Seems clear to me they where seduced by the features of the developer toolset rather than what was produced at the other end. It also appears that those tools did not scale to well either.
Ya the hero builder probably said what Ba wanted to ear and Ba forgot to check if this ass a sustainable engine for massive multiplayer.it is a great multiplayer engine but so far i haven't seen the massive part in it.i ll grant it to them tho for most it wont Mather cause they used what blizzard used open instance.(fireland daily are like that in wow)and i bet all game will be like that soon.it is a good compromise.but still doesn't help the engine
Originally posted by Praetalus Originally posted by drbaltazar Max amount of player supported by donnybrooks about 1000 (even Ms say a number)number of player supported on the same map none in hero engine.and when you read the multiplayer link it doesn't help either right then and there ea shouldn't have baught this a company that isn't capable of proving to you how many player on a single map the game support(ilum) i am sorry but if Ms can do it then it means this should be the first question answered if you are into buying an engine.
What is the maximum capacity of a single area server? "This depends on the hardware and the CPU budget for a user. For example, you can define the CPU budget on a per user basis to be 10 megahertz. If you have hardware that runs at 3 gigahertz, approximately 300 users could exist on the same area server. However, some of the scripting functions such as broadcast chat do not all scale linearly, so the actual capacity of the server would be somewhat less. If you have a game implementation that offloads some of the CPU budget to other servers, then more users will fit."
They break the areas up into "area servers" from what I understand. Still reading...lol.
If it was never their goal to have 300 hundred people standing around a single quest giver, then this wouldn't have been an issue. If their goal was a maximum of 200 people at a time in a particular area (Ilum), this would be within their target specs.
I can not remember winning or losing a single debate on the internet.
RIFT is capable of having a 100+ people battling giant NPCs and each other - with every spell effect you can imagine going off... and not imploding into a slideshow like Ilum.
Bioware obtaining HERO engine wasn't the wrong move, using it on SWTOR first was.
Oh really? Do you mind to elaborate why?
If you actually read my post I've actually explained it
While experimenting on an engine is good, it surely made the engine popular. But they should have tried the engine on a single player game first, to test the potential and figure out its limits. To use a engine in its primitive state, without knowing the full extend of what it can do, will simply restrict the game to what it evolves into. There is a reason why they state they are following the WoW successfuly model, not just because they will make loads of money, but also beacuse the engine limitations.
HERO engine is great for its use of cloud server to link different studios together and work coherently without using extra human resources. But the programming optimisation just isn't up to the 'next gen' level. SWTOR is the first AAA game to use HERO engine, and MMORPG should be the first to do many things, but testing a new engine is not one of them, too much is at stake there, development time is long, you can't really go back on your choice.
Press interviews should ask Bioware, if you have the choice again, would you use HERO engine again? If yes, why so? And see if they can give a convincing answer.
PS: HERO engine is good, but it isn't the greatest out there for SWTOR, or any MMO who wants to be next gen, the tech simply isn't optimised yet, cloud technology is still very primitive, and the programming support isn't well polish yet.
How much WoW could a WoWhater hate, if a WoWhater could hate WoW? As much WoW as a WoWhater would, if a WoWhater could hate WoW.
Originally posted by xKingdomx But they should have tried the engine on a single player game first, to test the potential and figure out its limits....To use a engine in its primitive state, without knowing the full extend of what it can do...There is a reason why they state they are following the WoW successfuly model, not just because they will make loads of money, but also beacuse the engine limitations....But the programming optimisation just isn't up to the 'next gen' level.
1) Can you backup your claim that BioWare is lacking knowledge about the engine and it's capabilities?
2) I assume you must have extensive knowledge about the engine and SWTOR code in order to make such claims. Do you mind sharing with us the details about this supposed "primitive" use of Hero engine in SWTOR and provide examples how it could have been done better?
3) Can you provide examples of those "limitations" you talk about?
4) What is "next gen programming optimization" supposed to be...? I guess you know a better engine that could be used and I expect you to provide examples of names and whys...
Just 2 more things:
There are points of "no return" in any programming, regardless whether you develop your own code or purchase one.
BioWare was already asked if they would use Hero engine again for SWTOR and they said: No.
Does that mean the engine is necessarily technically insufficient or inappropriate for SWTOR? No.
But they should have tried the engine on a single player game first, to test the potential and figure out its limits.
...
To use a engine in its primitive state, without knowing the full extend of what it can do
...
