more specialized in certain fields (you adapted your in-game skills to your preferred playstyle).
That would even allow for an easier implementation of a hard mode, by offering specializations that heavily restrict your other powers/abilities. A free to choose hard mode.

 

[Randomdrowner2015]

I agree that some enemies who once was a challenge logically becomes less challenging IF the character is getting stronger. This is not the case in games such as W2 though, when its simply a function of conventional gameplay(Geralt should not really start out as a newbie witcher with newbie gear, there are narrative functions that can balance this - such as the Dragon flying away with the legendary silver sword, the Legendary steel sword and armor is confiscated and received much later in a quest to recover the gear etc)

In game such as W2, a more effective route would be to introduce Geralt to situations where he (besides making short work of regular threats,) encounters more dangerous and numerous foes confronting him (large swarms of Nekkers instead of a handful, a group of elite soldiers with more efficient and faster combat moves instead of regular bandit grunts etc).

In games with some kind of "reputation system", this could manifest or be combined with the protagonist attracting the attention of dangerous people as he/she grows in power.

 

[Randomdrowner2015]

Personally I play for the story, I really don't care about challenges, I really prefer puzzles over challenge in combat.

This is all fine and well. A good RPG should give a character the option to prepare for certain difficult battles (making them easier)and possibly bypass them completely through stealth, wit and other form of strategy or manipulation. There is also a difference between games where you play an extremely skilled person (like Geralt) or a simple soldier raised on a farm trying to take on the world. The later logically would offer more difficult battles no doubt.

 

[dragonbird]

So why not offer both systems?

I would guess that the reason that nobody would offer a guided system is because nobody wants it. It's reducing the product to interactive entertainment, and I don't think that really counts as a "game" any more, if you remove the skill element. (And please remember that the use of tactics and turn-based RPGs are still games of skill. "Skill" is not a term that's restricted to real-time combat).

A choice of turn-based and real-time combat within the same game? That's been done often enough already and isn't a novelty. This may be where your DA/TW1 references came from, as they both allow you to pause or play in real-time. But somehow it never seems to result in a combat gameplay that satisfies anybody. Personally, I like turn-based on games where you control multiple characters, such as DA, but generally, if it's going to be there, it still has to be the player who sets up the tactics combo, not the AI, so that means a lot of extra development work. Costin's points on the development effort would then kick in.

I agree with your other points that balancing is the key factor. There's obviously a lot of room for improvement on this in a lot of games, and I think that this would be a much better way for developers to manage the needs of players with different gaming styles.

This is all fine and well. A good RPG should give a character the option to prepare for certain difficult battles (making them easier)and possibly bypass them completely through stealth, wit and other form of strategy or manipulation. There is also a difference between games where you play an extremely skilled person (like Geralt) or a simple soldier raised on a farm trying to take on the world. The later logically would offer more difficult battles no doubt.

^This.
I find it slightly ironic that we've been getting a bunch of Action games giving us the choice on lethal vs non-lethal playthroughs, whereas so many Story-based games still force combat at most milestones.

 

[freakie1one]

I find it slightly ironic that we've been getting a bunch of Action games giving us the choice on lethal vs non-lethal playthroughs, whereas so many Story-based games still force combat at most milestones.

Well, I would find it more ironic if you were to say that action games were giving you more options to role play than an open RPG. But having combat forced in certain areas in a story-driven game does make sense when the main point is for the devs to tell a story. If that story happens to include combat then so be it.