Game News Project Eternity Kickstarter Update #2: Stretch Goals Revealed

[Orchestra of Wolves]

Crooked Bee said: ↑

http://www.reddit.com/r/projecteter...ama_long_time_redditor_and_software_engineer/​

As far as enemy AI goes, in top down RPGs like this, the enemies don't typically need to be very sophisticated. The challenge usually comes from their stats (how hard they hit, how many hit points they have, etc) and what types of attacks they use.​

Straining... to... hope... for... best... (maybe he means in terms of the coding since I know nothing about coding? Please?)

 

[Infinitron]

Oh come on you guys. You honestly think an Obsidian code monkey means to go on Reddit and tell people their game's AI is going to be bad?

 

[Zed]

He's right. The AI doesn't have to be very sophisticated.

It only requires simple logic like:
Target prio - Ranged Low HP (if possible with ranged) > Melee Low HP > Melee High HP > Ranged High HP
If low on HP - Flee if opponent is high HP, Flee if outnumbered, Else fight 'til death.
Etc.

The challenge for the player is when the boss uses "summon multiheaded dicks" or "cone-shaped flame breath of fuck you", etc.

It worked in IE games and will work in any other.

I would love to see some advanced encounter AI though. Stuff like enemies using proper formations and shit.

 

[Excidium]

He's wrong. You people are just used to mediocrity.

It worked in IE games and will work in any other.

See what I mean?

 

[Zed]

He's wrong. You people are just used to mediocrity.

It worked in IE games and will work in any other.

See what I mean?

"work in any other" was not what I wanted to say, I should have written "any similar game". It really is good enough. I'm not saying it can't be better though.

 

[almondblight]

A good KickStarter isn't as much about math and pure numbers, as it is about creating enthusiasm about the project and not letting people the time to think things through while keeping the hype level high, both Schafer and Fargo managed this well with their "rebirth of Adventures/RPGs" events. Obsidian probably did the start with the countdown and speculation (what was it, like a news every day on the good ol' Kodex?) the best of all of them, but again instead of keeping that momentum and people engaged they ruined it with a big wet fart shortly after they reached their goal, despite having about double the amount of backers that Wasteland 2 had and a lot more money in the first two, they've managed to drop below that on the third day.

This Kickstarter took in more and had more backers on it's fourth day than both Wasteland 2 and Double Fine. And keep in mind that Double Fine's started later, so it's fourth day take has much more of the initial rush than Project Eternity did.

According to you, the project that is taking in more money and has more backers at a later date is the one that failed to keep the momentum, and the ones that took in less are the ones that kept it? What kind of logic is that?

And as for everybody saying that this would have made more if the goal was 2.2 million - just the other day everyone was talking about how their is a pledge rush after the goal is hit (because people see the product will actually come out, and want to jump on board). Now everyone's saying that pledges drop when the goal is hit?

I voiced my issues with this Kickstarter early on, but to say that one of the most successful videogame Kickstarters ever is a screwup because it follows the trajectory of every other Kickstarter ever is incredibly dense.

 

[Kz3r0]

This Kickstarter took in more and had more backers on it's fourth day than both Wasteland 2 and Double Fine. And keep in mind that Double Fine's started later, so it's fourth day take has much more of the initial rush than Project Eternity did.

I voiced my issues with this Kickstarter early on, but to say that one of the most successful videogame Kickstarters ever is a screwup because it follows the trajectory of every other Kickstarter ever is incredibly dense.

You are forgetting that Obsidian is a mainstream developer that thanks to Fallout and KOTOR is widely known, getting this kind of money shouldn't be surprising, it's not that they are doing bad, they could have done better.
Hopefully they will properly prepare for the final rush, that usually rakes in as much as the first days.

 

[almondblight]

You are forgetting that Obsidian is a mainstream developer that thanks to Fallout and KOTOR is widely known, getting this kind of money shouldn't be surprising, it's not that they are doing bad, they could have done better.
Hopefully they will properly prepare for the final rush, that usually rakes in as much as the first days.

And:

- Tim Schafer is legendary and the promise of of an old adventure game is something fans have wanted for years, and has developed well received games like Psychonauts, not to mention classics like Full Throttle and Secret of Monkey Island.
- Brian Fargo is famous for a plethora of RPG classics (including Fallout), and has a mainstream development studio.
- The Kickstarter hype has died down, and there isn't the excitement there was before (think there'll be another game like The Banner Saga getting 700k+?).
- We aren't starving for classic RPG's the way we were a year ago (or the way adventure gamers were for a mainstream old school adventure game). There's not only been a number announced recently, but many of them were also Kickstarter projects that people have already pledged for.
- There's been mixed reaction to a lot of Obsidian's games.

I'm not saying that Obsidian doesn't have a lot going for them, but you can just as easily make the case that Obsidian should be doing worse than Fargo and Schafer. They've done very well so far, as good as the best game Kickstarters. I don't see any indication of any big mistakes, even if I personally am not happy with all of the things that they've done.

 

[Excidium]

- Brian Fargo is famous for a plethora of RPG classics (including Fallout), and has a mainstream development studio.

