Crystal space anyone?
by Sean H. · in General Discussion · 08/12/2005 (11:19 am) · 11 replies
Ive been looking into some open source engines at dev masters and im getting interested in crystal space. however, im pretty reluctant to start the process of downloading, setting up environment variables, compiling etc. for an engine which wont be worth it. has anyone tried using this engine recently?
i know theres ogre as well but i downloaded a tech demo of it and the bsp level ran at about 0.8 fps on my system. somewhat disappointing. im thinking since crystal space is portal based, that it would be more efficient.
any opinions?
i know theres ogre as well but i downloaded a tech demo of it and the bsp level ran at about 0.8 fps on my system. somewhat disappointing. im thinking since crystal space is portal based, that it would be more efficient.
any opinions?
#2
The best engine besides Torque for handling BSP-style levels has been Irrlicht, if that's your focus. However, the outdoor handling isn't that great, and it's not a full game engine - and I'm not sure how good the tools support for it are. Feature-wise, all it's got over TGE is some additional shader support - it looks like TSE will clobber it on every front but price (Irrlicht is free).
Again - I could be a little out-of-date on my information, but when I last checked, Crystal Space really seemed to have crappy performance and development on it had slowed to a crawl. Potentially it could have improved by leaps and bounds over the last 18 months or so - but I don't expect so.
08/12/2005 (12:28 pm)
My information IS over a year out of date - but the last time I checked it out it was even more depressing performance-wise than Ogre. The best engine besides Torque for handling BSP-style levels has been Irrlicht, if that's your focus. However, the outdoor handling isn't that great, and it's not a full game engine - and I'm not sure how good the tools support for it are. Feature-wise, all it's got over TGE is some additional shader support - it looks like TSE will clobber it on every front but price (Irrlicht is free).
Again - I could be a little out-of-date on my information, but when I last checked, Crystal Space really seemed to have crappy performance and development on it had slowed to a crawl. Potentially it could have improved by leaps and bounds over the last 18 months or so - but I don't expect so.
#3
anyone else with recent experience with crystal space?
08/12/2005 (2:03 pm)
Thanks for the information jay. i'll take a look into irrlicht. a friend of mine was praising this engine awhile ago as well.anyone else with recent experience with crystal space?
#4
Not the best art pipeline, but if you wrtie new exporters, you're in business :P
08/12/2005 (2:07 pm)
CS was far less than useable in my experience. If you're not down for TGE or TSE, try Nebula Device 2. http://www.radonlabs.de/Not the best art pipeline, but if you wrtie new exporters, you're in business :P
#5
I stopped considering Crystal Space back in 1998 when after more than a year i was frustrated by the group going back and forth with different directions for the underlying tech and still had no stable 1.0 release. Here it is in 2005 and still no 1.0 release from this engine.
Save your self the trouble and scrounge up $100 for Torque.
08/12/2005 (2:19 pm)
Quote:Potentially it could have improved by leaps and bounds over the last 18 months or so - but I don't expect so.Ha... yea... don't expect it.
I stopped considering Crystal Space back in 1998 when after more than a year i was frustrated by the group going back and forth with different directions for the underlying tech and still had no stable 1.0 release. Here it is in 2005 and still no 1.0 release from this engine.
Save your self the trouble and scrounge up $100 for Torque.
#6
ok engine, looks ok, but hard to work with. The documentation is ok, but you will need a lot of c++ knowledge to get a game in there.
08/12/2005 (3:55 pm)
Ok in short:ok engine, looks ok, but hard to work with. The documentation is ok, but you will need a lot of c++ knowledge to get a game in there.
#8
08/12/2005 (9:35 pm)
The problem is that Ogre isn't a full game engine. Its just a graphics renderer, and if you feel it can't do that efficiently, then its probably not what you need for your game.
#9
I'm not saying this is better or worse than TGE/TSE. I honestly have no idea one way or the other. I just want to hear some opinions on this way of building a game vs. a full on game engine like Torque, good or bad.
