So, what, people are surprised at the "PVE is too hard" from the devs?

[willfarlow]

I've been reading a lot of stuff about how people seem shocked that BioWare basically said that they weren't going to add PVE to Starfighter because it would be "too hard".

 

Really? This is a surprise? What were they GOING to say, something closer to the truth, like "Our bosses have made it clear that any new content has to show at least a 300% return in subscription money on man-hours spent, or we're fired"?

 

Come on now...BioWare, ShmioWare...EA holds the purse strings, and EA's idea of cutting-edge game design is "Oh, hey, let's release a new 'Madden Football'...what are we up to now, like, 'Madden 2,347'? Oh, it'll be RADICALLY different than the LAST one...THIS one will have the new AWAY GAME HELMET DESIGNS and an improved MASCOT skin!"

 

I, for one, am kind of surprised that they even BOTHERED to do Galactic Starfighter instead of just adding on some "The Black Hole...PART 2!" crap.

 

So I'm grateful for what we actually have. Would I like to see PVE content added? Oh, sure, but don't get me started on what I'd LIKE to see. I'm just glad to have a ship that doesn't run on rails...

[psandak]

First let me just state from the outset: I am a player who mostly participates in PvE. I have one character that I do ground PvP with when I feel like it (that one character has a mix of Partisan to Oberan PvP gear). Since GSF launch I have at least done the daily quest and used up each ship's 2x reqs bonus almost every day. GSF has become part of my "routine" playing SWTOR.

 

My take on PvE GSF is this:

 

Let's assume for a moment that it does happen; some sort of flashpoint or operation is developed.

 

Now considering that ALL boss mechanics in PvE are scripted: all bosses do certain things at certain times, and it is up to the players to figure out what those things are and how to avoid/mitigate those effects and then execute those avoidance/mitigation tactics while doing enough damage to eventually kill that boss.

 

Now apply this common practice to PvE space combat: a handful of successful "strategies" become the norm. Everyone flies their starship in a certain pattern to maximize damage to the boss and minimize damage to themselves. That sounds awfully like "on rails" to me .

 

Given the above what is the real difference between the "on rails" space combat we've had since launch, and GSF PvE? The only difference I perceive is where the "rails" come from: the devs vs the players. And lets be completely honest here, when PvE content is created the devs have "strategies" in mind, it is only that they do not announce them and it is up to the players to figure them out; like any puzzle, it takes a while to solve it, but each time you "solve" the same puzzle it gets easier to solve.

 

The "too hard" part stems from BioWare realizing this and wanting to break the mold, but (at this time) is unable to do so.

 

"Our bosses have made it clear that any new content has to show at least a 300% return in subscription money on man-hours spent, or we're fired"?

 

18 months ago I would have agreed with this assessment 100%, but not today. If GSF is the SSSC project, they announced it about a year ago and have obviously been working on it since that time. Furthermore, EA has stated (albeit it eight months ago) that overall monthly revenue has doubled since F2P. And if I had to guess, the "early access" to GSF probably gave sub numbers a good bump, at least in the short term, and the Life Day and Gree events have probably caused a bump in overall population. My point is, at this time, the implied "restrictions" have been eased. EA was vainly hoping for a "WoW killer" and did not get it, but SWTOR is a solid MMO with a consistent/stable player base and the steady release of varied content will keep it that way; setting unrealistic goals like the one described above is the surest way to kill this game and the revenue stream.

[Helig]

To be specific, BW said that GSF combat AI worth using would be "too hard". It's not on-rails, so the AI would need to have a resemblance of being able to navigate in 3-d space without crashing into a ton of obstacles that litter the map, *and* be able to fight the players.

 

GSF is one of my favourite updates, so far, and I'm looking forward to what they add in February.

 

And I'm far from being the only one.

Edited December 22, 2013 by Helig

[BrianDavion]

yeah, 3d AI isnt exactly easy I imagine it'll come eventually but if it does I'd not expect anything much

[zechio]

yeah, 3d AI isnt exactly easy I imagine it'll come eventually but if it does I'd not expect anything much

 

I can't imagine AI ever coming around. Traditionally the calculations for AI hit your cpu the hardest. This is exactly whatMMOs do not have to spare. The servers already have to do lots of calculations, multiplied by all the people. For even a small subset of players to do content with an actual AI would require them to put FAR more power into their servers, costing a huge amount upfront, and multiplying their monthly expenses....for very minimal gain.

 

This is exactly why raid bosses and such are still scripted events, not true AI.

[Docmal]

I can't imagine AI ever coming around. Traditionally the calculations for AI hit your cpu the hardest. This is exactly whatMMOs do not have to spare. The servers already have to do lots of calculations, multiplied by all the people. For even a small subset of players to do content with an actual AI would require them to put FAR more power into their servers, costing a huge amount upfront, and multiplying their monthly expenses....for very minimal gain.