There is a reason why they state they are following the WoW successfuly model, not just because they will make loads of money, but also beacuse the engine limitations.
...
But the programming optimisation just isn't up to the 'next gen' level.
1) Can you backup your claim that BioWare is lacking knowledge about the engine and it's capabilities?
2) I assume you must have extensive knowledge about the engine and SWTOR code in order to make such claims. Do you mind sharing with us the details about this supposed "primitive" use of Hero engine in SWTOR and provide examples how it could have been done better?
3) Can you provide examples of those "limitations" you talk about?
4) What is "next gen programming optimization" supposed to be...? I guess you know a better engine that could be used and I expect you to provide examples of names and whys...
Just 2 more things:
There are points of "no return" in any programming, regardless whether you develop your own code or purchase one.
BioWare was already asked if they would use Hero engine again for SWTOR and they said: No.
Does that mean the engine is necessarily technically insufficient or inappropriate for SWTOR? No.
Nope, it does not as they did not specify the reasons...
But I guess you are as many others on these boards comfortable to fill the blanks with anything that suits your case, aren't you?
IMO the engine works fine for PvE and small skirmish type action.
If they are able to pull miracle patches like the latest (FPS went 20->40 on my rig in Imp Fleet) they may be able to make warfronts run fine.
I don't believe they are able to easily support massive PvP action in SW:ToR, but then again I'm capable to accept that. Those battles, although they've been very fun, have been a huge disappointment techwise in MMOs (AoC Sieges, Aion Reshanta madness, WAR Fort lagfests).
It seems like the Hero Engine has severe issues when there many players on the same place.
Or more than 10...
I drop to 10 fps on the fleet, with SWTORs terribad graphics and textures, there is no excuse for it, the game looks like crap and runs like crap.
Look at Tera...
I played K-Tera and the graphics are much much better than SWTOR, yet when i was part of 50 vs 50 PVP i had NO fps issues.
This isn't a "game vs game" crap or anything like that, just giving an example of a great looking game that runs better than SWTOR, even on huge pvp battles with 100+ players.
I have run tests in the hero engine with about 30 people in the same place all hitting on each other without any issues and that was on the build server that gets setup when you get a hero account. I am also sure the peopel behind dominus have done this, considering PVP is going to be a big part of their game and its also using the hero Engine. Bioware just blew their budget on voice acting and advertising..
But they should have tried the engine on a single player game first, to test the potential and figure out its limits.
To use a engine in its primitive state, without knowing the full extend of what it can do
There is a reason why they state they are following the WoW successfuly model, not just because they will make loads of money, but also beacuse the engine limitations.
But the programming optimisation just isn't up to the 'next gen' level.
1) Can you backup your claim that BioWare is lacking knowledge about the engine and it's capabilities?
2) I assume you must have extensive knowledge about the engine and SWTOR code in order to make such claims. Do you mind sharing with us the details about this supposed "primitive" use of Hero engine in SWTOR and provide examples how it could have been done better?
3) Can you provide examples of those "limitations" you talk about?
4) What is "next gen programming optimization" supposed to be...? I guess you know a better engine that could be used and I expect you to provide examples of names and whys...
Just 2 more things:
There are points of "no return" in any programming, regardless whether you develop your own code or purchase one.
BioWare was already asked if they would use Hero engine again for SWTOR and they said: No.
Does that mean the engine is necessarily technically insufficient or inappropriate for SWTOR? No.
It is a typical issue for any middleware...
-
1) I never said Bioware lacked knowledge about the engine, just its capabilities. Of course they have guys who can code and programming, but every engine is different. Every project starts with programmer understanding how the engine works, but the problem with the HERO engine is that the support simply isn't there. If I didn't remember wrong, HERO engine was licensed to Bioware before the first game from HERO engine (Hero's journey) even got released. No engine is perfect from the beginning, and to license and use this engine, when its first game hasn't even been finish, on a MMORPG, which is one of the hardest task in a game programming. (Note:
The early state of the engine warrents Bioware to use the engine on another, preferably online multiplayer (since the engine was made to service MMO), game to just simply know the full extend of engine, then to decide whether use it on SWTOR, before crossing the "no return" line you have mentioned.
