Heck, even CCP took Unreal 3 for their upcoming DUST thing. They tried to build their own engine for the world of darkness game and Walking in Stations in EVE (theres speculation they were beta testing the technology in EVE which angered people) and that has since fallen through. They themselves said they were overreaching doing all that an their company is (well was before the 20% layoffs) 600 ppl strong.
-
We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.
Heck, even CCP took Unreal 3 for their upcoming DUST thing. They tried to build their own engine for the world of darkness game and Walking in Stations in EVE (theres speculation they were beta testing the technology in EVE which angered people) and that has since fallen through. They themselves said they were overreaching doing all that an their company is (well was before the 20% layoffs) 600 ppl strong.
TBH they have given away copies of a product past it's shelf live and many already have and DLC's for this game well 2 of which alot already have and cost £3 each.
I don't think you could say they are exactly pushing the boat out here. lol
The ole art of appearing to give away something but actually giving away nothing that's really going to cost you much, that right Fred?
Removing large parts of the API is a big step, and it means that building an abstraction layer on top of both APIs is much harder. Some parts of the API, the DX10 counterparts were very different to their predecessors.They didn't really, it was more trimming out the bloat that DX9 (and which one, a, b or c?) suffered from. Of course, most of that bloat was due to backwards compatibility - worth remembering DX hadn't had a real overhaul since Windows 95.
It's a non-issue though. Most new games this year are also using DX10 as a baseline. It's only really the cross-platform releases that retain DX9, and that's simply because the current console generation is limited to DX9 (and even then, when they can be bothered to do proper ports it's turning up in DX10, it's getting to the point you can almost spot the shoddy ports by their DX9 support ...).
Also Mars 2 is based off the original Mars engine, so it's not like Kerberos built their own engine from the ground up for this - they upgraded the engine they built from the ground up for their previous release. I wouldn't speculate on the why; quite frankly there's as many downsides to licensing an engine as there is to building your own, not least of which is the cost.
In terms of 3D graphics, from what I've seen it's a big leap from Sots 1 to Sots 2. I would guess there was very little code of the original they could have reused. In the lower 3D graphics area the DX9 to DX10 move alone would mean a huge refactoring if not a rewrite. Slightly higher up Sots 1 was visually "OK", Sots 2 is visually impressive. This means many non-trivial rendering algorithms which weren't there before.
I'm not sure I agree with the licensing comment, what's available today is quite impressive and the costs are not that high. Looking at Ogre / Mogre alone which is open source shows you the starting point. In contrast, building an engine as visually impressive as Sots 2 looks from scratch is *a lot* of development time which has to be an order of magnitude higher than the licenses. I could be wrong on this though if anyone knows I'll be happy to learn some numbers.
The only good reason that I can think of for any company today to build it's own engine is if their requirements are unique. Which in this case I can't see.
I'd put money on them using the Mar2 engine for northstar.
IIRC Fort Zombie was the result of them using the Sots 1 engine for as a testbed for some of the environments they want to use in Northstar, and it wasn't very pretty. (total speculation here)
IIRC Fort Zombie was the result of them using the Sots 1 engine for as a testbed for some of the environments they want to use in Northstar, and it wasn't very pretty. (total speculation here)
There are some strong non-technical reasons for wanting your own engine if one can do it. Licensing fees, any issues if the owner goes under/gets bought out/etc., cheaper and easier work on other projects.
It's not just the graphics though - the physics, sound, everything else is also part of the engine.
It's a question of how well the engine does what you want it to do, and how much work you need to do to get it there. If you're using an off the peg engine you always have to make compromises in the design; by their very nature an engine designed for multiple applications will not be as good at a specific application as one which is designed purely for that specific application. In general, if you have the time and resources available building your own engine is always the better option.
Costs aren't that cheap for decent engines, especially not for a full commercial license. The bigger companies benefit simply through economies of scale - license the UT3 engine and they can probably get eight or nine games out of it. For a smaller developer who works on a single game at a time on the other hand you might get two games done before the engine is obsolete, but you'll pay the same price. Also bear in mind it's not just a single engine - you'll probably license one for the graphics, another for the physics, yet another for sound and another for network. Just take a look at the number of company logos on your average licensed AAA game.
