Hero Engine: why?

[VonAlex]

Actually you would. That's what companies do so if things break... It's not their fault.

 

Heh so they knew that it would fail and can blame some few persons and maybe make some spinoffs and make new money on them and betray millions of poor souls that just want their star wars fix

[ZedFighter]

If you know anything about the Austin team, you'll quickly realize why the game is in the shape it is. It's not just that the engine was a poor choice.

[Matte_Black]

Why did you choose an Engine that wasn't developed to your exact specifications? I mean, EA and BW make enough, you could have outsourced your Engine development and made the engine fit your game.

 

Instead, you chose the lazy way and are trying to make a game fit an engine. Not only that, you picked a terrible single-threaded engine that can't even process commands before animations finish, hence all of the ability delay rage you're seeing.

 

 

Seriously. Blizzard built an engine to fit their game, they're billions (trillions yet?) richer. You didn't, and you're missing out on Billions in part due to this decision.

Why?

 

At this point, does it matter?

[Adrenocrome]

I want to share an constructive outside personal opinion with Bioware and everyone else here. Just to be taken seriously and as someone with credentials, I am 3D artist (to not go in details) and have game industry experience. Ive been fan of SW (the original trilogy ) since I was a little kid and was really excited to play SWTOR. But from the opening, Ive noticed that there was something deeply wrong with this project and Im not talking about some bugs (which is expected at the openings of new MMOs)(You all know what am I thinking when Im talking about what is technically wrong here) . 3 things.

 

First and major thing is the game engine:

Game engine is 50% of entire game's performance (thats why its caled engine -something that unifies and drives all data together). I dont know why did anyone sain enough thought that Heroengine will be best choice for SWTOR and I dont know whos bright idea was to use this piece of c**p, but it turn out to be utter disaster. Maybe in theory Heroengine (and this is just my personal oppinion -Ive test it) sounds great, with all that cloud stuff and all content being made and manipulated at server in realtime, but it turned out that practical results were quite different in reality. And its obvious that BW had to rewrite like 50% of the engine just to be able work for SWTOR. Now, it is proven right, because I see that other MMOs which use this same game engine too, have same kind (and worse) of problems.

 

Bioware should have used custom made inhouse engine for SWTOR project. That way you'd know exactly what are you expecting and for what performance are you building it (and youll spend much less money). In any case franchise like Star Wars would deserved it. Also look at (I hate to say it) WoW. Blizzard made custom inhouse engine, thats why WoW have such a good performance (PvP battlegrounds have flawless performance) -because it is built exactly to suit the needs of that particular project. (should I have to mention 4fps graphics in WZs and Illum in SWTOR)?

 

Or at least you should have used some battle tested engine like Unreal engine 3 (which license you already posses --> Mass Effect etc.) (I dont even want to mention GW 2 in this case), or Gamebryo, which have tons of single player and MMO titles to prove its quality. Also, look at Mindark and their MMO project Entropia. They had a guts to say -ok this engine dont work for us anymore, lets use another one. Yes, they actually transferred entire game to the totally different game engine (from Gamebryo to Cryengine 2). And foremost, while they worked on transfer, the game was whole time live (on old engine), up to time they've finished it and just said -ok at this date well transfer game to a new engine and old one will no longer work. It took them a day or so to make rotation (also ION did the same thing). Yes read between the lines -I am suggesting actually to transfer whole SWTOR to new engine -it would be much less financially painfully than to keep this carcass alive.

 

 

 

STOP copying other MMO games:

Using some form of basic construction uniformity is ok (and using some good proven mechanics is ok too), but copying entire ideas and concepts just to apeal to wide population of players whos playing certain MMOs is just plain dumb. This is the place where marketing gimmick gone wrong. I dont know who told you that if you copy entire concepts of different MMOs (like WoW -yes I said it and you know it) will gather you x,xxx,xxx number of players. Stop thinking about $$$ for one second and do something right -do "your" ideas and concepts -it is hard to believe that BW dont have team capable of producing novelty concepts and ideas and pushing an envelope of MMO genre. And what its worst is that SW universe is so richfull now that you cant screw it up in that area. And whats funny is that, if BW made something unique and conceptually original (real StarWarsy) they would get much respect and admiration (that means subscriptions as well) in MMO community.

