If Project Eternity was turn-based...?

[Dream]

haha what a joke. Keep on making up crap. Professionals make mistakes all the time, just less than the average person. To pretend this doesn't exist or there aren't limitations to different people is some kind of joke.

 

The key point is that without any pressure they don't make mistakes; not that they never make mistakes. Turn based games take away the pressure of time which is why they are not inherently more challenging than real time games.

 

You guys can use the it's just bad A.I. excuse, but that's just your opinion. You can't prove that it can actually be done because it hasn't been done. For all we know it's simply a technical limitation at this point and my points still stand regardless. In the IE games, if I remember right, mobs just attacked the closest party member usually. If you moved out of range of that mob then it would give up chase and just attack the closest party member. No going specifically after your vulnerable weak characters. No way to close the distance when you were out of melee range.

 

 

The reason no one has done it yet is because a cRPG game in the style of the IE games hasn't been made in forever. Compare the AI of shooters from that era (Half-life, unreal, quake 2, etc.) to the AI of modern shooters; the difference is night and day. I suppose you could cite DA:O as being a modern version of the IE games, but Bioware, at least nowadays, isn't very good with AI (I'm looking at you Mass Effect). Also, TB games have plenty of their own abuses which are unique to that combat system, and which would not work in RT games.

Edited October 28, 2012 by Dream

[Sedrefilos]

It is not that i don't like turn-based combat but i prefer real-time/pause to rpgs. The battles should be faster and more realistic. I mostly like turn-based combat in strategy games.

Baldur's Gates, IWDs and DA: O made a great job in battles i believe, where in fallout 1 & 2 and TOEE where too slow and some times boring (clicking all the time just to get 1 hit out of 3 misses for example).

[BetrayTheWorld]

I'd have prefered turn bassed.

I liked ToEE and Fallout systems better than those in BG, IWD and whatever.

 

But no big deal either way. NWN and NWN2 were just fine already, so I'm good.

 

Pretty much my thoughts. For it's time, the combat system of fallout was awesome, and I really enjoyed it. I do prefer the turn-based system rather than RTwP, especially when the RTwP system is using behind the scenes turns anyhow. It would allow you to consider more elements in your tactical decision-making, such as movespeed of the character in question. But, as you say, nwn and nwn2 were good, so either way works. Turn-based is just better. :D

[Gatt9]

haha what a joke. Keep on making up crap. Professionals make mistakes all the time, just less than the average person. To pretend this doesn't exist or there aren't limitations to different people is some kind of joke.

 

The key point is that without any pressure they don't make mistakes; not that they never make mistakes. Turn based games take away the pressure of time which is why they are not inherently more challenging than real time games.

 

Two points to make...

 

First, the "Pressure of time" is largely illusionary. All you're doing is clicking on unit A, then click on enemy B. You're not under any pressure, so long as you performed that action, it's automated from there. In fact, most games have this so heavily automated that even without instruction a unit will engage an enemy within it's range or sphere of awareness. Honestly, you're more under pressure to keep the AI from getting you killed than you are from any time factor. In any given RT game, the biggest pressure is making sure the AI doesn't crap out on you and decide the shortest path is a 2 mile hike around the edge of the map, or that the automation algorithm doesn't crap out and leave your guy standing there.

 

Second, turn based games can easily be an order of magnitude more challenging than a RT one. In a RT game, the AI has to process it's decisions in fractions of a second, which means it must have small contrived choices.

 

"Attack the nearest enemy"

"Attack whoever hits hardest"

"Attack the mage"

 

TB games, OTOH, have significantly more time to process their decisions. Not only does the AI have more time to process it's decisions after you make your move, but you can even thread out the AI and have it process it's decisions while the Player is taking their move. Since decisions are permutations and evaluation, the AI in a TB game will *always* be stronger than a RT one, simply because the AI has more resources (CPU time).

 

There's a reason why no RTS has ever had AI that tries to use advanced tactics on you, it's because it doesn't have time to do the calculations, and some of the calculations involved are NP-problems (Shortest path is a common example).