For any backup you want, this are the documentation of HERO Engine compare to Cryengine which released the engine to public last year with documentation
You can see CryEngine have a lot more in terms of documentation, physics simulation, graphics and rendering, asset creation and plugins to third party software, HERO Engine doesn't have much in documentation for developers to solve problems. Any game developers can tell you how much they communicate with the engine developers for support and upgrades. Not to mention CryEngine also have a much larger modding community to solve problem (yes devs look at those forums for help as well). This proves how lessen known are HERO engines capabilities to the devs. HERO engine was used because it was made for MMO (doesn't mean it is the best choice) and the rapid prototyping abilities of HERO engine.
2) I answered quite a bit of this question above, but will repeat the main points. They licensed the engine in 2005 before HERO's journey was even released (if I'm not wrong, it never got released or finished)
This was an article how Bioware ended up using HERO engine
“It’s not productized yet,” we told Gordon Walton. “There are whole sections of code that is only roughed in and not optimized for performance or security. And there are very few comments and very little documentation.”
He didn’t care. “We are going to have tons of engineers. We can finish it ourselves. We’re going to want to modify your source code for our special project anyway.”
3) As for limitations, its hard to say to be exact, since they have heavily modified the HERO engine, and SWTOR's engine is not publicly released, but I'll refer to the HERO engine at its current state
"In the current release, simulation and rendering run on a single shared core."
So basically the entire game runs on one core, a huge limitation when almost all mid to low range computers will run dual core processors by now.
You can say Bioware might have made a lot of changes, but Bioware was never known for their programming prowess like EPIC games or Crytek. But this information is based on HERO engine in Sep 2011.
4) The fact that their engine cannot process high amount of polygons due to many reports of high lag in heavy populate areas, the multicore optimization or the shader optimization. Engines like Source and UnReal (used in Mass Effect already) would've been a better choice. Due to the amount of official and community support (CryEngine would've made the game too power demanding I think unless they work really hard on optimization)
As for Bioware not using HERO engine again, can you think of a reason other than "engine is NOT necessarily technically insufficient or inappropriate for SWTOR"
I'm not saying Bioware shouldn't use HERO engine, but instead made a smaller scale game first, not a MMORPG. Let the engine develope further before locking themselves into HERO engine.
How much WoW could a WoWhater hate, if a WoWhater could hate WoW? As much WoW as a WoWhater would, if a WoWhater could hate WoW.
Originally posted by xKingdomx As for Bioware not using HERO engine again, can you think of a reason other than "engine is NOT necessarily technically insufficient or inappropriate for SWTOR"
I'm not saying Bioware shouldn't use HERO engine, but instead made a smaller scale game first, not a MMORPG. Let the engine develope further before locking themselves into HERO engine.
I won't address your wall of text specifically as I do not think there is a need to nitpick and point out some obvious loopholes, I would say that your whole point can be summed up:
"Hero engine is less popular and used than CryEngine"
Which is fairly meaningless statement as it does not imply that CryEngine would be better engine for SWTOR...
No ground there.
2) Yep, standard practice. Middleware is meant to be customized and in many case pretty much all re-written...
3) You have just described how all MMOs operate.
There is no real utilization for multi-core. You do not really need more than 2 threads...and only 1 thread will cause significant load anyway. In fact, for gaming purposes it is still better to have more power per core rather than number of less powerful cores...
4) You are again comparing apples and oranges. MMO is server/client architecture, do not compare it to stand alone apps.
However, this is something I am wondering myself about too. No hint what the cause of an issue might be though...
Why no Hero engine for SWOTR?
Hero engine might have been proven to be technically plausible but time and resources wise unfeasible.
Many things might have made them change their mind, technical parameters of the engine aren't the only thing to weigh.
Making "small scale game" would not tell you about intricacies of MMO development, moot point again...
As for Bioware not using HERO engine again, can you think of a reason other than "engine is NOT necessarily technically insufficient or inappropriate for SWTOR"
I'm not saying Bioware shouldn't use HERO engine, but instead made a smaller scale game first, not a MMORPG. Let the engine develope further before locking themselves into HERO engine.
I won't address your wall of text specifically as I do not think there is a need to nitpick and point out some obvious loopholes, I would say that your whole point can be summed up:
"Hero engine is less popular and used than CryEngine"
Which is fairly meaningless statement as it does not imply that CryEngine would be better engine for SWTOR...
Funny, I specifically said they shouldn't use CryEngine since it will make the game too power demanding, which is extremely bad for a MMORPG who wants a large playerbase.
No ground there.
2) Yep, standard practice. Middleware is meant to be customized and in many case pretty much all re-written...
I don't think you got what I'm saying, they licensd an unfinished engine.