 

Dont do single player stuff in MMOs:

MMO is an social event and should be treated as such in building process. Ok to be fair, its alright to have that kind of content to some sertain point of degree like an personal story -but to overcrank it to become the main PVE thing, is ridiculous. Linear gaming is fun for the first time, but when you have 2,3 or more chars it became pretty boring in MMOs. Thats why we still have single player games. MMO genre is not suitable for linear gameplay. I really liked KOTOR series but if you wanted to put that much storyline in the project you should gone for KOTOR 3 the single player RPG and not KOTOR 3 the MMORPG.

 

[terminova]

I want to share a constructive outside personal opinion with Bioware and everyone else here. Just to be taken seriously and as someone with credentials, I am 3D artist (to not go in details) and have game industry experience. Ive been fan of SW (the original trilogy ) since I was a little kid and was really excited to play SWTOR. But from the opening, Ive noticed that there was something deeply wrong with this project and Im not talking about some bugs (which is expected at the openings of new MMOs)(You all know what am I thinking when I'm talking about what is technically wrong here) . 3 things.

 

If you are going to claim credentials, please list them.

 

First and major thing is the game engine:

Game engine is 50% of entire game's performance (that's why its called engine -something that unifies and drives all data together). I don't know why did anyone sane enough thought that Heroengine will be best choice for SWTOR and I dont know who's bright idea was to use this piece of c**p, but it turn out to be utter disaster. Maybe in theory Heroengine (and this is just my personal oppinion -Ive test it) sounds great, with all that cloud stuff and all content being made and manipulated at server in realtime, but it turned out that practical results were quite different in reality. And its obvious that BW had to rewrite like 50% of the engine just to be able work for SWTOR. Now, it is proven right, because I see that other MMOs which use this same game engine too, have same kind (and worse) of problems.

 

HeroCloud's Hero-engine != Bioware's Hero-engine. Maybe you're right and Bioware's modified Hero-engine is as crappy as the Free to use Herocloud Heroengine, and maybe it isn't. I think it's too early to tell if the performance issues can be resolved.

 

Tell you what, you go work with Bioware's Hero-engine and compare results with the Herocloud Heroengine, and tell us if there is any difference.

 

Bioware should have used custom made inhouse engine for SWTOR project. That way you'd know exactly what are you expecting and for what performance are you building it (and youll spend much less money). In any case franchise like Star Wars would deserved it. Also look at (I hate to say it) WoW. Blizzard made custom inhouse engine, thats why WoW have such a good performance (PvP battlegrounds have flawless performance) -because it is built exactly to suit the needs of that particular project. (should I have to mention 4fps graphics in WZs and Illum in SWTOR)?

 

And they've been working the bugs out of the engine for 7 years, and you have no real proof that developing an engine in house from the ground up would have actually made any difference.

 

Yes read between the lines -I am suggesting actually to transfer whole SWTOR to new engine -it would be much less financially painfully than to keep this carcass alive.

 

If you honestly had the experience in the industry you claim to have you would know that would take too much development time and money to do, or it would be more expensive financially, than less.

 

STOP copying other MMO games:

 

Only to have the players demand you do so.

 

Dont do single player stuff in MMOs:

 

That's impossible unless you want to drive away your player base.

[Maltra]

With money out of the picture they were either lazy or stupid, or both. There's really no other explanation.

[terminova]

With money out of the picture they were either lazy or stupid, or both. There's really no other explanation.

 

Or neither.

[tkshinoda]

Rift: Game Engine of Choice - Modified Gamebryo (Runs smooth even on toasters) - 8vs20 (28 people at screen same time no fps drops at all, no stuttering, no lag)

 

Swtor: Game Engine of Choice - Modified HeroEngine (Runs like crap on high end systems, for some it gives no problems) - Op vs Op it would seem (about 36ish people on screen, 1 - 5 fps, major stuttering, terrible low fps)

http://www.youtube.com/watch?v=6KUNaMUNnzQ

 

 

Tell me again why they choose HeroEngine and not Unreal3 or gamebryo for that matter they both run better then HE ever will, and they both have been developed for a longer time then HE has, that in return = better coding, better support, better tools.

Edited March 4, 2012 by tkshinoda

[Kubernetic]

Blizzard didn't build a new custom game engine from scratch for World of Warcraft.

