3d gamestudio user considering Torque
by Matthew Hoesterey · in Torque Game Engine · 09/12/2004 (8:58 am) · 11 replies
Hello. I'm currently using A6 and was wondering what if any advantages their were to switching engines. Does torque allow for higher poly counts? Better features? Normal maps? physics? dynamic lighting with stenciled shadows?
Honestly I'm not overly impressed with torque engine screen shots and am wondering if the current screens are the limit of its capabilities.
Honestly I'm not overly impressed with torque engine screen shots and am wondering if the current screens are the limit of its capabilities.
#2
It's nearly impossible for me to help you without in-depth knowledge of your project, where you're at in development, and your team's dedication and experience. If the answer is "my team consists of me, I have no programming experience but I'm learning C-Script, and I'm using MED, WED, and the built-in script editor for all of my current project's development", then I'd recommend staying with A6 and expanding your experience base. If the answer is "I have two knowledgable C++ programmers, a modeler who uses GameSpace and a level designer that uses QuArK and WED and Radiant interchangably", then I'd say "download the demos and see which engine fits you better".
Ask yourself a couple of questions:
Does your gametype fit well within stock A6 or Torque?
Will you have to make major modifications to either engine for key features?
What are the limitations of your team? (this is extremely important question)
How far along are you in development? If you're going to have to completely re-recreate art assets PLUS programming, then you should stick with A6 for this project). If you're in the design stages, then you'll need to evaluate the engines. Remember to look at games created in the engines with a critical eye. If you look at what a good team has done, don't automatically assume that you can do it with minimal effort. It may have taken them six months for 8-12 hour days to make the modifications to give them a professional edge.
That said, one of the biggest things to realize when moving from A6 to Torque is your workflow change. If you use the included tools with A6, then you'll notice a marked difference in that you'll be using third-party programs. If you're a Milkshape (or Lightwave, Maya, 3DS Max) user, then you'll already be a step ahead of the modeling curve. If you use MED, then you'll have the learning curve of using another application be it Blender, Milkshape, GameSpace, or whatever you desire. WED is one of my favorite CSG editors ever, and the move to QuArK will most likely throw you for a major loop at first. Once you figure out the QuArK workflow, you'll be fine, though.
If your game is network oriented, then you'll be extremely pleased at Torque's capabilities, even when compared to the insanely priced professional version. Tokamak and Novodex physics can be incorporated (windows only), and there are people working on ODE integration (Windows, Linux, Mac). If your game is physics heavy (Half-Life 2, PainkilleR, PSI-Ops), then you'll definitely want to look at integrating one of these products into either Torque or A6. The physics model is nice in A6, but it doesn't offer even half the response and ability of these dedicated physics libraries. The physics engine is the only reason to consider the professional price over commercial in A6, IMO.
Another key consideration is your target audience. If you're looking for cross-platform distribution, then A6 is not an option at all. If you don't care about targeting multiple OS's, then one of Torque's key features won't matter to you.
The fundamental difference between the type of engine comes down to Torque being a source engine and A6 being a modification engine. You can change anything (for good or bad) in Torque. You can only change the areas that Conitec allows you to change in A6 (though very creative people have found some interesting "work-arounds" to make the SDK to unintended things with the A6 engine).
I don't know if that helps or muddles things more.
09/12/2004 (10:18 am)
TGE and TSE can be modified to look and act like Doom 3 or Far Cry as long as you have a team of dedicated programmers. Much like how 90% of the games using A6 look like extremly rigid Tomb Raider clones, and yet there are that 10% of extremly high quality titles. Both come down to the dedication of the team and the members of the team that you pull on board. Both engines are capable of doing amazing things as long as you put together a team to make it happen. Both can look like complete and utter crap as well.It's nearly impossible for me to help you without in-depth knowledge of your project, where you're at in development, and your team's dedication and experience. If the answer is "my team consists of me, I have no programming experience but I'm learning C-Script, and I'm using MED, WED, and the built-in script editor for all of my current project's development", then I'd recommend staying with A6 and expanding your experience base. If the answer is "I have two knowledgable C++ programmers, a modeler who uses GameSpace and a level designer that uses QuArK and WED and Radiant interchangably", then I'd say "download the demos and see which engine fits you better".
Ask yourself a couple of questions:
Does your gametype fit well within stock A6 or Torque?
Will you have to make major modifications to either engine for key features?