 

You guys can use the it's just bad A.I. excuse, but that's just your opinion. You can't prove that it can actually be done because it hasn't been done. For all we know it's simply a technical limitation at this point and my points still stand regardless. In the IE games, if I remember right, mobs just attacked the closest party member usually. If you moved out of range of that mob then it would give up chase and just attack the closest party member. No going specifically after your vulnerable weak characters. No way to close the distance when you were out of melee range.

 

 

The reason no one has done it yet is because a cRPG game in the style of the IE games hasn't been made in forever. Compare the AI of shooters from that era (Half-life, unreal, quake 2, etc.) to the AI of modern shooters; the difference is night and day. I suppose you could cite DA:O as being a modern version of the IE games, but Bioware, at least nowadays, isn't very good with AI (I'm looking at you Mass Effect). Also, TB games have plenty of their own abuses which are unique to that combat system, and which would not work in RT games.

 

The reason no one has done it is because you're talking very small amounts of resources, very large permutations, and NP-level calculations. You're asking the AI to determine a strategy for the best plan of attack, by processing all of the variables in a couple hundred milliseconds at best.

 

"If I move here and my ally stays there, then I can do this, but what if I move here and my ally stays there, what's the chances that I'll expose myself to that enemy over there and can I survive it, or would it be better for me to move to my ally or..."

 

Every unit increases the complexity at an ~factorial rate at least (Worsened by each ability they might have), combined with all of the calculations involved in pathing. The time slices involved in RT gameplay makes it virtually impossible to do any significantly complex AI, and it gets worse with each successive AI controlled unit introduced.

 

That's why the AI in RT systems uses "Dumb" algorithms like predefined scripts, scripts with randomized behaviors (Attack the closest, attack the furtherest, attack whoever is attacking the most damaged AI), or just plain dumb "Attack whoever's in my range".

 

Also, please note, the AI in modern shooters is at least as dumb today as it was years ago. In fact, some AI's are so dumb that you can literally reload and have it behave exactly the same each time (Gears of War, Mass Effect). Other AI's just use very simple algorithms such as "I was shot, I should move". Halo is a great example, the critters will stand in plain sight until they get shot, then move to cover.

 

TBH, if the AI was any good at all in any modern shooter, in some game somewhere it would've figured out that all it has to do to kill you is sacrifice one unit to slow you down, fall back, and gather overwhelming numbers since any given level will have several dozen of them and 1 of you.

 

The problem with any, and every, RT system is Time.

[Valverrin]

If I wanted to play realtime action pressure etc etc I'd pick a shooter with plot. The best way to have real tactics in combat is to go turn-based and realtime pause is in no way a good workaround. Jagged Alliance 2 has the best combat system out there and if you put it into an RPG... man I can only dream about it.

[jivex5k]

The game is RTwP not turn based.

They said this from the get go.

[Bhaal21]

i honestly cant say i care which one it is i like both ways i do hope that if they do real time with pause we get a lot of options for weather or not the game pauses automatically when such and such happens just like in baldur's gate

[light487]

XCOM: Enemy Unknown did some great stuff with turn-based combat.. but personally I would be extremely and utterly disappointed in the game and the development team (and the community for allowing it) if this was anything but RTwP.

[PK htiw klaw eriF]

Where is the "It isn't turn-based, please stop creating these threads" option?

 

I don't mean to knock turn-based(I really do like it), but PE was advertised (not exactly the right word, I know) as being a Real Time with Pause game, just like the old IE titles. I think that it should simply be the best RTwP game it can be, and was happy to throw in $140 to make it happen.

[Dream]

snip

 

 

What? That may have been the most ridiculous thing I've read in this thread yet; surpassing even Grimlorn's "facts."

 

This isn't 1982, it's 30 years later. Computers are powerful enough that they can do massive calculations in infinitesimally small time frames. Expert level SC, Q3, D3, etc. bots play the game better than most living breathing people, and half the time they appear more human as well. The reason that RT games don't make use of perfect AI isn't because of some lack of computing power, but because it would make the games impossible.

Edited October 31, 2012 by Dream

[paasi]

ToEE style turn based was great. It gave possibility to perform more complex tactics without spoiling the momentum of the game.

Though it did get annoying when there were so many creatures that every time i opened a door combat started.