 

They reused the existing Warcraft engine and heavily modified it.

 

___

 

The other thing I don't get is the constant mixing of two mutually exclusive ideas, as if they are rational to be shoved together.

 

1) BioWare used a third party engine, and the engine is flawed and cannot be repaired.

 

2) BioWare has had to heavily modify the engine and by some estimates has modified 50% or more of the engine to suit their needs for the game.

 

Folks, if they have the skills to modify the engine to that extent, they have the ability to fix most of its issues, whatever they might be. If it's a core memory issue, they have the ability to rewrite the code so that they can optimize it and change how it operates.

 

You cannot simultaneously say that they have thoroughly modified the engine, but that the engine is frozen and cannot be repaired. If they can modify it to the extent they did, and they have the licensing permission to do so, then they can fix whatever needs to be fixed.

 

But it will... take... time...

[Terko_Koslah]

Rift: Game Engine of Choice - Modified Gamebryo (Runs smooth even on toasters) - 8vs20 (28 people at screen same time no fps drops at all, no stuttering, no lag)

 

Swtor: Game Engine of Choice - Modified HeroEngine (Runs like crap on high end systems, for some it gives no problems) - Op vs Op it would seem (about 36ish people on screen, 1 - 5 fps, major stuttering, terrible low fps)

http://www.youtube.com/watch?v=6KUNaMUNnzQ

 

 

Tell me again why they choose HeroEngine and not Unreal3 or gamebryo for that matter they both run better then HE ever will.

 

That's awful and the cartoonish style doesn't even look good.

 

ps. They should have gone kotor style. SWtor is a big disappointment for me.

Edited March 4, 2012 by Terko_Koslah

[DarthKhaos]

Heh so they knew that it would fail and can blame some few persons and maybe make some spinoffs and make new money on them and betray millions of poor souls that just want their star wars fix

That is not what I said. I said IF. Come on companies are out sourcing phone tech support to another country. Why? It's cheaper than hiring and training their own people also if you got a problem you gotta take it up with them. We'll just send you to ask them anyway.

 

Server hosting, backups, any and everything these days for IT are being handled more and more by external parties. Why? Because it's cheaper and they can pass the buck IN-CASE something bad happens. If nothing bad happens it's a good investment.

[VonAlex]

Well of that I will agree with you... We can just hope that everybody does their job well and BW is capable of taking the work home and do it right if somebody outside the fence mess up

[hurriedcorgi]

Blizzard didn't build a new custom game engine from scratch for World of Warcraft.

 

They reused the existing Warcraft engine and heavily modified it.

 

___

 

The other thing I don't get is the constant mixing of two mutually exclusive ideas, as if they are rational to be shoved together.

 

1) BioWare used a third party engine, and the engine is flawed and cannot be repaired.

 

2) BioWare has had to heavily modify the engine and by some estimates has modified 50% or more of the engine to suit their needs for the game.

 

Folks, if they have the skills to modify the engine to that extent, they have the ability to fix most of its issues, whatever they might be. If it's a core memory issue, they have the ability to rewrite the code so that they can optimize it and change how it operates.

 

You cannot simultaneously say that they have thoroughly modified the engine, but that the engine is frozen and cannot be repaired. If they can modify it to the extent they did, and they have the licensing permission to do so, then they can fix whatever needs to be fixed.

 

But it will... take... time...

 

They had five years, no shortage of money, and plenty of industry standards to compare themselves to. I don't think it would be a stretch to argue that the only reason any one is really playing this game is because of the setting of the game.

[DarthKhaos]

They had five years, no shortage of money, and plenty of industry standards to compare themselves to. I don't think it would be a stretch to argue that the only reason any one is really playing this game is because of the setting of the game.

You really do not understand how programming works, how resourceful players are when it comes to finding exploits and how the nature of the system changes with more users.

 

Tell me this. You never written a document, read it over 5 times then someone else looks over your stuff and find, to them an obvious typo or error? When players were introduced into the equation things popped up that they never considered to test. It will take time to fix this because that error wasn't found out at the start of development. You also gotta remember other things are going on and ensuring any and all fixes don't severely break other things.