What are the limitations of your team? (this is extremely important question)
How far along are you in development? If you're going to have to completely re-recreate art assets PLUS programming, then you should stick with A6 for this project). If you're in the design stages, then you'll need to evaluate the engines. Remember to look at games created in the engines with a critical eye. If you look at what a good team has done, don't automatically assume that you can do it with minimal effort. It may have taken them six months for 8-12 hour days to make the modifications to give them a professional edge.
That said, one of the biggest things to realize when moving from A6 to Torque is your workflow change. If you use the included tools with A6, then you'll notice a marked difference in that you'll be using third-party programs. If you're a Milkshape (or Lightwave, Maya, 3DS Max) user, then you'll already be a step ahead of the modeling curve. If you use MED, then you'll have the learning curve of using another application be it Blender, Milkshape, GameSpace, or whatever you desire. WED is one of my favorite CSG editors ever, and the move to QuArK will most likely throw you for a major loop at first. Once you figure out the QuArK workflow, you'll be fine, though.
If your game is network oriented, then you'll be extremely pleased at Torque's capabilities, even when compared to the insanely priced professional version. Tokamak and Novodex physics can be incorporated (windows only), and there are people working on ODE integration (Windows, Linux, Mac). If your game is physics heavy (Half-Life 2, PainkilleR, PSI-Ops), then you'll definitely want to look at integrating one of these products into either Torque or A6. The physics model is nice in A6, but it doesn't offer even half the response and ability of these dedicated physics libraries. The physics engine is the only reason to consider the professional price over commercial in A6, IMO.
Another key consideration is your target audience. If you're looking for cross-platform distribution, then A6 is not an option at all. If you don't care about targeting multiple OS's, then one of Torque's key features won't matter to you.
The fundamental difference between the type of engine comes down to Torque being a source engine and A6 being a modification engine. You can change anything (for good or bad) in Torque. You can only change the areas that Conitec allows you to change in A6 (though very creative people have found some interesting "work-arounds" to make the SDK to unintended things with the A6 engine).
I don't know if that helps or muddles things more.
#3
Torque also has vehicle physics already integrated into the engine, as well as rigid body physics.
Here's another topic about A6 vs Torque, and here's another one.
Those links should help you better decide.
-Jase
09/12/2004 (10:41 am)
On the subject of using A6 or Torque in its stock form, you are better off using Torque. A6's template scripts are complete utter crap. Torque takes a bit more to get something up and going using only its scripting language, but the results are far better.Torque also has vehicle physics already integrated into the engine, as well as rigid body physics.
Here's another topic about A6 vs Torque, and here's another one.
Those links should help you better decide.
-Jase
#4
I've seen this question come up enough times now that I can't resist answering for once. I'm probably biased, but below is my honest, frank opinion.
Bottom-line for me is this: if you're looking to create a professional game that you can actually make money with, Torque is the only way to go.
If you're serious about creating a real, marketable game, I don't think there's much comparison between the two engines. Especially if you need multiplayer gameplay-- only the hundred-thousand+ dollar engines get close to Torque's networking.
A6 has been used on some successful games, but I think anybody comparing the two engines side-by-side would have to agree that Torque is far, far more likely to help breed success.
Try to compare the two engines. Taking info straight off of A6's website, check out how limited their $89 (closest to Torque's price) engine is:
-No networking whatsoever.
-No dynamic shadows whatsoever.
-No bone animation whatsoever
-No physics whatsoever.
-limited, 16-bit static shadows; no mirroring; no customizable start-up / menu screen; etc; etc; etc
Those are huge holes. Is this thing even in Torque's ballpark?
What happens if you pay double for their next step up? The $199 version of A6 still has severely restricted networking (only 4 players in a game), and severely restricted physics (only 1 object can have physics).
To get full networking and physics support, you have to buy their most expensive product, which weighs in at a hefty $899.
That's *nine times* the cost of Torque, but it still doesn't stack up. Don't forget all the stuff not listed on the A6 site. Torque is deeply multiplatform, which means you can market your game on Windows, Mac, Linux... and even XBox in the future. This multi-platform support is crucial to the success of many games-- Mac sales alone sometimes account for more than half of our game sales.
Next, compare Torque's terrain engine to A6's. Compare the animation systems. Compare the water engine, particle effects, GUI tools, etc, etc. Observe their notes on limited support, as compared to what happens here. And again, there's no way their networking is even in the same ballpark as Torque's.
To seal the deal... don't forget that you get all the source code for Torque. Even if you never change a line of C++ code, just being able to investigate and learn from the codebase is invaluable for many people.