“It’s not productized yet,” we told Gordon Walton. “There are whole sections of code that is only roughed in and not optimized for performance or security. And there are very few comments and very little documentation.”
3) You have just described how all MMOs operate.
There is no real utilization for multi-core. You do not really need more than 2 threads...and only 1 thread will cause significant load anyway. In fact, for gaming purposes it is still better to have more power per core rather than number of less powerful cores...
Well I'm no tech guru, but I thought rendering and simulation can be done on two different cores. Maybe you are right then.
Not to mention the background processes on a lot of the home computers.
4) You are again comparing apples and oranges. MMO is server/client architecture, do not compare it to stand alone apps.
Unreal and Source have been used in MMO developement before, with a few coming as well (well Unreal mostly, I still find Vindictus hard to be called as a MMO lol)
However, this is something I am wondering myself about too. No hint what the cause of an issue might be though...
Why no Hero engine for SWOTR?
Hero engine might have been proven to be technically plausible but time and resources wise unfeasible.
HERO engine main attraction is the HERO cloud feature, which utilising cloud technology to let two different studios to work on the same project from two different geographical location, efficiently speeding up the prototyping process. If HERO engine is the best at anything, its probably time and resource management.
Many things might have made them change their mind, technical parameters of the engine aren't the only thing to weigh.
There aren't many parameters in choosing an engine, technical difficulties are usually the big one, because techinical difficulties can lead to time and resource consumption, leading to financial problems.
I dont think you choose an engine based on one specific problem or advantage, but in terms of what is the best for the project overall. Its like choosing a tennis racquet, you don't just use it for the new technology, but also the feel and stability
Making "small scale game" would not tell you about intricacies of MMO development, moot point again...
Smaller scale, not small scale. Meaning a normal Bioware game, maybe a Dragon Age game or something. There is a lot of developement process that is the same between MMO and singleplayer. Polygon limitation, shader optimization, graphical rendering optimization, physics simulation, and maybe even programming/scripts optimization.
These are things Bioware would've learned if they have used it on another game beforehand. Even network traffic can be touched on if they integrate online elements into the game. Maybe something like online instances Dragon Age game like Guild Wars 1. Is it a full fledge MMORPG, no, but it is testing the engine and learning the engine while also be able to earn some profits.
How much WoW could a WoWhater hate, if a WoWhater could hate WoW? As much WoW as a WoWhater would, if a WoWhater could hate WoW.
Out of curiosity ive been to the hero engine website and yes, as others stated before one of its selling features was this developer cloud technology. If Bioware is correct in its statement that 800 ppl from all continets were working on the game, its becoming clear that they probably picked the engine because of this single feature. I dont think its good to pick something for a singular reason.
Lots of engines provide ways of donig this, Hell even Unity 3d pro has this feature..
I dont even think EA/Bioware used the cloud system as they brought full access to the engine and probally run it all in house.
The main reason the team im working in decided against using hero cloud was because our prgrammers did not want to learn the hero script language.. Other than that the engine seemed to be very capable and there are some really nice upcoming indie games using the hero engine..
Originally posted by xKingdomx Funny, I specifically said they shouldn't use CryEngine since it will make the game too power demanding, which is extremely bad for a MMORPG who wants a large playerbase.
...I don't think you got what I'm saying, they licensd an unfinished engine. ...Unreal and Source have been used in MMO developement before, with a few coming as well (well Unreal mostly, I still find Vindictus hard to be called as a MMO lol) HERO engine main attraction is the HERO cloud feature, which utilising cloud technology to let two different studios to work on the same project from two different geographical location, efficiently speeding up the prototyping process. If HERO engine is the best at anything, its probably time and resource management.
There aren't many parameters in choosing an engine, technical difficulties are usually the big one, because techinical difficulties can lead to time and resource consumption, leading to financial problems.I dont think you choose an engine based on one specific problem or advantage, but in terms of what is the best for the project overall. Its like choosing a tennis racquet, you don't just use it for the new technology, but also the feel and stability Smaller scale, not small scale. Meaning a normal Bioware game, maybe a Dragon Age game or something. There is a lot of developement process that is the same between MMO and singleplayer. Polygon limitation, shader optimization, graphical rendering optimization, physics simulation, and maybe even programming/scripts optimization. These are things Bioware would've learned if they have used it on another game beforehand. Even network traffic can be touched on if they integrate online elements into the game. Maybe something like online instances Dragon Age game like Guild Wars 1. Is it a full fledge MMORPG, no, but it is testing the engine and learning the engine while also be able to earn some profits.