[Lyssarel]

I don't understand why this discussion, don't you all have problems with game performance?I am not an expert I am just a simple user,player and I can see what horrid things this game does to my computer lol and to my nerves especially in warzones, so from my point of view it must have to do something with the core of this bloody game...engine perhaps... Aside everything I don't care what has to do with it I just know that after all this waiting for this game to come out I am very disappointed that I cannot even play it right due to this issues...lag, horrid fps, bugs etc...And I truly try to play lol.

Arent we all just users that deserve a decent product?

Its exhausting to think about what went wrong...In perfect world our only job would be to buy and enjoy the game, and not to think what went wrong...there should be some committee that would ban this type of unfinished games to charge for subscriptions or even sell the damn thing on first place....this is still in beta version-I think its obvious that we are paying just so we could test it

All I can do is waste a bit more time here until some decent game comes out...something like guild wars2 :D

[owenthorn]

do remember this is the FIRST MAJOR GAME released on this engine....that says something to me....

[Kubernetic]

They had five years, no shortage of money, and plenty of industry standards to compare themselves to. I don't think it would be a stretch to argue that the only reason any one is really playing this game is because of the setting of the game.

 

They spent the early parts of those years BUILDING the game. You aren't going to find out any shortcomings in anything to a great extent when the worlds are first being built.

 

They spent the later parts of those years modifying what was there to suit their purposes to launch the game. Only when the beta periods begin are you going to get a sense for where the bottlenecks are, and only after that are you going to start seeing how the system can change to resolve those issues.

 

And even then, you have to figure out how to resolve the issues, and set those actions into motion while still not affecting everything else.

 

So no, they didn't have "five years" to fix it all. They've had three months since launch to fix it, and a few months before that to even know that people were having issues with it.

 

To say they had "no shortage of money" is just preposterous. I don't care how big your budgets are, you're still going to have to optimize your spend and limit what you can do within budgetary boundaries, especially when you're talking about licensing fees from George Lucas.

 

Granted, it was Star Wars content that pulled me into this game, but even if it had all of the actual canon Star Wars elements, I wouldn't keep playing a piece of garbage just because it threw images of C3-P0 or R2-D2 at me. I keep playing it because it's enjoyable and fun to play.

[Toweleeeie]

They spent the early parts of those years BUILDING the game. You aren't going to find out any shortcomings in anything to a great extent when the worlds are first being built.

 

They spent the later parts of those years modifying what was there to suit their purposes to launch the game. Only when the beta periods begin are you going to get a sense for where the bottlenecks are, and only after that are you going to start seeing how the system can change to resolve those issues.

 

And even then, you have to figure out how to resolve the issues, and set those actions into motion while still not affecting everything else.

 

So no, they didn't have "five years" to fix it all. They've had three months since launch to fix it, and a few months before that to even know that people were having issues with it.

 

To say they had "no shortage of money" is just preposterous. I don't care how big your budgets are, you're still going to have to optimize your spend and limit what you can do within budgetary boundaries, especially when you're talking about licensing fees from George Lucas.

 

Granted, it was Star Wars content that pulled me into this game, but even if it had all of the actual canon Star Wars elements, I wouldn't keep playing a piece of garbage just because it threw images of C3-P0 or R2-D2 at me. I keep playing it because it's enjoyable and fun to play.

 

I would argue that the theme of Star Wars is what makes it fun to play. I mean, I went and saw Episode 1 in 3D at the midnight showing and knew it would be trash. So why did I go? It was freaking Star Wars.

Edited March 5, 2012 by Toweleeeie

[DarthKhaos]

I would argue that the theme of Star Wars is what makes it fun to play. I mean, I went and saw Episode 1 in 3D at the midnight showing and knew it would be trash. So why did I go? It was freaking Star Wars.

I would disagree. There have been many Star Wars games released that were crap. Even the fans agree it was crap.

 

I think a lot of people have insane expectations of a new MMO. Also they may be a bit tired of WoW and anything with similarities. Which is amusing cause if it's missing what WoW has they also complain.

 

IMO this game has a solid foundation for a new MMO. Now they will add stuff to make it better. Leveling is refreshing and fun. I enjoy the various classes and PvPing. I like how the game looks. I do not need super realistic graphics, I appreciate the art style. Everyone is different though. The thing is no one lied to gamers as to what is going to be in the game. Hell a lot of players been claiming since beta they're going to cancel their preorder yet they didn't and took a multi month subscription. I don't get that.