Of course, the conditionals David brings up in his post above are important to consider. If you're starting to learn basic game development skills, you may not worry about having all of Torque's power.
Some engines are easier to learn than Torque. That fact may never go away completely. Still, the difference in learning curves is shrinking at a faster and faster rate lately, and we're working to make Torque as welcoming to developers of all experience levels as *any* engine out there.
Even if you are simply trying to learn, Torque can give you better experience and, ultimately, more professionally useful knowledge than engines like A6. It may take longer to get started, but in the end you'll get a lot more out of working with the full source of a AAA game engine that has been continuously, dedicatedly updated and enhanced for years.
A6's tools are easy to use and the basics are easy to learn, but the engine itself is a handicap. As an experienced engineer who's serious about game development, that is a deal-breaker for me.
Most importantly though, once you do get your footing in Torque, you won't hit the unfortunate limitations you inevitably would in A6 (and other engines in a similar price range, especially those which don't offer their full source code).
To me, it's not hard to decide which way to go.
Sorry to type so much. :) I just like to see good, enthusiastic game developers get tools that are most likely to help them make games for a living.
09/12/2004 (1:04 pm)
It's cool that people give pretty fair analyses of the engines even here. Good stuff.I've seen this question come up enough times now that I can't resist answering for once. I'm probably biased, but below is my honest, frank opinion.
Bottom-line for me is this: if you're looking to create a professional game that you can actually make money with, Torque is the only way to go.
If you're serious about creating a real, marketable game, I don't think there's much comparison between the two engines. Especially if you need multiplayer gameplay-- only the hundred-thousand+ dollar engines get close to Torque's networking.
A6 has been used on some successful games, but I think anybody comparing the two engines side-by-side would have to agree that Torque is far, far more likely to help breed success.
Try to compare the two engines. Taking info straight off of A6's website, check out how limited their $89 (closest to Torque's price) engine is:
-No networking whatsoever.
-No dynamic shadows whatsoever.
-No bone animation whatsoever
-No physics whatsoever.
-limited, 16-bit static shadows; no mirroring; no customizable start-up / menu screen; etc; etc; etc
Those are huge holes. Is this thing even in Torque's ballpark?
What happens if you pay double for their next step up? The $199 version of A6 still has severely restricted networking (only 4 players in a game), and severely restricted physics (only 1 object can have physics).
To get full networking and physics support, you have to buy their most expensive product, which weighs in at a hefty $899.
That's *nine times* the cost of Torque, but it still doesn't stack up. Don't forget all the stuff not listed on the A6 site. Torque is deeply multiplatform, which means you can market your game on Windows, Mac, Linux... and even XBox in the future. This multi-platform support is crucial to the success of many games-- Mac sales alone sometimes account for more than half of our game sales.
Next, compare Torque's terrain engine to A6's. Compare the animation systems. Compare the water engine, particle effects, GUI tools, etc, etc. Observe their notes on limited support, as compared to what happens here. And again, there's no way their networking is even in the same ballpark as Torque's.
To seal the deal... don't forget that you get all the source code for Torque. Even if you never change a line of C++ code, just being able to investigate and learn from the codebase is invaluable for many people.
Of course, the conditionals David brings up in his post above are important to consider. If you're starting to learn basic game development skills, you may not worry about having all of Torque's power.
Some engines are easier to learn than Torque. That fact may never go away completely. Still, the difference in learning curves is shrinking at a faster and faster rate lately, and we're working to make Torque as welcoming to developers of all experience levels as *any* engine out there.
Even if you are simply trying to learn, Torque can give you better experience and, ultimately, more professionally useful knowledge than engines like A6. It may take longer to get started, but in the end you'll get a lot more out of working with the full source of a AAA game engine that has been continuously, dedicatedly updated and enhanced for years.
A6's tools are easy to use and the basics are easy to learn, but the engine itself is a handicap. As an experienced engineer who's serious about game development, that is a deal-breaker for me.
Most importantly though, once you do get your footing in Torque, you won't hit the unfortunate limitations you inevitably would in A6 (and other engines in a similar price range, especially those which don't offer their full source code).
To me, it's not hard to decide which way to go.
Sorry to type so much. :) I just like to see good, enthusiastic game developers get tools that are most likely to help them make games for a living.
#5
not to mention a6 is not just 100 usd. tge is.
cheers.
09/12/2004 (1:22 pm)
A6 lacks documentation. and simple scripts old enough to love my granny doesn't count it. gg has tons of documentation (i need glasses by now after reading so much and i switched 2 ink cartridges on my printer). with tge you can make games, although without shiny features or dinamics. i had to quit a6 right after i managet to implement newton dinamics in it onlly because i found it impossible to learn forward.not to mention a6 is not just 100 usd. tge is.
cheers.