1) Why did you bring up CryEngine as an counter to Hero Engine in the first place then?
To show that there are some supposedly better, yet non-viable options?
2) And I said that is does not matter, you gonna rewrite it anyway. What was important to them was the parts that were finished. Not picking an engine because it does not supports things you have no interest in would be silly...
3) CLOUD feature is marketing gimmick only, very much unimportant to majority of developers.
What I had on mind is feasibility determined by amount of time invested in engine customization, faced challenges and expectations.
4) There aren't? Why people use Windows instead of more technically sophisticated NIX platform then?
If you do not think that an engine can't be picked up just for one specific feature, why did you consider CLOUD tech a main attraction of Hero engine?
You will not know what engine will be better overall until you actually use one and make a final product.
That's the problem especially of middleware - you do not know what your needs will be, and that's why you usually pick a few things you can identify as crucial and try to find a solution for them.
BioWare has a large team of people who can supplement for unfinished parts of the engine and do other modifications but as other poster pointed out, there might be parts such as network tools that BioWare simply preferred to purchase instead of trying to develop on their own.
5) You say that BioWare should spent 50M USD and 4 years of development just to try out the engine? Do you seriously think that makes any sense?
Engine is still just a set of tools and if you are a bit experienced programmer you can swap tools with ease. You do not need 4 years and 50M USD to learn new API or language...
At the end, it is programmers and designers making the game, not the engine.
Honestly, I do not know what your beef is and why are you so much advert that using Hero engine was a bad move...
I don't think should have built their own engine necessarily. They could have easily went with some other engines mentioned in this thread. I think what they should have done was not go with the Hero Engine obviously. It's flawed in so many ways, and it's limitations are not only clear in the planet design, but also in the performance on ilum. So should they have built their own engine over Hero is the real question and it's pretty damn obvious that anything would have been better than what currently exists.
The weather issue is just one of many other comments from the devs that show their laziness in designing an immersive game. If you ever wonder why something isn't in the game the response is, it didn't really make sense to do that. And yet they claim they were innovative. Except that even planets without life have complex weather systems in our own solar system, but these don't? Don't even get me started on day/night. Apparently the sun rising on a planet doesn't fit in with a planet having a dreadful feel... lmao
Even if they had spent only 50 million on this game, I would still be left wondering where exactly all the money went.
I really dont think this is fair to say. I never noticed or felt a need for there to be either a day/night cycle or weather system in SWTOR. The game is a new step in a direction that hasnt really been taken before, the small things like weather and day/night are really minor and I dont think make the package any less if they were there in the first place.
You have to take a look at the priority of there stuff, do you really think Bioware couldnt do those things in there game if they felt it was nessasary, of course they could. This game isnt ment to be a simulator like many other mmos, its like reading a book, you dont have to constantly know what time of day it is or what the weather is doing, the story is what compels you.
I don't think should have built their own engine necessarily. They could have easily went with some other engines mentioned in this thread. I think what they should have done was not go with the Hero Engine obviously. It's flawed in so many ways, and it's limitations are not only clear in the planet design, but also in the performance on ilum. So should they have built their own engine over Hero is the real question and it's pretty damn obvious that anything would have been better than what currently exists.
The weather issue is just one of many other comments from the devs that show their laziness in designing an immersive game. If you ever wonder why something isn't in the game the response is, it didn't really make sense to do that. And yet they claim they were innovative. Except that even planets without life have complex weather systems in our own solar system, but these don't? Don't even get me started on day/night. Apparently the sun rising on a planet doesn't fit in with a planet having a dreadful feel... lmao
Even if they had spent only 50 million on this game, I would still be left wondering where exactly all the money went.
I really dont think this is fair to say. I never noticed or felt a need for there to be either a day/night cycle or weather system in SWTOR. The game is a new step in a direction that hasnt really been taken before, the small things like weather and day/night are really minor and I dont think make the package any less if they were there in the first place.
You have to take a look at the priority of there stuff, do you really think Bioware couldnt do those things in there game if they felt it was nessasary, of course they could. This game isnt ment to be a simulator like many other mmos, its like reading a book, you dont have to constantly know what time of day it is or what the weather is doing, the story is what compels you.
It is clear that most of you have no idea what you're talking about on this subject. Many of you keep saying they should have used the Unreal Engine or CryEngine. These are graphics engines... they lack the MMO multiplayer framework and netcode, which is the hardest thing to design when it comes to an MMO. If Bioware had gone with one of these engines, they would have had to write the server net code from scratch.