 

This is exactly why raid bosses and such are still scripted events, not true AI.

 

Why does everyone say 3d AI is so hard? It is no different than AI in any 3d game... Specifically, decent space 3d AI has been around since the Wing Commander series in the 90s. There are hundreds of console and pc games since then that have done acceptable space ai. It doesn't have to be anything that can compete 1v1 but a dogfighting AI isn't something that would take years to develop.

[zechio]

Why does everyone say 3d AI is so hard? It is no different than AI in any 3d game... Specifically, decent space 3d AI has been around since the Wing Commander series in the 90s. There are hundreds of console and pc games since then that have done acceptable space ai. It doesn't have to be anything that can compete 1v1 but a dogfighting AI isn't something that would take years to develop.

 

I didn't say anything about developing the AI, just that it would need more resources than they have to spare. Perhaps you meant to quote someone else? Also, 3d AI is a bit mkre advanced, it has a number more possibilities than normal, such as pitchup/down and rotate. Its very doable though...for smaller scale.

[Ramalina]

It's true, flight sim AIs have been done before, but sims are often built to run in a standalone single player mode, so for multiplayer all the server really does is play traffic cop for communications.

 

Implementing a good pilot ai in SWTOR on the other hand might be a lot harder.

 

Making boats is pretty well understood, that doesn't mean that making boats in a factory for building railway locomotives is going to be easy. To make boats, you want to be working in a boatbuilder's shop.

 

There's also the little matter of personnel and payroll. If you're making a sim you hire people with relevant experience and have them work on the sim full time. If you're maintaining a pre-existing MMORPG you're likely to be a bit short on time and expertise unless your MMO was so profitable that you have money to spare for new development.

Edited December 22, 2013 by Ramalina

[Docmal]

I didn't say anything about developing the AI, just that it would need more resources than they have to spare. Perhaps you meant to quote someone else? Also, 3d AI is a bit mkre advanced, it has a number more possibilities than normal, such as pitchup/down and rotate. Its very doable though...for smaller scale.

 

Yeah I meant to quote who you were quoting. My bad.

[Docmal]

It's true, flight sim AIs have been done before, but sims are often built to run in a standalone single player mode, so for multiplayer all the server really does is play traffic cop for communications.

 

Implementing a good pilot ai in SWTOR on the other hand might be a lot harder.

 

Making boats is pretty well understood, that doesn't mean that making boats in a factory for building railway locomotives is going to be easy. To make boats, you want to be working in a boatbuilder's shop.

 

There's also the little matter of personnel and payroll. If you're making a sim you hire people with relevant experience and have them work on the sim full time. If you're maintaining a pre-existing MMORPG you're likely to be a bit short on time and expertise unless your MMO was so profitable that you have money to spare for new development.

 

If I were a locomotive shop and I needed some boats I would look at purchasing some boats from someone who already made them That was the "gist" of what I was saying. It wouldn't be that hard to leverage something within EA that has the algorithms already. Then it is simply a job for an intern to port the code across platforms.

[Arkerus]

I've been reading a lot of stuff about how people seem shocked that BioWare basically said that they weren't going to add PVE to Starfighter because it would be "too hard".

 

Really? This is a surprise? What were they GOING to say, something closer to the truth, like "Our bosses have made it clear that any new content has to show at least a 300% return in subscription money on man-hours spent, or we're fired"?

 

Come on now...BioWare, ShmioWare...EA holds the purse strings, and EA's idea of cutting-edge game design is "Oh, hey, let's release a new 'Madden Football'...what are we up to now, like, 'Madden 2,347'? Oh, it'll be RADICALLY different than the LAST one...THIS one will have the new AWAY GAME HELMET DESIGNS and an improved MASCOT skin!"

 

I, for one, am kind of surprised that they even BOTHERED to do Galactic Starfighter instead of just adding on some "The Black Hole...PART 2!" crap.

 

So I'm grateful for what we actually have. Would I like to see PVE content added? Oh, sure, but don't get me started on what I'd LIKE to see. I'm just glad to have a ship that doesn't run on rails...

 

As soon as you finish your 3d space combat game let us know.

[BrianDavion]

If I were a locomotive shop and I needed some boats I would look at purchasing some boats from someone who already made them That was the "gist" of what I was saying. It wouldn't be that hard to leverage something within EA that has the algorithms already. Then it is simply a job for an intern to port the code across platforms.

 

I'd be willing to bet money that AI is specific to the game/engine.