#6
09/12/2004 (1:28 pm)
I have been using A5, A6 for about three years and Torque for about a year. The A6 have lot better documation to get start. You have to reseach Torque alot to get start, I like torque but you can get start with A6 alot faster. Plus A6 have more AI, They come with A6. Torque AI, you better be ready do some programmer. A6 is good for games that take place inside. If you have Bryce 5, it is very easy to build Terrain for A6. If Torque came with better AI than would be different. Read the article in gamedeveloper September 2004. It much easy to set up game in A6. Plus unless you have hours go throw form. I would stick with A6, nothing against Torque.
#7
And you don't want to bottom out a month before you're ready to ship.
Sometimes the investment in time to learn how to use a more complex tool is worthwhile.
09/12/2004 (2:40 pm)
A6 is easier to start with. Torque lets you go a lot deeper.And you don't want to bottom out a month before you're ready to ship.
Sometimes the investment in time to learn how to use a more complex tool is worthwhile.
#8
The Josh and Ben whipped in and offered better commentary in less words (well, maybe no Josh in terms of word count, but definitely in terms of content ;-) ).
I think that a point that people often get caught up on is seeing what other people do with an engine and automatically assuming that they can do the same thing. Some of the people who have been using Dark Basic Pro for a long time have made it do some extremely odd things that the creators never expected (such as a clone of Final Fantasy VI, though it had often been stated by the designers that 2D functionality was extremely limited because DBPro is a 3D engine). But just because there are programmers than can create the exact type of game you want to make doesn't mean that you have the knowhow to do it. It simply means that it can be done. Torque can be used to make just about anything on the planet as long as you have Melv on your team. :-) I mean, as long as you have good people (like Melv) on your team who are knowledgable about the internals of the engine.
It's like when someone who is new to 3D modeling brings up the "which package is best" and start comparing screenshots from different packages from people with 10+ years of experience (though often the images will be one 10 year veteran in package X and the other one being a second week modeler from the forum who just learned to use the material system and say something exciting like "nuff sayed"). Then they get frustrated when people who have used the packages for a significant amount of time say "download the demo and see which one speaks to you more clearly." While that's the best advice you can give to someone who is new and interested, it makes them frustrated because they want an easy answer that no one can give them. They want to know what they can do with the package. People want the same answer from Blitz 3D, Torque, and A6. While people here at GG are going to say "dude, buy Torque" and people on the Conitec forums will say "dude, buy A6", the answer really comes down to what the developer asking the question is willing to sacrifice in terms of time and learning.
@Michael Prewer
I agree that Torque's main issue for new developers is the learning curve. I think that stems from its roots as a professional engine created for a team rather than a general purpose engine. This is one of the key areas that GG has been working on. Torque is many times more accessible now than when I first started here. There's still a ways to go, but at least it's a realizable goal and one that GG are working towards.
The AI in A6 was extremely basic when I played with it. Much the same stuff I would have had to do in C-Script I'd have to replicate in Torque using TorqueScript. I'm extremely excited about the AI pack that's being worked on for Torque because it is miles ahead of what both stock Torque and A6 have to offer (and my own knowledge of AI programming).
09/12/2004 (4:28 pm)
I ran up against the 4096 posting limitation, and like usual decided to edit rather than post two large posts that basically said the same thing in many more words than needed.The Josh and Ben whipped in and offered better commentary in less words (well, maybe no Josh in terms of word count, but definitely in terms of content ;-) ).
I think that a point that people often get caught up on is seeing what other people do with an engine and automatically assuming that they can do the same thing. Some of the people who have been using Dark Basic Pro for a long time have made it do some extremely odd things that the creators never expected (such as a clone of Final Fantasy VI, though it had often been stated by the designers that 2D functionality was extremely limited because DBPro is a 3D engine). But just because there are programmers than can create the exact type of game you want to make doesn't mean that you have the knowhow to do it. It simply means that it can be done. Torque can be used to make just about anything on the planet as long as you have Melv on your team. :-) I mean, as long as you have good people (like Melv) on your team who are knowledgable about the internals of the engine.