Are you serious? Games like Fallout, which were barely relevant to the community at large before Fallout 3 are more often associated with people like MCA and Tim Cain. I'm p. sure most people didn't even know who Brian Fargo was before Wasteland 2.

 

[Esquilax]

You are forgetting that Obsidian is a mainstream developer that thanks to Fallout and KOTOR is widely known, getting this kind of money shouldn't be surprising, it's not that they are doing bad, they could have done better.
Hopefully they will properly prepare for the final rush, that usually rakes in as much as the first days.

Absolutely, they need to organize things and present them better. That first day was fantastic, but they really dropped the ball once they hit the stretch goal (how the fuck did they not know it would hit the goal so quickly?!), I'm hoping that they really push and get people to dig deep. According to Kicktraq, the average pledged so far is $40 compared to Wasteland 2's $48. Even when you take into account that the discounted edition for Project Eternity went for $20, that's still bad because Wasteland was being sold for $15!

Really, with the amount of big names and resources that they're bringing to the table, they really should be bringing in the same amount of donations per pledger as Wasteland 2 at least.

Wouldn't hurt to try and attract some of the $5k and $10k pledgers as well. The "tavern design" thing at $5k was great, and it's no wonder that it sold out so quickly, but they need to do more shit like that.

Seriously, $5 million should be very doable given the tools and the presence that they have at their disposal. Anything less than that is a serious disappointment.

 

[almondblight]

Are you serious? Games like Fallout, which were barely relevant to the community at large before Fallout 3 are more often associated with people like MCA and Tim Cain. I'm p. sure most people didn't even know who Brian Fargo was before Wasteland 2.

Look at the Kickstarter comments and the Wasteland 2 boards. There are a ton of people who said that they knew nothing about Wasteland but pledged because they were hoping they were going to get a Fallout type game, and the news media said this was connected to Fallout.

Now, I don't doubt that a lot of people haven't heard of Fargo, but I don't think Tim Cain or Josh Sawyer are much better known. At least, I rarely see people saying, "Oh, Tim Cain! Brian WHO?"

 

[tuluse]

- Brian Fargo is famous for a plethora of RPG classics (including Fallout), and has a mainstream development studio.

Are you serious? Games like Fallout, which were barely relevant to the community at large before Fallout 3 are more often associated with people like MCA and Tim Cain. I'm p. sure most people didn't even know who Brian Fargo was before Wasteland 2.

The community at large aren't the ones donating to kickstarter. They're happy with their popamole. It's the fringes that are donating.

 

[Excidium]

- Brian Fargo is famous for a plethora of RPG classics (including Fallout), and has a mainstream development studio.

Are you serious? Games like Fallout, which were barely relevant to the community at large before Fallout 3 are more often associated with people like MCA and Tim Cain. I'm p. sure most people didn't even know who Brian Fargo was before Wasteland 2.

The community at large aren't the ones donating to kickstarter. They're happy with their popamole. It's the fringes that are donating.

Or so you believe.

 

[mondblut]

He's wrong.

Nope. A "sophisticated" AI would gang up together in thousands and buttrape you the moment the introduction sequence is over. Failing that, he would run away and never show up for another encounter.

Some people tend to forget that a lifetime of a single AI unit is one encounter (and normally a small fraction of it), and its only raison d'etre, to harm player in some way it can then get killed. Anything more complicated is a waste of time. It needs no self-preservation, and anything other than glass cannon/suicide bomber mentality is counterproductive to its real purpose. If these people desire an opposition that fights to survive and win, they are ought to play MMOs. Or, at best, strategy games where global AI toils to produce its units from the same pool of resources you do, and has goals identical to yours.

 

[J_C]

WTF with the J-C fanboy shit? Most of us want this game to do well, but the hiveresque shit posts from him lately are an embarrassment. Goddamn man, use your head a little more.

You say fanboy shit. Why? Because I don't speak stupid things like:
- this kickstarter is doing terrible (it is not)
- their fuckup will hurt cRPGs, the games, the industry as a whole (it won't)
- they would have sustained their original pledge rate if they don't fuck up (they wouldn't)
- they will fuck up a game, based on some informations (we don't know that).
- they would gather 10m dollars if they didn't fuck up (they wouldn't)

I only called out on the dumbfucks who are histerically bashing Obsidian because of a small mistake, or had impossible expectations, and they play the drama queen when those are not met. And yeah, you were among them.

 

[Mrowak]

He's right. The AI doesn't have to be very sophisticated.

It only requires simple logic like:
Target prio - Ranged Low HP (if possible with ranged) > Melee Low HP > Melee High HP > Ranged High HP
If low on HP - Flee if opponent is high HP, Flee if outnumbered, Else fight 'til death.
Etc.

That's why 7/10 games with this type of procedural "AI" is shit. All the challenge from the encounter is relageted to stats, which essentially play the game for you. The 3/10 are the very scripted ones with handcrafted fights and AI "redesigned" for every encounter - they start using "smart" tactics, like using special items or certain spells which disable your party. The problem here is how it is easy to break such an "AI".

The challenge for the player is when the boss uses "summon multiheaded dicks" or "cone-shaped flame breath of fuck you", etc.