 

Bottom line all new mmos will take time to develop. They will optimize both the client, server & engine, fix bugs and add content. No one can argue that they are not working on things. So chillax.

[Blackweb]

I would like to repost this answer my mgriffith it's really informative.

 

There is this lovely link floating around in several places that points to MrLizards' blog post (loosely) titled "Why didn't they fix this bug yet (and why am I an idiot for asking)?"

 

About 5 years ago, shortly after they secured funding, the development leads sat down and discussed the main game systems they wanted. Based on that they looked around for a game engine that they expected would be able to support those features they planned to include. This engine of necessity had to be in development (as existing engines would inevitably be dated in 5 years when the game would be released), and licensing terms and cost would be a major issue as well.

After that the developers would spend the next 5 years making modifications to the engine, to support the specific features they would need, the built the actual game on top of it and an entire database structure and custom networking code to work in tandom with the game engine. And they created hundreds of man-years worth of art assests (e.g. models, textures, animations, lip-syncing and so on).

 

And somewhere towards the end of beta, when everything was coming together they realised they overestimated the abilities of graphics cards, or they miscommunicated between development teams. It does not really matter at this point. The game has too many assets that need to be pushed to the graphics card out of caches or cabinet files and it's bogging computers down.

Now most people will not see much of it. As long as they keep the number of simultaneous player characters visible on screen down the amount of data is manageable. So they suspended the high resolution texture option for regular gameplay, considerably reducing the memory footprint, and they lowered the threshold for sharding a zone, reducing the chance that a player would get past that number of player characters where they would see performance degradation.

What they could not fix was the PvP part, where having other players around is kind of the entire point. And that is where almost everybody who comes here complaining is pointing at as the problem area. Ilum and the warzones.

 

But there is not going to be a quick fix possible and you (and other complainers like you) can whine, threaten or foam at the mouth, but that is not going to change the fact that any solution is going to take time and a lot of effort to implement.

For the problems reported with the graphics data cache getting trashed, leading to degraded performance and eventually crashes, that will require figuring out what goes wrong, trying to find a solution that will not make matters worse and the implementing and testing it.

For the memory footprint, that gets down to coding again (they need to be more aggressive in their cutting down the level of detail on more distant objects, but preferably only for objects that are unlikely to need a more detailed representation soon, and as things getting closer their higher quality representations should be phased in before it is actually needed to prevent hitching). And of course those extra low quality geometry and textures need to be created for everything which in itself is a mammoth undertaking.

For PvP specific optimisations one of the biggest and most difficult things is to try and come up with better predictive algorithsm so that the game has to load less models and textures that it ends up not or barely using. Of course that shifts the burden away from the GPU to the CPU to an extent and introduces more vulnerabilities for client side exploits.

 

I probably used words that were a bit too big, so here is the executive summary: It took 5 years of development to get the game in its current state. They are not going to fix anything in 5 days, so stop whining.

 

(and yes, if you have fps issues with the game outside of the PvP specific situations then you have a problem with your gaming computer, in that something else is bottlenecking the game from running smoothly. The game should be able to run with decent fps (outside of PvP that is) on a P45 motherboard, Core 2 duo E8500 processor and HD6850, which is hardly very high end. You will need plenty of memory both on the computer and on the graphics card though).

 

 

In other words, patience is a virtue and it won't take a shorter time fixing it just because you keep whining about it.

[MGriffith]

You are not the only one on these forums who has knowledge and/or experience with software coding.

As someone who claims to you should know that game type and genre have little impact, ultimately on game engine.

Yes, its a trite example, but look at the Unreal 3 engine.

 

I never claimed that the Unreal engine could not have been used. It would have meant more work in creating content.

And at the time that development on SW:TOR started the unreal engine was in an equal state of premature bugginess. Just read back what the developers of Tera had to say about it. It can be done, it obviously has been done. That does not mean it is automatically the best choice. But I have no idea -why- Bioware chose the Hero Engine. Nobody here does, and Bioware is not going to tell.

I do however doubt it was not based on at least the expectation that it would meet the requirements laid out in the design brief at least as well or better than the alternatives

 

You claim there is a difference; but there is not. The differences come in how easy the engine is to develop with, how difficult it is to modify, and the technical specifications of what it can do, such as incorporation of physx/havoc.