The HERO engine provides both the multiplayer netcode AND the graphics engine all in one and it allows many different developers to work on the same project at the same time without the need for nightly builds or re-compiles. The engine has a lot of positives and I can totally understand why Bioware went with it. They saved themselves easily two years of development time by electing not to write their own MMO server code. HERO allowed them to get working on the game right away, rather than having to deal with the multiplayer netcode... something Bioware has almost no prior experience in.
So no, they should not have built their own engine. If they did, the game would probably not release before 2015.
Comments
Seems clear to me they where seduced by the features of the developer toolset rather than what was produced at the other end. It also appears that those tools did not scale to well either.
I agree that no one can for sure say its the engine but its very very suspect!
And ofc no way will BW admit it as its the heart of the game.
Go to his house play it on his computer and then you will see why he doesnt have to. Copy his specs while you're at it. Get yourself a new pc.
What is the maximum capacity of a single area server?
"This depends on the hardware and the CPU budget for a user. For example, you can define the CPU budget on a per user basis to be 10 megahertz. If you have hardware that runs at 3 gigahertz, approximately 300 users could exist on the same area server. However, some of the scripting functions such as broadcast chat do not all scale linearly, so the actual capacity of the server would be somewhat less. If you have a game implementation that offloads some of the CPU budget to other servers, then more users will fit."
They break the areas up into "area servers" from what I understand. Still reading...lol.
If it was never their goal to have 300 hundred people standing around a single quest giver, then this wouldn't have been an issue. If their goal was a maximum of 200 people at a time in a particular area (Ilum), this would be within their target specs.
I can not remember winning or losing a single debate on the internet.
RIFT is capable of having a 100+ people battling giant NPCs and each other - with every spell effect you can imagine going off... and not imploding into a slideshow like Ilum.
The engine is gimpy for an MMO.
If you actually read my post I've actually explained it
While experimenting on an engine is good, it surely made the engine popular. But they should have tried the engine on a single player game first, to test the potential and figure out its limits. To use a engine in its primitive state, without knowing the full extend of what it can do, will simply restrict the game to what it evolves into. There is a reason why they state they are following the WoW successfuly model, not just because they will make loads of money, but also beacuse the engine limitations.
HERO engine is great for its use of cloud server to link different studios together and work coherently without using extra human resources. But the programming optimisation just isn't up to the 'next gen' level. SWTOR is the first AAA game to use HERO engine, and MMORPG should be the first to do many things, but testing a new engine is not one of them, too much is at stake there, development time is long, you can't really go back on your choice.
Press interviews should ask Bioware, if you have the choice again, would you use HERO engine again? If yes, why so? And see if they can give a convincing answer.
PS: HERO engine is good, but it isn't the greatest out there for SWTOR, or any MMO who wants to be next gen, the tech simply isn't optimised yet, cloud technology is still very primitive, and the programming support isn't well polish yet.
How much WoW could a WoWhater hate, if a WoWhater could hate WoW?
As much WoW as a WoWhater would, if a WoWhater could hate WoW.
1) Can you backup your claim that BioWare is lacking knowledge about the engine and it's capabilities?
2) I assume you must have extensive knowledge about the engine and SWTOR code in order to make such claims. Do you mind sharing with us the details about this supposed "primitive" use of Hero engine in SWTOR and provide examples how it could have been done better?
3) Can you provide examples of those "limitations" you talk about?
4) What is "next gen programming optimization" supposed to be...? I guess you know a better engine that could be used and I expect you to provide examples of names and whys...
Just 2 more things:
There are points of "no return" in any programming, regardless whether you develop your own code or purchase one.
BioWare was already asked if they would use Hero engine again for SWTOR and they said: No.
Does that mean the engine is necessarily technically insufficient or inappropriate for SWTOR? No.
It is a typical issue for any middleware...
-
Well that speaks volumes, doesn't it?
An honest review of SW:TOR 6/10 (Danny Wojcicki)
Nope, it does not as they did not specify the reasons...
But I guess you are as many others on these boards comfortable to fill the blanks with anything that suits your case, aren't you?
IMO the engine works fine for PvE and small skirmish type action.
If they are able to pull miracle patches like the latest (FPS went 20->40 on my rig in Imp Fleet) they may be able to make warfronts run fine.
