What do you want in a game engine
by Mike Kuklinski · in General Discussion · 11/01/2007 (11:28 pm) · 10 replies
For the purposes of collecting information...
What would be what YOU would want in a game engine? What features would you want the engine to have? What would you have it do? Be creative! Be abstract, also: I am not asking for an answer such as "I want to see SM99 supported on the Geforce QAOSFNAKLSdnkl", I am looking for an answer along the lines of "I feel as though the scripting engine in Torque is perfected, particularly in how it handles arrays", or as such.
What would be what YOU would want in a game engine? What features would you want the engine to have? What would you have it do? Be creative! Be abstract, also: I am not asking for an answer such as "I want to see SM99 supported on the Geforce QAOSFNAKLSdnkl", I am looking for an answer along the lines of "I feel as though the scripting engine in Torque is perfected, particularly in how it handles arrays", or as such.
About the author
http://dev.stackheap.com/
#2
You want to buy your game engine from Staples???
11/02/2007 (8:09 am)
Quote:I want the Big Red Button.
You want to buy your game engine from Staples???
#3
Short answer is - I'll take your engine, version 3.1. :-)
11/02/2007 (12:36 pm)
I want a game engine that lets me rapidly prototype. It is not a feature set that achieves that - it's a tool set and engine code that is written with work flow in mind. There is probably no way to design this up front and get it 100% right. It will probably take a back-and-forth between the engine and the users of the engine.Short answer is - I'll take your engine, version 3.1. :-)
#4
11/02/2007 (12:46 pm)
Complete documents that are updated when changes are made.
#5
I don't see why it has some of the dopy naming conventions that it does.
I feel the datablocks are implemented in a silly way, as well. It would be easy to have static and nonstatic methods and automatically transport the correct ones. This implementation is way too exposed and in your face when there is no need for it to be something anyone ever thinks of.
So, I wish we could easily switch out scripting languages for one.
I also wish that BASIC documentation such as how to compile without searching for hours on end...such as what file do I compile? Where does it build things? These are not obvious in the least and from the forums I am not the only one who was puzzled. Once you know how it works it's easy, so that means the documentation is even more lacking.
To me, having the demo and the engine linked together like that is just weird and kinda confusing, too.
Having that as an example is fine but starting by editing on it is just kinda weird and not what people expect at all.
I want it easy to do easy things and possible to do hard things.
I'd like to see some more camera code put in as example stuff so that things like 3rd person overhead view and a few others are basically precoded and you can just adjust them.
11/05/2007 (10:18 am)
Well...I feel torquescript is lacking in some ways. I know it is not supposed to be a torque critique and I am also new to gaming engines in general but....I don't see why it has some of the dopy naming conventions that it does.
I feel the datablocks are implemented in a silly way, as well. It would be easy to have static and nonstatic methods and automatically transport the correct ones. This implementation is way too exposed and in your face when there is no need for it to be something anyone ever thinks of.
So, I wish we could easily switch out scripting languages for one.
I also wish that BASIC documentation such as how to compile without searching for hours on end...such as what file do I compile? Where does it build things? These are not obvious in the least and from the forums I am not the only one who was puzzled. Once you know how it works it's easy, so that means the documentation is even more lacking.
To me, having the demo and the engine linked together like that is just weird and kinda confusing, too.
Having that as an example is fine but starting by editing on it is just kinda weird and not what people expect at all.
I want it easy to do easy things and possible to do hard things.
I'd like to see some more camera code put in as example stuff so that things like 3rd person overhead view and a few others are basically precoded and you can just adjust them.
#6
But, in Torque 2 you will (maybe) be able to implement yourself other scripting languages, so everyone could be happy in the end.
11/05/2007 (11:12 am)
Quote:I feel torquescript is lacking in some ways.Well, there's a lot of people here doing great with torquescript. I'm pretty satisfied by it myself, and I don't want to switch for something else just now (I'm speaking about TGB, maybe the TGE version is a bit different).
But, in Torque 2 you will (maybe) be able to implement yourself other scripting languages, so everyone could be happy in the end.
#7
Regardless if it's TS or any other type of language, I'm pretty sure we are going to need to learn new ways of coding as the system is so much different in T2. With that being said, I hope they allow all the people out there who like to implement their own script set the chance to do so while they are still in beta. That way, when the product is released, people will have more options on how they want to go about writing their code.
Just my opinion
11/05/2007 (11:35 am)
There are alternatives to Torquesscript out there already. Like Ben G (wasn't trying for a play of words) says, Torque 2 could have other alternatives considering the component type model.Regardless if it's TS or any other type of language, I'm pretty sure we are going to need to learn new ways of coding as the system is so much different in T2. With that being said, I hope they allow all the people out there who like to implement their own script set the chance to do so while they are still in beta. That way, when the product is released, people will have more options on how they want to go about writing their code.