 

Plus memory management, networking support, resistance to third party tampering, licensing structure and fees and so on. There is more to a game engine than the technical aspects of getting an image on screen.

Again, I have no idea how exactly the Hero Engine compared to the Unreal 3 engine, 5 years ago. It has (had) some advantages in how easy it was to work with and how well developed its backend suite was. It also has (had) some disadvantages in that it has a heftier system requirement than the Unreal Engine. For the Hero Engine the stated minimum specs are a bit of wishful thinking. For Unreal 3 they allow for a (barely) functional gameplay at minimum quality settings.

 

Now, the EVE developers have shown us that with proper netcode (which TOR seems to lack) all the client does is interact with the server databases, then render those changes in a graphical display for the client. As such, if one has client side assets in a format that is usable by more then one client, modifying one engine to make the neccessary calls and translate those calls and the responses it sees as appropriate, you could have both a hero engine client for those with low end machines and an unreal/dragon age client for those with high end machines.

 

Actually, EVE is struggling with its large scale combat support (granted, large scale on that game is an order of magnitude bigger than entire planets on SW:TOR, and what on EVE is considered a big fight is an entire server at heavy load in SW:TOR), and in part this has to do with its networking code (all those communications between different servers really bog things down server side, as opposed to the client side gridlock of SW:TOR.

 

Your point however is well made. SW:TOR has some serious issues with its memory management, and there seem to be some server side problems (some possibly relating to collision detection, and others related to your companion not being where it should be according to the client). That is at least my impression but it is a bit difficult to get a clearer picture of the cause of the occasional server lag.

These are not simple problems to identify and solve, and it is a bit too easy to just shout 'it is the Hero Engine and if Bioware just ripped it out and replaced it with Unreal 3 then next week all our problems could be solved'. Which is the gist of too many contributions on this type of discussion thread.

It is important that there are some posts as well that explain why the problem might occur, and why fixing it is not going to be a matter of a magic nor of flipping a switch. Which is what I have been trying to do. Feel free to do a better job than me

[Grammarye]

It is perhaps important to remember that motivations for choosing given technologies are many, and lots of different factors weigh in on that decision. 20-20 hindsight of 'this area of your engine sucks' is the worst way to approach the discussion, because you're glossing over all the ways in which the technology works well (and this applies to any such discussion, not just this one) - usually because the poster has insufficient background to actually see those.

 

Writing your own engine from scratch (because that is the most oft-quoted alternative) is not exactly simple; it's not like you can just go 'hey Blizzard, you did that great engine, can we have a look?'. Every time a company builds an MMO for the first time, you have a new set of developers who must pretty much make the same mistakes and learning experiences as all their competitors. That is just the way this stuff works. That sets aside that every MMO genuinely is different; it doesn't take much to diverge (in software terms) away from what works in a given engine context.

 

I continue to be astonished at how so many people latch onto phrases like single-threaded and blow it all out of proportion, forgetting that some of the most complex advanced games of today are still running on exactly one thread. Threads are an ideal way to spread parallel work across cores. Just what in a game engine is genuinely parallel? That's a much tougher question than most realise. Sure, some engines offer multithreading for some tasks, but ultimately your input & render loop isn't going anywhere except your main thread, not even with DX11.

 

It's also way, way, way too easy for people to look at an 'engine' and believe that does everything e.g. Gamebryo has been mentioned. It is a graphics/physics/audio engine people - that's it. Hero does substantially more than just a pretty simulation. In many ways, I wish people would stop using the term engine; it's been so grossly abused as to have no meaning left.

Edited March 5, 2012 by Grammarye

[ctghost]

For anyone interested, here's a demo video of the Unreal Engine 3. I stumbled on this video when browsing Kotaku.

 

[Jederix]

Isn't Tera kind of a joke? Also, you can't really comment due to it not being released.

 

And w.t.f is blade and soul?

 

Just another generic Korean grinder MMO from what I've heard.

[Jederix]

If you know anything about the Austin team, you'll quickly realize why the game is in the shape it is. It's not just that the engine was a poor choice.

 

Clarify? Unless it is wild speculation or some vendetta you have against Bioware. Then no one here is interested in your clarification.