So again - it's not the opponent playing smart with minimal resources - in order to become a challenge he has to have access to much larger resource pool than the party (in most games it's tonnes of mooks). Otherwise he doesn't stand a chance.

And so we land into all manner of problems - artificial level thresholds, the need to level-scale enemies, repetetive use of certain abilities (encouraged by the dumb AI - why to use anything other than what's always effective) rendering other ones useless, the consistency of the setting, etc. etc. Frankly I've got the impression that people at the 'Dex had problem with these issues more than with actual combat system in IE games when proclaiming "RTwP in BG2 sucks".

It worked in IE games and will work in any other.

And it phailed hard in NWN2 with all expansions. If they are going to have an AI like that they'd better invest their goldz into good encounter design. I'd really rather avoid having a clusterfuck like in their earlier games.

I would love to see some advanced encounter AI though. Stuff like enemies using proper formations and shit.

Ditto.

 

[Crooked Bee]

Some people tend to forget that a lifetime of a single AI unit is one encounter (and normally a small fraction of it), and its only raison d'etre, to harm player in some way it can then get killed. Anything more complicated is a waste of time. It needs no self-preservation, and anything other than glass cannon/suicide bomber mentality is counterproductive to its real purpose. If these people desire an opposition that fights to survive and win, they are ought to play MMOs. Or, at best, strategy games where global AI toils to produce its units from the same pool of resources you do, and has goals identical to yours.

Josh Sawyer seems to agree with you on this, as far as monster AI is concerned:

There's really a big difference between the needs of party members (or enemies built with party member-type stats, in the D&D sense) and monsters. In many cases, monsters can be extremely simple, not because they're sacks of HP, but because they present a distinct type of threat that is made more complex by allied creatures that present a different type of threat.

Party members are a typically a much different matter due to how varied they can be. I believe both caster types and non-caster types need to have active things to do and they also need to have good default behavior. In IE games, fighter-types effectively just needed to change position and target. Caster types needed to be micromanaged constantly because they were consuming limited resources every round.

 

[piydek]

In many cases, monsters can be extremely simple, not because they're sacks of HP, but because they present a distinct type of threat that is made more complex by allied creatures that present a different type of threat.

Yeah, to me this seems to exactly describe how things worked in IWD2. And they worked great IMO. It all comes down to intelligently mixing distinct monsters within a certain context. None of them is complex within themselves. But combining and context provide lots of possible diversity. It seems to me that's what good encounter design is about.

Also, i don't think I've ever really encountered good group AI and intelligent interactivity between enemies in any RPG. Group "intelligence" comes strictly from intelligently mixing distinct enemies in a group.

 

[Rivmusique]

Some mainstream media are hoping on a figure of 10M$ , it's a nice read, considering it's not a specialized magazine: http://techland.time.com/2012/09/18...ty-rpg-gets-funded-could-it-raise-10-million/

it linked to here.

 

[hoverdog]

villain of the story went full mainstream now.

 

[Mrowak]

Josh Sawyer seems to agree with you on this, as far as monster AI is concerned:

There's really a big difference between the needs of party members (or enemies built with party member-type stats, in the D&D sense) and monsters. In many cases, monsters can be extremely simple, not because they're sacks of HP, but because they present a distinct type of threat that is made more complex by allied creatures that present a different type of threat.

Party members are a typically a much different matter due to how varied they can be. I believe both caster types and non-caster types need to have active things to do and they also need to have good default behavior. In IE games, fighter-types effectively just needed to change position and target. Caster types needed to be micromanaged constantly because they were consuming limited resources every round.

Ehh, that's fine and all but generating mixed setups isn't everything that makes up an encounter. Sure it introduces variety, but do it wrong and you won't stray far from repetition - look at NWN2: SOZ and MotB.

Also, it is essentially just masking the problem - instead of creating one enemy type with a variety of options at his disposal, you create many enemy types with very limited options. So it consumes resources that have to be put into creating the new enemy type (e.g. models). I fondly remember all those mage battles in BG2 SCS2 where mages used spells and items at their disposal quite intelligently - they tapped their potential to the fullest -and it was mostly one enemy type. The reverse of that would be ToEE enemies, who despite having a wide range of abilities offered by the game mechanics could not take advantage of them, thus reducing their variety (even though there were many enemy types in that game).

However, I have to say I am fine with enemies presenting distinct type of threat - improving AI to be more adaptable is one of them; adapting the encounter itself is another.

 

[jewboy]

BG2 w SCS2 did have pretty good enemy AI. If you play some BG2 with the default scripting and then with SCS2 scripting it's easy to see the difference. Assuming you have at least halfway decent strategic combat, more intelligent enemies are more fun. Increasing their intelligence is also one of the best ways of increasing the game difficulty.

 

[Mrowak]

Project Eternity DRM-free version revealed!

It will be made available to all minimum level backers via GOG!! : x : x: x: x: x: x: x: x

 

[Jaesun]

EDIT: Fuck you Mrowak you beat me to it.

 

[Jarpie]

They need to get those updates out there in the newssites like Rock Paper Shotgun.