Impossible to give a definite answer, since it really depends on how much you need to twist and bend the licensed engine.
Well, I disagree, but to explain why I would be simply repeating my previous posts in this thread and I'm trying to convince myself that I have a life.
Sorry, I take that back. The money part is somewhat different to what was said before.
I disagree because for a small title you have plenty of stuff around. A license for Unity is 1500$. Next to that there are Ogre / Mogre (open source). Physics simulation - Newton (free) and PhysiX by Nvidia (I think it's free now?). Sound is next to nothing, 3D audio libs have ridiculous costs. In short for a small title you can assemble parts for the engine with an indie budget, which is why so many of them are springing up now. Even if you need to write more on top of that, there is no way the costs of the licenses equate to the costs of building all that on your own.
For bigger titles yes the higher-end engines would be more expensive, but bigger titles can afford it. Didn't Epic offer the UT3 engine for free to indies or something like that ?
But more importantly my argument isn't that a full-fledged engine where you just drop your content and game-logic at the level of Epic or Id's engines is cheap (It might be, but I don't know). But parts that give you a lot of that (for example high-end rendering) are available for prices that an existing game developer with a successful title under its belt can afford. And these would cut down development costs by more than what they are worth.
I disagree because for a small title you have plenty of stuff around. A license for Unity is 1500$. Next to that there are Ogre / Mogre (open source). Physics simulation - Newton (free) and PhysiX by Nvidia (I think it's free now?). Sound is next to nothing, 3D audio libs have ridiculous costs. In short for a small title you can assemble parts for the engine with an indie budget, which is why so many of them are springing up now. Even if you need to write more on top of that, there is no way the costs of the licenses equate to the costs of building all that on your own.
For bigger titles yes the higher-end engines would be more expensive, but bigger titles can afford it. Didn't Epic offer the UT3 engine for free to indies or something like that ?
But more importantly my argument isn't that a full-fledged engine where you just drop your content and game-logic at the level of Epic or Id's engines is cheap (It might be, but I don't know). But parts that give you a lot of that (for example high-end rendering) are available for prices that an existing game developer with a successful title under its belt can afford. And these would cut down development costs by more than what they are worth.
I know the hate this in gonna engender for me but imo they should go back to Lighthouse for Northstar.
Whatever happened to them anyway?
Nope - not Lighthouse almost case of ANYTHING but Lighthouse.
Hopefully Kerberos have held onto the Intellectual Property rights to SOTS as they did when they were published by Lighthouse.
almost case of ANYTHING but Lighthouse.Hopefully Kerberos have held onto the Intellectual Property rights to SOTS as they did when they were published by Lighthouse.
Nope - not Lighthouse
Hopefully Kerberos have held onto the Intellectual Property rights to SOTS as they did when they were published by Lighthouse.
Huh looked like it was a good setup for the first game. I cant remember if I was there for launch, but I was there well before BoB.
I would think Paradox itself would be the most likely party to work on the game if Kerberos did not for whatever reason.
Building it on your own only costs whatever the salary is for the coders doing it, for the length of time they're doing it.
Licensing adds in the licensing cost, but doesn't necessarily save any time. You still have to tweak the engine to do what you want, and if you're licensing out other parts then you also need to spend time bringing them together.
Not exactly. It's free provided you don't sell the end product IIRC.
Not for a small studio. The only costs to develop the engine for Kerberos would be the salaries of the coders. Given they're paying them anyway whether they're working on something or twiddling their thumbs, you could say development costs nothing. It's only expensive for larger developers because they tend to have other projects those coders could be busy on - Kerberos AFAIK only has one game in development at a time. So it's basically get the code monkeys doing something, or pay them for doing nothing.
They went under. Declared bankrupt in 2009.
Anything else than unity please. It works nice and dandy on browser games but for sots type game, EEEK.
Hi amirabiri,
What you're saying about game engines is correct... up to a point. Assuming for a moment that they are competent developers, they are probably building layers on the Graphics engine. When making the choice between 3rd party code and rolling your own, it always comes down to:
1) Most important: can the engine even deliver the required features?
2) How easy is it to use the API of the 3rd party engine?
3) How stable is the engine?
4) How well does the engine perform?
3) What does it cost in terms of licensing, vs. us developing our own engine?