I don't believe they are able to easily support massive PvP action in SW:ToR, but then again I'm capable to accept that. Those battles, although they've been very fun, have been a huge disappointment techwise in MMOs (AoC Sieges, Aion Reshanta madness, WAR Fort lagfests).
You mean maybe they didn't like the colors of the interface? That why they said they woulnd't use it again?
An honest review of SW:TOR 6/10 (Danny Wojcicki)
Thank you for proving my point.
I have run tests in the hero engine with about 30 people in the same place all hitting on each other without any issues and that was on the build server that gets setup when you get a hero account. I am also sure the peopel behind dominus have done this, considering PVP is going to be a big part of their game and its also using the hero Engine. Bioware just blew their budget on voice acting and advertising..
1) I never said Bioware lacked knowledge about the engine, just its capabilities. Of course they have guys who can code and programming, but every engine is different. Every project starts with programmer understanding how the engine works, but the problem with the HERO engine is that the support simply isn't there. If I didn't remember wrong, HERO engine was licensed to Bioware before the first game from HERO engine (Hero's journey) even got released. No engine is perfect from the beginning, and to license and use this engine, when its first game hasn't even been finish, on a MMORPG, which is one of the hardest task in a game programming. (Note:
The early state of the engine warrents Bioware to use the engine on another, preferably online multiplayer (since the engine was made to service MMO), game to just simply know the full extend of engine, then to decide whether use it on SWTOR, before crossing the "no return" line you have mentioned.
For any backup you want, this are the documentation of HERO Engine compare to Cryengine which released the engine to public last year with documentation
http://wiki.heroengine.com/wiki/Main_Page
http://freesdk.crydev.net/dashboard.action
You can see CryEngine have a lot more in terms of documentation, physics simulation, graphics and rendering, asset creation and plugins to third party software, HERO Engine doesn't have much in documentation for developers to solve problems. Any game developers can tell you how much they communicate with the engine developers for support and upgrades. Not to mention CryEngine also have a much larger modding community to solve problem (yes devs look at those forums for help as well). This proves how lessen known are HERO engines capabilities to the devs. HERO engine was used because it was made for MMO (doesn't mean it is the best choice) and the rapid prototyping abilities of HERO engine.
2) I answered quite a bit of this question above, but will repeat the main points. They licensed the engine in 2005 before HERO's journey was even released (if I'm not wrong, it never got released or finished)
This was an article how Bioware ended up using HERO engine
“It’s not productized yet,” we told Gordon Walton. “There are whole sections of code that is only roughed in and not optimized for performance or security. And there are very few comments and very little documentation.”
He didn’t care. “We are going to have tons of engineers. We can finish it ourselves. We’re going to want to modify your source code for our special project anyway.”
Source: http://www.heroengine.com/2011/11/heroengine-meets-starwars/
3) As for limitations, its hard to say to be exact, since they have heavily modified the HERO engine, and SWTOR's engine is not publicly released, but I'll refer to the HERO engine at its current state
"In the current release, simulation and rendering run on a single shared core."
Source: http://community.heroengine.com/forums/index.php?topic=889.msg4155#msg4155
So basically the entire game runs on one core, a huge limitation when almost all mid to low range computers will run dual core processors by now.
You can say Bioware might have made a lot of changes, but Bioware was never known for their programming prowess like EPIC games or Crytek. But this information is based on HERO engine in Sep 2011.
4) The fact that their engine cannot process high amount of polygons due to many reports of high lag in heavy populate areas, the multicore optimization or the shader optimization. Engines like Source and UnReal (used in Mass Effect already) would've been a better choice. Due to the amount of official and community support (CryEngine would've made the game too power demanding I think unless they work really hard on optimization)
As for Bioware not using HERO engine again, can you think of a reason other than "engine is NOT necessarily technically insufficient or inappropriate for SWTOR"
I'm not saying Bioware shouldn't use HERO engine, but instead made a smaller scale game first, not a MMORPG. Let the engine develope further before locking themselves into HERO engine.
How much WoW could a WoWhater hate, if a WoWhater could hate WoW?
As much WoW as a WoWhater would, if a WoWhater could hate WoW.
Damn you won.
An honest review of SW:TOR 6/10 (Danny Wojcicki)
I won't address your wall of text specifically as I do not think there is a need to nitpick and point out some obvious loopholes, I would say that your whole point can be summed up:
"Hero engine is less popular and used than CryEngine"
Which is fairly meaningless statement as it does not imply that CryEngine would be better engine for SWTOR...
No ground there.