That opposed with the attacked, pause, engage of the BG. I really like both, so I vote for an in-game switch from where to enable/disable TBC.

[Nerei]

[long post cut down due to size end post size]

 

 

Is making an AI that makes an über stack of doom from units spread over the map hard? It is not, it is actually fairly easy. Any moderately decent AI programmer can do that.

It would not be hard to make the Starcraft 2 AI select all units, assemble them into one big stack and go hunt down the player. You could make it even simpler by just giving it omniscience and repeatedly attacking nearest player unit after making the stack.

However do they want to make an AI like that?

Would people want an AI like that?

What would that do for gameplay?

If the AI use cheap tricks like instant rushes you can be pretty sure that the players will complain. Even exploiting that type of AI would be somewhat simple in a fairly even scenario.

An AI like that is just bad. It is fairly irrelevant what type of game it is, be it Civilization or Starcraft.

 

The challenge for an AI programmer is to make an AI that is competitive without it effectively cheating. An AI that play like a human (generally without trying to use cheap tactics like the human probably would) and provides a challenge without having omniscience and/or X% more resources than a human. That is the goal, not an cheap unbeatable construct.

 

If we are going to claim that the AI in a TB game like say the original Xcom (or just about any TB game from the 90's) can do a decent job simply due to it having more time, then a modern AI can do just as well in real time simply due to better hardware. Many of the TB games from the early to mid 90's could run on a I486.

 

If the AI fails at handling basic tasks it is not about it not having the the calculation power but bad programming or a lack of time to make it. Compare the calculating power of a good I7 quad core to that of a 300 MHZ P II the latter easily being capable of running early RT games like Command & conquer (C&C can run on a 100 MHZ I486).

 

 

You forget one of the main problem of making a good AI for a TB game. The (generally) higher complexity it has to handle. This however is not just down to the amount of calculations but also a matter of programming workload.

If you have 100 tools at your disposal, you may very well be able to handle the majority if you are good, but odds are the AI will not (the programmers might not even predict all the scenarios). If the AI does not handle the additional options well you are likely to just hand the player more ways to easily beat it. That I would say makes the game a lot easier.

 

Many people bring up ToEE as a nice example of good TB combat (and on that I agree, it is a good system). However the standard AI is not all that great. It works okay for basic combat but not special situations. I remember effectively breaking it without even trying simply due to the things it could not properly handle. One of the improvements the Co8 mod includes is improved AI.

 

 

As for the pressure of time being illusionary, that highly depends on the game. If you can manage it largely with auto-attacks or basic AI, then the challenge would probably not be too hard no matter if it is TB or RT. That tactic is little more than overrunning your enemy with larger numbers or stronger forces.

If we stick with Starcraft 2 as example pulling a large number of special abilities targeted at different enemies in a few seconds very much can make a difference (especially if your opponent tries to do the same) and is an example of time pressure in RT. Alternatively just try fighting 3-4 close battles spread out across the map (for added challenge make the 3-4 attackers human players).

Edited October 31, 2012 by Nerei

[Starglider]

In a RT game, the AI has to process it's decisions in fractions of a second, which means it must have small contrived choices.

 

I have done extensive game AI programming and I can state confidently that this is not a significant limitation in 2D games. Back in the 1990s yes it was a struggle to get 100s of units in an RTS to act intelligently, but computing power is a complete non-issue for a few tens of actors on a modern multi-core multi-gigahertz machine. The problem is overwhelmingly algorithms, design and implementation; or test/dev man hours if you chose to try and solve the problem with a ton of unit and location specific scripting.

 

TB games, OTOH, have significantly more time to process their decisions.

 

The vast majority of real-time games including the classic infinity engine stuff do not use an 'all AI starts from scratch and completes on every single cycle' model. The basic engine executes simple actions set by both the player and the AI; the AI runs asynchronously and merely updates the current actions & targets from time to time the same way the human player does. A turn-based AI is still required to complete moves of individual units in a fraction of a second to avoid annoying players, that is not significantly more time than a real-time version (and with multi-core machines and all graphics done on the GPU, doing game and AI at the same time is not an issue).