Even assuming that the 3rd party engine passes the business test, the features needed for SotS II require additional coding. You talked about how different ship sections are no different from an FPS character holding a different gun. This is true at a low level, but there still needs to be a lot of code written in a layer above the basic 3d / physics engine. Many FPS games do not care about where you are hit (some do), but even assuming the engine does this, you still have to map those 3d collisions onto actual entities. What I find intimidating about the SotS features were always 2 things:
1) Ship sections don't appear to be just single entities. There seems to be a hierarchy of objects here. Individual guns can be shot off of sections if I remember correctly from SotS prime.
2) The physics model of hit / collision detection looks fairly complex.
Ships seem to react to getting hit; projectiles and ships both have mass. Most of the FPS games I've played, you get thrown around by explosions but that's it. Consider the complex case of oblivion, where you shoot an arrow into a bucket and it starts swinging from the rope it hangs from. I know you mentioned PhysX but you still have to code in all your game entities by assigning mass and other physical properties. Each individual task may not seem like a lot of work but when you consider all the types of ship sections they have...
If it were my development team, I would have assigned some devs to investigate what's out there. What I would have accepted would be a basic engine that allows us to give solid objects momentum, support collision detection, and provide a way to plug-in basic physics rules (as per PhysX). IF that saves me development time and money, I would still have to put all the SotS-specific features into layers on top of that. And those rules are daunting. Most real time strategy games I have played treat a ship as one solid mass, without location-specific damage. I always thought SotS was rather ambitious with its damage model--and it's one of the reasons I loved prime. I never even got very far with GalCiv II because I couldn't get over the lack of tactical combat.
What you're saying about game engines is correct... up to a point. Assuming for a moment that they are competent developers, they are probably building layers on the Graphics engine. When making the choice between 3rd party code and rolling your own, it always comes down to:
1) Most important: can the engine even deliver the required features?
2) How easy is it to use the API of the 3rd party engine?
3) How stable is the engine?
4) How well does the engine perform?
3) What does it cost in terms of licensing, vs. us developing our own engine?
Even assuming that the 3rd party engine passes the business test, the features needed for SotS II require additional coding. You talked about how different ship sections are no different from an FPS character holding a different gun. This is true at a low level, but there still needs to be a lot of code written in a layer above the basic 3d / physics engine. Many FPS games do not care about where you are hit (some do), but even assuming the engine does this, you still have to map those 3d collisions onto actual entities. What I find intimidating about the SotS features were always 2 things:
1) Ship sections don't appear to be just single entities. There seems to be a hierarchy of objects here. Individual guns can be shot off of sections if I remember correctly from SotS prime.
2) The physics model of hit / collision detection looks fairly complex.
Ships seem to react to getting hit; projectiles and ships both have mass. Most of the FPS games I've played, you get thrown around by explosions but that's it. Consider the complex case of oblivion, where you shoot an arrow into a bucket and it starts swinging from the rope it hangs from. I know you mentioned PhysX but you still have to code in all your game entities by assigning mass and other physical properties. Each individual task may not seem like a lot of work but when you consider all the types of ship sections they have...
If it were my development team, I would have assigned some devs to investigate what's out there. What I would have accepted would be a basic engine that allows us to give solid objects momentum, support collision detection, and provide a way to plug-in basic physics rules (as per PhysX). IF that saves me development time and money, I would still have to put all the SotS-specific features into layers on top of that. And those rules are daunting. Most real time strategy games I have played treat a ship as one solid mass, without location-specific damage. I always thought SotS was rather ambitious with its damage model--and it's one of the reasons I loved prime. I never even got very far with GalCiv II because I couldn't get over the lack of tactical combat.
What is frustrating to me is that even now I still can't get their game to EVEN START. It was working until they released a patch on monday and now the window opens and closes and logs say it cant find the game directory. i tried everything they say in the forum for same result even deleting and reinstalling the game. this is so pathetic.
There are still a whole lot of people still using win XP .. I was using it until I got this game.. lol then upgraded my second hardrive for a dual boot system. Anyway point is while microsoft stopped supporting XP there are still a lot of games that do.
I had an issue with that where steam would say the game was unavailable. I fixed this by verifying that all the files were there through steam and the problem went away.