2) Yep, standard practice. Middleware is meant to be customized and in many case pretty much all re-written...
3) You have just described how all MMOs operate.
There is no real utilization for multi-core. You do not really need more than 2 threads...and only 1 thread will cause significant load anyway. In fact, for gaming purposes it is still better to have more power per core rather than number of less powerful cores...
4) You are again comparing apples and oranges. MMO is server/client architecture, do not compare it to stand alone apps.
However, this is something I am wondering myself about too. No hint what the cause of an issue might be though...
Why no Hero engine for SWOTR?
Hero engine might have been proven to be technically plausible but time and resources wise unfeasible.
Many things might have made them change their mind, technical parameters of the engine aren't the only thing to weigh.
Making "small scale game" would not tell you about intricacies of MMO development, moot point again...
How much WoW could a WoWhater hate, if a WoWhater could hate WoW?
As much WoW as a WoWhater would, if a WoWhater could hate WoW.
Out of curiosity ive been to the hero engine website and yes, as others stated before one of its selling features was this developer cloud technology. If Bioware is correct in its statement that 800 ppl from all continets were working on the game, its becoming clear that they probably picked the engine because of this single feature. I dont think its good to pick something for a singular reason.
Lots of engines provide ways of donig this, Hell even Unity 3d pro has this feature..
I dont even think EA/Bioware used the cloud system as they brought full access to the engine and probally run it all in house.
The main reason the team im working in decided against using hero cloud was because our prgrammers did not want to learn the hero script language.. Other than that the engine seemed to be very capable and there are some really nice upcoming indie games using the hero engine..
1) Why did you bring up CryEngine as an counter to Hero Engine in the first place then?
To show that there are some supposedly better, yet non-viable options?
2) And I said that is does not matter, you gonna rewrite it anyway. What was important to them was the parts that were finished. Not picking an engine because it does not supports things you have no interest in would be silly...
3) CLOUD feature is marketing gimmick only, very much unimportant to majority of developers.
What I had on mind is feasibility determined by amount of time invested in engine customization, faced challenges and expectations.
4) There aren't? Why people use Windows instead of more technically sophisticated NIX platform then?
If you do not think that an engine can't be picked up just for one specific feature, why did you consider CLOUD tech a main attraction of Hero engine?
You will not know what engine will be better overall until you actually use one and make a final product.
That's the problem especially of middleware - you do not know what your needs will be, and that's why you usually pick a few things you can identify as crucial and try to find a solution for them.
BioWare has a large team of people who can supplement for unfinished parts of the engine and do other modifications but as other poster pointed out, there might be parts such as network tools that BioWare simply preferred to purchase instead of trying to develop on their own.
5) You say that BioWare should spent 50M USD and 4 years of development just to try out the engine? Do you seriously think that makes any sense?
Engine is still just a set of tools and if you are a bit experienced programmer you can swap tools with ease. You do not need 4 years and 50M USD to learn new API or language...
At the end, it is programmers and designers making the game, not the engine.
Honestly, I do not know what your beef is and why are you so much advert that using Hero engine was a bad move...
I really dont think this is fair to say. I never noticed or felt a need for there to be either a day/night cycle or weather system in SWTOR. The game is a new step in a direction that hasnt really been taken before, the small things like weather and day/night are really minor and I dont think make the package any less if they were there in the first place.
You have to take a look at the priority of there stuff, do you really think Bioware couldnt do those things in there game if they felt it was nessasary, of course they could. This game isnt ment to be a simulator like many other mmos, its like reading a book, you dont have to constantly know what time of day it is or what the weather is doing, the story is what compels you.
I'm speechless.
An honest review of SW:TOR 6/10 (Danny Wojcicki)
It is clear that most of you have no idea what you're talking about on this subject. Many of you keep saying they should have used the Unreal Engine or CryEngine. These are graphics engines... they lack the MMO multiplayer framework and netcode, which is the hardest thing to design when it comes to an MMO. If Bioware had gone with one of these engines, they would have had to write the server net code from scratch.
The HERO engine provides both the multiplayer netcode AND the graphics engine all in one and it allows many different developers to work on the same project at the same time without the need for nightly builds or re-compiles. The engine has a lot of positives and I can totally understand why Bioware went with it. They saved themselves easily two years of development time by electing not to write their own MMO server code. HERO allowed them to get working on the game right away, rather than having to deal with the multiplayer netcode... something Bioware has almost no prior experience in.
So no, they should not have built their own engine. If they did, the game would probably not release before 2015.