 

Since decisions are permutations and evaluation, the AI in a TB game will *always* be stronger than a RT one, simply because the AI has more resources (CPU time).

 

Again CPU time is not an issue for any algorithm you'd implement in this kind of game. Pathing (for example) is trivial in the first place for these relatively small and simple maps and not any harder for real-time since the path is only regenerated when the destination is changed or the path is blocked, not on every frame. No one uses chess-style deep lookahead to billions of gamestates in RPGs.

 

There's a reason why no RTS has ever had AI that tries to use advanced tactics on you, it's because it doesn't have time to do the calculations, and some of the calculations involved are NP-problems (Shortest path is a common example).

 

No, it's because no one knows what calculations to do. Even full-time AI researchers with access to near-unlimited grid compute can't recreate human play ability in games as complex as Go, and the game state in Go is vastly simpler than the average RPG.

 

The reason no one has done it is because you're talking very small amounts of resources, very large permutations, and NP-level calculations. You're asking the AI to determine a strategy for the best plan of attack, by processing all of the variables in a couple hundred milliseconds at best.

 

A couple of hundred milliseconds is a relative eternity. Modern game engines draw hundreds of millions of polygons and shade several billion pixels in that time, as well as processing a couple dozen time steps on several thousand complex physics objects and a few thousand animated bones. Again, no one making these kind of games (turn based or not) uses algorithms that would lock up a modern PC for seconds. Maybe for Civilisation style games you might do that, but even there usually no, just heuristic analysis of a set of possible moves, no attempt to model player response.

 

The limiting factor for AI in Infinity Engine games was primarily the amount of developer time available, secondly the very slow scripting engine, thirdly the limited engine support and least importantly the fact it had to run on late 90s CPUs. The last three should all be a non-factor now so it is almost entirely down to how good the programmers are (at writing game AI) and how much time they have to spend on it.

[Continuum]

I love turn-based games in that I enjoy being able to stop and tactically plan out everything.

 

Pausable real-time like Infinity Engine games and Dragon Age is good, but how good is very dependent on the AI of the NPCs. Crappy AI dramatically impacts RPGs. Even the mediocre AI in Dragon Age was very problematic. You had to create all these rules and conditions to basically program each character, and if you wanted to go beyond a few simple rules you had to sacrifice skills in things like crafting and combat skills. Things like area of effect spells and traps become EXTREMELY problematic... either the game has to make your characters immune to them or you have to massively micromanage the NPCs to keep them out of the AoEs, effectively making things in to a half-baked turn based system of hitting the pause button. It's been a long time since I've played an IE game, but I remember this being a problem there as well.

 

Any time you have to juggle more than one or two characters in tactical combat, the system either requires intense micromanagement, an extremely complex AI, or increasing simplification of the combat systems. Good AI is a rarity. If the combat system is going to be simplistic, then why bother making a top-down, full-party RPG? And if I need to micromanage everything, I'd rather just have it be turn based.

 

There is a happy medium to be hit with pausable, tactical, real-time, it's just hard to hit. I'd prefer to see P:E error towards the turn-based side of that continuum. Nothing annoys me more than being dependent on AI companions that basically get in the way of my attempts at strategy.

[K1113R]

I definitely like the real time with pause system more, though of course the game wouldn't be worse if it was turn-based.

 

[Karkarov]

If it has a pause feature they can make it as real time as they want, I can still turn it into turned based by just pausing a lot. In other words... don't really care.

[Mikaw]

Well I voted "no" only because they're was no option saying " I'd support it, but I really prefer realtime with active pause"

[Lynx88]

Turn-based is nice, but I think a real-time with pause, or perhaps a simultaneous-turn based system would be preferable.

 

I really, really don't care for waiting for the enemy to take their turn 1 by 1, especially neutral/civilian NPCs like in Fallout.

[VixRaine]

I have to ask myself, "What is the point of a discussion about what type of combat system you would like better in a game that has an Officially announced and confirmed type of combat system? It seems an awful lot like watching a horse race, and then heading down and placing a bet on the horse that was in Last Place just because you're a masochist like that."

Edited November 2, 2012 by VixRaine