[Gavin_Kelvar]

well I will say that I've never tried or had any interest in trying the ground game PvP of TOR so the first time I hit "launch" I wasn't sure what to expect with GSF. I used to play Star Wars spaces games all the time so I was also very skeptical that it would be done right in an MMO. While GSF isn't flawless and there are some things I think should've been left out (accuracy/evasion for example, I prefer X-Wing/Rogue Squadron where a hit lands if your aim is good, no RNG misses) but GSF is now part of my routine on my main and currently the focus of what I do when I log on to my main (dailies are now just for filling time between GSF rounds).

 

I'm not saying that I'm going to try ground game PvP but I will say I'm impressed and while I'd probably try GSF PvE I think unless it was mind blowing I'd probably stick with the PvP GSF.

[NemoSD]

It's true, flight sim AIs have been done before, but sims are often built to run in a standalone single player mode, so for multiplayer all the server really does is play traffic cop for communications.

 

Implementing a good pilot ai in SWTOR on the other hand might be a lot harder.

 

Making boats is pretty well understood, that doesn't mean that making boats in a factory for building railway locomotives is going to be easy. To make boats, you want to be working in a boatbuilder's shop.

 

There's also the little matter of personnel and payroll. If you're making a sim you hire people with relevant experience and have them work on the sim full time. If you're maintaining a pre-existing MMORPG you're likely to be a bit short on time and expertise unless your MMO was so profitable that you have money to spare for new development.

 

Also, most flight sims that have been published in the last decade with combat in them have been multiplayer only, or helplessly arcadey.

[Telos]

Then it is simply a job for an intern to port the code across platforms.

 

no. interns are not allowed to perform work. they are there to observe and learn. if a business is using interns to actually work, they are breaking federal internship guidelines and should be fined.

 

that being said... porting code from one game to another ain't exactly easy.

[SwiftDuck]

yeah, 3d AI isnt exactly easy I imagine it'll come eventually but if it does I'd not expect anything much

 

have you people ever played a flight combat game? lol! AI is not as hard to program as one would think. People have been doing it for over a decade now.

[Lymain]

These are the devs who took forever to implement automated copies to PTS and still haven't figured out cross-server queues. Nothing they can't do surprises me anymore...doesn't matter how long other games have been doing it.

[Zoom_VI]

have you people ever played a flight combat game? lol! AI is not as hard to program as one would think. People have been doing it for over a decade now.

 

LoL I have looked at encoding for flight AI, and that stuff is a nightmare. The *only* viable way to do it for swtor that would overload the servers would be to do the SWBF2 style, where the AI are not even AI but rather following invisible predrawn lines called "splines." Of course they are also derpingly easy to shootdown and hardly ever hit the player.

[Zoom_VI]

These are the devs who took forever to implement automated copies to PTS and still haven't figured out cross-server queues. Nothing they can't do surprises me anymore...doesn't matter how long other games have been doing it.

 

Point me to the full MMO games that have been doing challenging 3d dogfighting AI for a decade.

 

I'm waiting.

[Lymain]

Point me to the full MMO games that have been doing challenging 3d dogfighting AI for a decade.

 

I'm waiting.

 

What does that have to do with anything? I mean Jump to Lightspeed had decent space PvE over 9 years ago, but my point is that that has no bearing on whether or not Bioware will ever be able to do it.

[Zoom_VI]

no. interns are not allowed to perform work. they are there to observe and learn. if a business is using interns to actually work, they are breaking federal internship guidelines and should be fined.

 

that being said... porting code from one game to another ain't exactly easy.

 

Code going from one game to another is almost never ever done unless its to make a sequel game. For starters games ARE WRITTEN IN DIFFERENT CODE LANGUAGES. I have never seen the same two games use the same language EVER. They may use heavily modified versions of some base language but never the same.

 

SW:BF2 used a heavily modified LUA script

SW:EAW used a modified XML language

the jedi knight series used a completely different language that those.

 

And because of that you can't switch ai like that. First issue is a large part of game AI is based on how it interacts with the environment, each map is filled with tons of invisible AI markers tailor for the games needs. GSF is completely different from any other flight game and would not be compatible.

[Telos]

 

And because of that you can't switch ai like that. First issue is a large part of game AI is based on how it interacts with the environment, each map is filled with tons of invisible AI markers tailor for the games needs. GSF is completely different from any other flight game and would not be compatible.

 

ya, people just don't realize what games are. everything in the game has a name like obj_asteroid_01 and each object has rules assigned to it and each one of those rules has to be programmed by a guy who knows the game's base code etc.

 

like I said. ain't easy. really really really hard would be more accurate.

[Zoom_VI]

.

like I said. ain't easy. really really really hard would be more accurate.

 

More like its never been done.

I mean really just for licensing it would have to be from the same studio and company so they could only *if they could technically* use code from another game that was made by both EA&BW.

[Hitomo_x]

Play a round of PvE in star.conflict ... swtor really does not Need this, it has the rail shooter