Sorry your browser is not supported!

Im trying to pick from all of these.

Im a bit confused as to which i should choose. But i\'ve already started this with DarkGDK..
http://forum.thegamecreators.com/?m=forum_view&t=162265&b=22
I just want to know what other people\'s opinion are about these engines/libraries/language?

(Which is the best thing to use?!)<-a more flame prone question..

So far i have this list of advantages and disadvantages that standout among others
+ advantage
-disadvantage

Natively without plug-ins::

dbPro
+easy to use and understand
+can use video animations
-slow arithmetics

DarkGDK
+C++ port for dbPro(has advantages of dbPro and C++ except for animations)

Ogre3D
+Fast shadow rendering
+Fast Fresnel,reflection,refraction,post process effects
-Poor support for 2D

Irrlicht
+Support many 3d model formats
+OpenGL,DirectX9,DirectX8,Software Rendering
+GUI support

OpenGL
+Direct LowLevel Access
+Platform independent

DirectX
+Direct LowLevel Access


Please add some things you can think of..

I also want to know which engine/library/language to use after i have my fair share of using DarkGDK.

But for the record, the greatness of a game mostly depends on the skills and the capabilities of the developer/artist/designer than the engine/language being used.

DarkGDK
-You have to try and learn two coding languages to be successful


I reckon, start with DBPro, move to DarkGDK then go into full C++. You would then have a large coding knowledge.

Why are you comparing languages to engines?

DBPro is a language.
Irrlicht is an engine.

It's crazy to compare them, because it depends on your experience with the actual language you'll use with these engines. It's a broken comparison. I can guarantee that using C++ with any of those engines will be far more complex than GDK.

So really you are comparing DBPro or GDK, to C++ with these engines, or other incarnations of Basic which tend to be more complex than DBPro as well.

IMO you should stick with GDK if you are getting results, because that will sharpen your C++ skills for going onto other engines if that's how it ends up. Usually, GDK or DBPro is fine for our projects here - there's little benefit in having an engine that can display millions of polygons, when you don't need them - I find development speed is just as important as engine speed, I can develop really quickly in DBPro so I use that more than anything.

You might want to add as a disadvantage for GL that ATi's support for GLSL is just hopeless. Stuff that works fine on nVidia cards and some Intel cards just doesn't on ATi systems because they couldn't be bothered to implement some small function. If you don't set a variable explicitly it's full of garbage. No support for GL_TEXTURE_RECTANGLE which is useful for screen space effects. The alternate, non-power-of-two, isn't supported by other ATi cards. It's a bloody gamble. My friend's ATi system would crash my app whenever a display list was compiled and a shader had been compiled. Even if the shader had nothing to do with the display list and was disabled. That's my experience anyway.

Yes, C++ and oh wait, nothing else. Of course you could use the .net version thus use C# or something.



+ easy/memorable syntax and identifiers
- slow compilation
- slow runtime speed
- too verbose; impedes development time and requires you remember a lot of very long-winded names
- terrible error handling and general lack of debugging ability/tools
- doesn't scale well for large projects due to a lack of objects/namespaces/libraries/proper scopes/bacon



+ easy to use compared to most other libraries that offer similar
+ due to (likely)using C++ it'll be way faster than DBPro in most scenarios
- compiled using retarded settings, and the wizard doesn't even properly set the project up to work with all of the standard library(each release seemingly randomly alternates between release and debug versions without providing both)
- poor documentation, it's literally a find/replace of the DBPro docs and fails to document vital information such as how its returned strings are stored
- it lacks usage of even basic and fundamental C/C++ principles such as namespaces, const-correctness, objects(by large) etc



+ cross platform
+ has third-party bindings for a ridiculous amount of libraries
+ is well-written and has many games under its belt, including several on steam
- is well-written, for C++, thus anyone not proficient in the language may get confused
- requires you understand quite a lot of concepts before you can really do much
* though those aren't really negatives of the library



+ cross platform
+ comes with lots of libraries as well as has bindings for others, but it's not like coming with them is much better than not
+ is well-written and I have no idea what's been made with it
- same as above I guess, but I have very little knowledge about it


+ you can't practically get much lower-level than them, thus you have access to everything without much abstraction
- lower-level, thus you have to abstract it to get any of the advantages of the above things
* there's not much point in using these directly as you'll end up wasting loads of time, there's plenty of libraries out there that are cross platform and wrap everything very nicely, so why reinvent the wheel? Especially since many of the libraries allow you to fork off them if you want some crap version of it

I'd personally recommend NuclearFusion, actually, but as it's not yet available -- what good is that recommendation?

http://nuclearglory.com/products/nf/

It comes down to one thing: what's your goal with the project?

Do you want to make your own engine from scratch, looking forward to some four years of development time or more, then DirectX and OpenGL are both good and bad in their own way. DirectX wins through on documentation, I'd say, but OpenGL has the (slight) advantage of multi-platform support. A good engine can swap its rendering pipeline, however, and shouldn't be dependent on anything else. But again -- years in the making.