It's like when someone who is new to 3D modeling brings up the "which package is best" and start comparing screenshots from different packages from people with 10+ years of experience (though often the images will be one 10 year veteran in package X and the other one being a second week modeler from the forum who just learned to use the material system and say something exciting like "nuff sayed"). Then they get frustrated when people who have used the packages for a significant amount of time say "download the demo and see which one speaks to you more clearly." While that's the best advice you can give to someone who is new and interested, it makes them frustrated because they want an easy answer that no one can give them. They want to know what they can do with the package. People want the same answer from Blitz 3D, Torque, and A6. While people here at GG are going to say "dude, buy Torque" and people on the Conitec forums will say "dude, buy A6", the answer really comes down to what the developer asking the question is willing to sacrifice in terms of time and learning.
@Michael Prewer
I agree that Torque's main issue for new developers is the learning curve. I think that stems from its roots as a professional engine created for a team rather than a general purpose engine. This is one of the key areas that GG has been working on. Torque is many times more accessible now than when I first started here. There's still a ways to go, but at least it's a realizable goal and one that GG are working towards.
The AI in A6 was extremely basic when I played with it. Much the same stuff I would have had to do in C-Script I'd have to replicate in Torque using TorqueScript. I'm extremely excited about the AI pack that's being worked on for Torque because it is miles ahead of what both stock Torque and A6 have to offer (and my own knowledge of AI programming).
#9
Actually A6 had a great number of walkthroughs for all kinds of things that were easy to setup. Everything from camera control, particle effects, day & night effects, AI, ect. But Torque presents much more opportunity(thats the key) for a lower price with the only overhead is that its a bit more complicated.
-Jase
09/12/2004 (9:28 pm)
Quote:A6 lacks documentation.
Actually A6 had a great number of walkthroughs for all kinds of things that were easy to setup. Everything from camera control, particle effects, day & night effects, AI, ect. But Torque presents much more opportunity(thats the key) for a lower price with the only overhead is that its a bit more complicated.
-Jase
#10
I'm still in the design phase of my project and am shopping around. I own an older version of a6 and was planing on upgrading but thought I should check out torque.
Just to answer David's questions about my team's dedication and goals.
I will be graduating soon and would like to design a prototype in the hopes of finding a publisher to fund production on a full title. I know C++ though I am much better at 3d animation. I use 3d studio max and Maya depending on the project. (I like 3ds better).
As far as modifying the engine that is beyond me. I am capable of programming Kick a*# AI ect... but want to stay away from modifying the engine its self. I'm not really invested in A6 though I must say I love the customer and user support. I must say I'm pleasantly surprised with the number of responses I received and will be seriously considering torque. I simply want the engine that is "best" for making a prototype of an RPG with a real time fighting system that would be worthy of approaching a producer with and getting funding to buy rights to a commercial engine and to fund development. Thanks again.
My current 3d work in progress (high poly)
I may have rambled sorry if I did. Just got done with a 12 hour shift :(
[img]http://www.rit.edu/~mrh9661/GargoyalTailPersp.jpg[/img]
09/12/2004 (9:39 pm)
Thanks for the feedback :)I'm still in the design phase of my project and am shopping around. I own an older version of a6 and was planing on upgrading but thought I should check out torque.
Just to answer David's questions about my team's dedication and goals.
I will be graduating soon and would like to design a prototype in the hopes of finding a publisher to fund production on a full title. I know C++ though I am much better at 3d animation. I use 3d studio max and Maya depending on the project. (I like 3ds better).
As far as modifying the engine that is beyond me. I am capable of programming Kick a*# AI ect... but want to stay away from modifying the engine its self. I'm not really invested in A6 though I must say I love the customer and user support. I must say I'm pleasantly surprised with the number of responses I received and will be seriously considering torque. I simply want the engine that is "best" for making a prototype of an RPG with a real time fighting system that would be worthy of approaching a producer with and getting funding to buy rights to a commercial engine and to fund development. Thanks again.
My current 3d work in progress (high poly)
I may have rambled sorry if I did. Just got done with a 12 hour shift :(
[img]http://www.rit.edu/~mrh9661/GargoyalTailPersp.jpg[/img]
#11
Live and learn, I'm happier here...
:P
Nick
09/13/2004 (5:43 am)
Just my two cents... I own A5 (but I've forgotten my password so I haven't updated or patched it in years :P ) I never use it any more. I bought Torque cos the demo impressed me much more than the capabilities of A5. I also like the community here much better then the community over at conitec, where people yell at each other a lot. I've played around a lot with both, and obviously there are learning curves on both, but I think what you get out of torque is much much better and it was half the price I paid for A5.Live and learn, I'm happier here...
:P
Nick
Torque Owner Kyle Goodwin