Just my opinion
#8
be to expensive for an indie, and wouldn't be nearly as much fun.
2. I do want an engine that allows me to create a prototype quickly. Say, within a month
or two depending on the complexity of the prototype.
3. I do want an engine that can be modified with relative ease. (that relative part being
in relation to ones programming expertise, but does not imply that if I'm really really good
that I should have to pick through someone else's spaghetti code.)
4. I want an engine that has been optimized for performance. There's nothing that kills a
game faster than an engine that bogs down the system. 20 frames per second would be
a nice minimum rate to maintain no matter what is happeneing in the scene. (Rocket Bowl
drops to it's minimum graphics settings on my machine and it's impossible to play the
challenges. It's so choppy that I can't aim the ball at all before the timer expires.
Dark Horizons LORE is the same way. By the time I've aimed at an enemy they've already blown
me away while I'm waiting for the mouse movements to catch up, and my machine is actually
pretty up to date. 2.8Ghz, 1gb RAM, GeForce 6800.)
5. I want an engine with very robust networking. Capable of handling hundreds, if not thousands,
of players at once.
6. I'd like an engine that has continuous flow terrain. I want to walk the entire planet without
having to stop and load up the next chunk of terrain. However, I would also like to be able to
have a mission without a terrain. (space missions)
Some of these things I know are dreams that may be unattainable for an engine company that's
catering to the indie community. There are several things I didn't list, but I suspect with the
modularity of Torque2 we will see several 3rd parties come out with components that fill the needs.
From everything that I've heard so far, I want Torque 2. If it doesn't do this or that out of the box
it's going to be much easier to write a component that does do it, than it would have ever been
to write the same capability into Torque 1.
11/05/2007 (12:06 pm)
1. I don't want an engine that does it all for me. That would take to long to develop, wouldbe to expensive for an indie, and wouldn't be nearly as much fun.
2. I do want an engine that allows me to create a prototype quickly. Say, within a month
or two depending on the complexity of the prototype.
3. I do want an engine that can be modified with relative ease. (that relative part being
in relation to ones programming expertise, but does not imply that if I'm really really good
that I should have to pick through someone else's spaghetti code.)
4. I want an engine that has been optimized for performance. There's nothing that kills a
game faster than an engine that bogs down the system. 20 frames per second would be
a nice minimum rate to maintain no matter what is happeneing in the scene. (Rocket Bowl
drops to it's minimum graphics settings on my machine and it's impossible to play the
challenges. It's so choppy that I can't aim the ball at all before the timer expires.
Dark Horizons LORE is the same way. By the time I've aimed at an enemy they've already blown
me away while I'm waiting for the mouse movements to catch up, and my machine is actually
pretty up to date. 2.8Ghz, 1gb RAM, GeForce 6800.)
5. I want an engine with very robust networking. Capable of handling hundreds, if not thousands,
of players at once.
6. I'd like an engine that has continuous flow terrain. I want to walk the entire planet without
having to stop and load up the next chunk of terrain. However, I would also like to be able to
have a mission without a terrain. (space missions)
Some of these things I know are dreams that may be unattainable for an engine company that's
catering to the indie community. There are several things I didn't list, but I suspect with the
modularity of Torque2 we will see several 3rd parties come out with components that fill the needs.
From everything that I've heard so far, I want Torque 2. If it doesn't do this or that out of the box
it's going to be much easier to write a component that does do it, than it would have ever been
to write the same capability into Torque 1.
#9
Torque 2 sounds like a much cleaner design overall. I wonder how the scripting will be handled.
11/05/2007 (3:18 pm)
Ben, the accessibility to the engine is good and I have no complaints there so far, but just the basics of the language itself are a little lacking. There's no reason not to have support for the basics of other languages that can be similarly precompiled with simple API interfaces for them to use.Torque 2 sounds like a much cleaner design overall. I wonder how the scripting will be handled.
#10
-Stability. I want it to do what it says on the tin, no or very little "coming soon" features please.
-Features that cater for modern development. Shaders, large terrain, good collision sys, physics etc.
Overall just a good, well made engine that is capable of decent looking games by todays standards.
11/05/2007 (3:46 pm)
-Good documentation, that is up to date and without holes or gaps.-Stability. I want it to do what it says on the tin, no or very little "coming soon" features please.
-Features that cater for modern development. Shaders, large terrain, good collision sys, physics etc.
Overall just a good, well made engine that is capable of decent looking games by todays standards.
Torque Owner DragonSix
Apart from that, I guess that an engine that would look likes TGB but that support both 2d and 3d and WITHOUT any bug would be already perfect for me.