Are you looking to write complete games with complex project structure and what amounts to thousands of lines of code? In that case, I'd recommend XNA Game Studio and C#. A lot better than the GDK in every single way. This is the level where you find ready-made engines (Irrlicht, Ogre, etc.), as well, but C# is a magical wonderland made of candy if you compare it to the swampy marshes of C++. Especially for a fledgling programmer.

Do you want to "test an idea" and have some knowledge of C++ and a few dollars to spend on plugins? In that case, the GDK is really neat and can be made a fairly complete package if you're willing to spend money on it. And learn how to get plugins to cooperate. And fight against your non-3D Studio MAX 3D program until it yields to your demands. (I've never succeeded with this myself.)

Finally, if all you desire is to "test an idea" that doesn't rely heavily on graphical power or optimization (or debugging capacity), DBP is the lightweight method. It's clumsy, as the commands are often ungainly, but it does the job and it does the job really fast.

That's difficult to answer without some type of context, so what exactly do you want to make ??

Its a wrong one, NF is out and working fine, infact i have it running using codeblocks right now, the basic version of the engine, Nuclear Basic is not out yet, but will be soon. I suggest NF as its extreamly powerfull and the developer is extreamly dedicated to his products.

Irrlicht is a lot less feature rich than Ogre3D. Irrlicht will suffice, but it's way of handling shaders is just not practical. Ogre3D is a little better, and would have my vote. DGDK is not, in my opinion, practical for commercial game development, but hey, if you are going to aim for commercial development you already know that. Nuclear Fusion has never caught my attention and still does not. If you want the best possible and you are capable of programming well in C++, I'd say go with OpenGL or Direct3D. You will have to learn Direct3D or OpenGL, and they are definitely nowhere as easy as DarkBasic or DGDK. I do, however, believe that they are worth it to learn.

If you want to write a low-level engine: yes. If you want to try out ideas: no.

Learning Direct3D or OpenGL is only the tip of the iceberg. There's always sound, GUI, game logic, scripting, animation, threading, asset management and a million other things that you must build on top of the renderer. All ready-made engines from Irrlicht to DBP or Unity already has this.

Assuming you want to write an engine, which isn't a very good idea. Everyone and their grandmother are trying to make engines. The word is overused beyond belief. And if you read, these two sentences complement each other:

It was a simple warning that if you want to use Direct3D or OpenGL then you must learn how to program with them first, and that task is not easy compared to basic. Of course, you will have to gather the other components as well though, and teach yourself how to use those.

No, you used a value term. You said:



I definitely know that I'd personally want the "best possible"!

I'm just saying that an "engine" is so much more than a renderer.

I dont really look at languages used bcoz i can already code a dark basic game in Ruby and Lua language using the applications i made.


Development speed is really important. Thanks for that. Im sticking with DarkGDK since im getting results and development is really fast even with the few time i spend on it.



Hmm..I want to make mind-blogging and heart warming game.
Some of my designs plans.

Learning DirectX and OpenGL does take some time and i can't even assure myself that i can make a quality renderer with them. So i guess, I'll be learning them only when i have lots of free time. But then again, it takes time away from my development time.
So using libraries and engines wrapping them is a better option.

Ogre is the fastest rendering engine i know. It can display numerous amount of 3d objects and high polygon models a time with less fps down. It supports shadows without the harsh fps penalty. But i has poor support for sprites management. Heck I dont know how to scale,rotate,flip or mirror a sprite in Ogre.But Ogre has the best support for shaders and post processing. Im not amazed why commercial quality games are made with them. I just had to say it.

First things first, make the best game for DarkGDK. Hehe. ^__^
I havent really seen a good DarkGDK Game since i started in the forums. I'll be making the best game for DGDK. ^_- Hehe

Good call on the DarkGDK.

When I first started looking for a game design app/engine, I looked them all over, before deciding which one to go with.

I tried Havoc (Awesome!, unfortunately only three parts of that multipart app are free (The physics, the content tools, and behavior tool). For the rest of the toolset, it's big $$ and there are also license fees if you plan on distributing any decent quantity of games.

Installed Unity as well... That was a bit sparse, and also a bit expensive. I have heard they redid their pricing, and it's much more affordable now. Haven't had time to look at it again.

Loaded XNA Game Studio for PC/Wii/Xbox 360. Nice tools, gorgeous for arcade, side-screen scrolling, Browser and UI type games. Not so good for RPG / 3d game development. Amazing price: Totally Free. Also you can sell games on the Xbox live XNA gameclub site in addition to direct selling Xbox games, but you only get points, which goes to buying other Xbox games. Not the ideal model for compensation.

Blender Stunning and Impressive Capabilities, Amazingly Free. The absolutely worst user interface design and vertex and point selection tools I have ever come across. It took me a whole week to build one 3d landscape, I could only repeat the process with the blessings of the Gods and copious notes, with at least one drinking binge thrown in, and because I missed a few points on the backside of one of my animated characters (the caterpillar), on my first animation try with this, the mesh deformed and the caterpiller detonated all over the gamespace. Sure, you can use this tool to design games, but only if you want your grandkids to inherit a half-completed game. They might finish it in their lifetime. I still use Blender, but only to import models which I then export into DirectX. I wouldn't dream of using this as a game engine.