-Ajari-
08/13/2005 (12:37 am)
This is from Ogre's web site as to why they are not a full on game engine.Quote:...even within the games industry, requirements can vary widely; for example a MMORPG will need a very different kind of network library than an FPS, and a flight sim will need a different kind of collision / physics system to fighting game. If OGRE included all these features, we would be enforcing a particular set of libraries on you, with an inbuilt set of assumed requirements, and that's not good design. Instead, we provide a very integration friendly API and let YOU choose the other libraries, if you want them. Many experienced game developers have expressed their approval of this approach, because there are no inbuilt constraints.
I'm not saying this is better or worse than TGE/TSE. I honestly have no idea one way or the other. I just want to hear some opinions on this way of building a game vs. a full on game engine like Torque, good or bad.
-Ajari-
#10
It's a solid methodology that focuses on a localization of sophistication. Just like Newton, Novodex, ODE, Havok, etc focus on physics, OGRE focuses on being a tight rendering engine versus a feature-rich game engine. And it works very well for what it provides. What it does not provide, the development team is left to incorporate themselves (or find something they can add on). Torque is in an intermediary position between polarities. It provides a set group of functionality that can make development of certain title easier or more difficult depending on the particulars of a project. Using OGRE, the particulars would have to be built from the ground up rather than extended. On the other end of the pole are engines like 3D Game Maker, which allow absolutely no extension to the engine. They provide an all-in-one solution. A6 is close to this side of the spectrum since they allow extension of the engine through a SDK but not through complete source control.
With a team of competant programmers, creative designers, and exceptional artists, nearly any engine and development model can be utilized to create a game. Whether creating features from the ground-up or incorporating elements (OGRE renderer, ODE physics, etc) or utilizing (and often extending) pre-built functionality, the team and their experience are stronger keys than the tools they are utilizing.
08/13/2005 (11:27 am)
Quote:I just want to hear some opinions on this way of building a game vs. a full on game engine like Torque, good or bad.
It's a solid methodology that focuses on a localization of sophistication. Just like Newton, Novodex, ODE, Havok, etc focus on physics, OGRE focuses on being a tight rendering engine versus a feature-rich game engine. And it works very well for what it provides. What it does not provide, the development team is left to incorporate themselves (or find something they can add on). Torque is in an intermediary position between polarities. It provides a set group of functionality that can make development of certain title easier or more difficult depending on the particulars of a project. Using OGRE, the particulars would have to be built from the ground up rather than extended. On the other end of the pole are engines like 3D Game Maker, which allow absolutely no extension to the engine. They provide an all-in-one solution. A6 is close to this side of the spectrum since they allow extension of the engine through a SDK but not through complete source control.
With a team of competant programmers, creative designers, and exceptional artists, nearly any engine and development model can be utilized to create a game. Whether creating features from the ground-up or incorporating elements (OGRE renderer, ODE physics, etc) or utilizing (and often extending) pre-built functionality, the team and their experience are stronger keys than the tools they are utilizing.
#11
Void War was done pretty much "from scratch" with no engine at ALL other than DirectX. I've worked with several custom 3D solutions in my career, so startng with nothing at all (or with just a 3D graphics engine) doesn't scare me much.
But it is enough of a pain in the butt that I switched over to Torque for my next game.
Why? The Void War engine *works* as it stands now. It COULD be re-purposed. What made my decision for me?
Three reasons:
#1 - there's a heck of a lot of work that goes into making a game engine (or re-purposing an engine) that has already gone into Torque. Things that you normally don't think of, things that aren't too sexy - like tools pipelines, input devices, UI functionality, dropped packets in multiplayer not screwing up game synchronization, handling configuration files & "mission files" (or *GASP* scripting!) for data-driven game development, and so forth. Being able to start with a stable and solid, already data-driven game engine means you get to jump right into the "fun stuff" - the actual game logic unique to your game.
#2 - A mature engine means a lot of the bugs have been ironed out already. If you have a graphics engine that has already been used in several commercial games, then you should have this hold true for some of the nastiest bugs (the graphics problems, like videocard compatability - the bane of MY existence for a few months). But most of the Open Source solutions out there haven't actually been used in any serious game projects yet, so they haven't reached that level of maturity. I *LOVE* having other people find and fix my game's bugs.
#3 - Community. This is something you WOULD get (to a certain degree) with a popular pure graphics-engine system, but once again you won't get the breadth of resources and suggestions for a game that doesn't have a common code base. There were some dismal points during the development of Void War when I was REALLY exhausted where I'd surf the web and have this vague dream that I'd stumble across some site where some random guy had written the code I needed to finish Void War, and left it there with an open license so I could just plug it in and be good to go.
Well, that kinda-sorta happens with Torque. There are a lot of great resources - some free, some available for a reasonable price (like the RTS pack and lighting pack, in my case - the RTS pack having very little to do with graphics-engine type modifications). While they are never EXACTLY what I need, they do provide me with some useful hooks and examples to shine a light on my path and make my job easier.
Do these answers make up for the fact that learning Torque can be almost as time-consuming as creating your own (albeit much simpler) game engine? Well, I'm banking on the answer being "yes," and so far it's being born out. In spite of the six-months or so I spent learning the system (and working on an aborted, smaller project) I feel like I've shaved many months off my dev cycle so far, and I'm seeing the possibility of shaving several more months off the END of development - which is where it matters the most to me. The end of the dev cycle is always the longest and most painful.
08/13/2005 (10:07 pm)
Ajari:Void War was done pretty much "from scratch" with no engine at ALL other than DirectX. I've worked with several custom 3D solutions in my career, so startng with nothing at all (or with just a 3D graphics engine) doesn't scare me much.
But it is enough of a pain in the butt that I switched over to Torque for my next game.
Why? The Void War engine *works* as it stands now. It COULD be re-purposed. What made my decision for me?
Three reasons:
#1 - there's a heck of a lot of work that goes into making a game engine (or re-purposing an engine) that has already gone into Torque. Things that you normally don't think of, things that aren't too sexy - like tools pipelines, input devices, UI functionality, dropped packets in multiplayer not screwing up game synchronization, handling configuration files & "mission files" (or *GASP* scripting!) for data-driven game development, and so forth. Being able to start with a stable and solid, already data-driven game engine means you get to jump right into the "fun stuff" - the actual game logic unique to your game.
#2 - A mature engine means a lot of the bugs have been ironed out already. If you have a graphics engine that has already been used in several commercial games, then you should have this hold true for some of the nastiest bugs (the graphics problems, like videocard compatability - the bane of MY existence for a few months). But most of the Open Source solutions out there haven't actually been used in any serious game projects yet, so they haven't reached that level of maturity. I *LOVE* having other people find and fix my game's bugs.
#3 - Community. This is something you WOULD get (to a certain degree) with a popular pure graphics-engine system, but once again you won't get the breadth of resources and suggestions for a game that doesn't have a common code base. There were some dismal points during the development of Void War when I was REALLY exhausted where I'd surf the web and have this vague dream that I'd stumble across some site where some random guy had written the code I needed to finish Void War, and left it there with an open license so I could just plug it in and be good to go.
Well, that kinda-sorta happens with Torque. There are a lot of great resources - some free, some available for a reasonable price (like the RTS pack and lighting pack, in my case - the RTS pack having very little to do with graphics-engine type modifications). While they are never EXACTLY what I need, they do provide me with some useful hooks and examples to shine a light on my path and make my job easier.
Do these answers make up for the fact that learning Torque can be almost as time-consuming as creating your own (albeit much simpler) game engine? Well, I'm banking on the answer being "yes," and so far it's being born out. In spite of the six-months or so I spent learning the system (and working on an aborted, smaller project) I feel like I've shaved many months off my dev cycle so far, and I'm seeing the possibility of shaving several more months off the END of development - which is where it matters the most to me. The end of the dev cycle is always the longest and most painful.
Torque Owner Unsung Zero
It is easier to find something you would be willing to work with (or around) when you know what you wouldn't be willing to work with (or around).