Tried Irrlicht. Didn't like it. It didn't like my XP much either. Haven't even tried it with the Vista.

Working with Ogre3d was a total pain. I spent a week trying to get a working build of the game engine going on my XP, which was about six days longer than any of the other game engines I had experimented with. Found out later it wasn't working right because of the graphics card I had installed at the time. (the note on this were buried deep in the code). If I can't make games which will work on older machines, I'm not much interested. Only about 20% of the computers out there running right now are multi-core processer, advanced with graphics card & GPU's. While designing exclusively for that is awesome, I didn't think that cutting out 80% of my potential market was a really good idea. If I recall, I had a similar problem with Irrlicht, and Unity as well... supporting games for legacy systems.

I would design games with the PS3 SDK in a heartbeat, however the cost for that toolset is way beyond my meager hobbyist budget (It's like $25,000 a seat, yeah?).

I looked at Eclipse and JAVA - Also free, Really great for web based games, and turn based simulations not so hot for multiplayer 3d.

I also tried Python Free, very nice, actually, still have several versions on my system. Would have went with this and Visual C++ however, they are endlessly morphing the programming language, and the with every new codefix/release more older code gets broken requiring extensive rework of existing code. the Python 2.5 release almost brought me to tears when it broke my game Windows User Interface, and also took out the easy-to-use splash-screen and menu generation subroutines. I still use it now, just to test coding, and to solve some cross-platform issues (linux/Windows) but not much anymore. Mark Hammond pretty much wrapped the entire Win32 API and included that in Python so you can do pretty much anything on a Win32 system using Python. I don't think anyone has done that for the Win64 API though... And I'm using a 64 bit Vista now as my primary.

Visual C++ Free, Awesome!, Installed and in use. Used to support DarkGDK, and for creating dlls. Plenty of good tutorials, lots of code samples laying around for to rapidly create all kinds of useful tools. Not so much support for linux and mac OS's though. With Resources to support both DirectX and OpenGL games.

DarkGDK Free, Awesome! Installed and in use. This is my go-to choice for building games for market. With Visual C++ support and a solid library for supporting DirectX D3D games, building a working game is a snap, and the object oriented code routines are reuseable for other new games. Build it good once here, and you won't need to rebuild your game objects for a very long-long time and can concentrate on optimizing the performance of your game, or making better games.

DBPro Affordable, Awesome! I spend 70% of my time actually coding using DBPro becuase it's lightning fast to build prototype games, and test extensive game variants, and iterations.
This allows me to build an optimal game model that is then ported over to VisualC++/DarkGDK.

OpenGL and DirectX are two different languages and toolsets for supporting and manipulating graphics within a game. I believe (correct me if I'm wrong) OpenGL got its start with linux and DirectX was built originally for Windows. In my opionion OpenGL is faster, and DirectX more stable, with more graphics cards (especially legacy graphics cards) that will support it.

The various game engines and languages mentioned above support OpenGL, or DirectX, or both in some combination. Most of the current graphics cards that are being sold today support both OpenGL and DirectX.

DirectX is advancing with the impending release of DirectX11/12 and this is not compatible with the older versions of DirectX. Any games designed now using DirectX9 or 10 will need to be reworked (and in some cases you might not be able to rework it, and you might have to write new game code) to be useable on DirectX 11 and 12 systems. But hey, DirectX has the largest installed base of gamers at the moment.

OpenGL, on the other hand is just a part of a new open graphics standard OpenCL that is sponsored by a Computer Industry Roundtable/Group Khronos. OpenCL supports multiple graphics core processors, as well as multi-core CPUs. They are taking their speed rendering advantage in games to new heights to make games blindingly fast by taking advantage of all the processing power available on the newest machines. Collada is also being supported by the Khronos special interest group.

Collada (part of OpenCL, just like OpenGL is a standard digital exchange for models and artwork and the Developers kit includes code so you can wrap your 3ds models, and anyone who supports Collada can then use your models in their games or game engines. Collada maintains a public library of models and art for games, plus alot of advanced game designers automatically wrap their game models using Collada and trade with each other.

I think I would really prefer to use OPenGL (Fast, Fewer Users) for my games, but I'm currently using DirectX9/10 (Stable, Many Users).

Cheers!

Ah, I was under the impression we has talking renderers as he mentioned only two items which were not. Ogre3D, Irrlicht, Direct3D, and OpenGL are all renderers.

Well, in my opinion it is'nt always necessary to have the best graphics or top-engine, but rather good and fun gameplay. There are plenty of games that have breath-taking graphics, but eventually end up in the bin (sort of speak) due to poor gameplay and a low fun-factor. Look at al those oldschool games that so many try to remake with better graphics. Some of them are even worse then they where before (gameplay).

Speaking of good game engines, did anyone of you tried the Esenthel Engine?
It's free for non-commercial games.

It can't hurt, and in any case, we are also considering the flexibility and ease of use of the engines. If anything, graphical capability is so equal now it's not really worth mentioning. Just about every engine can read GLSL or HLSL and multitexture properly.

i read this article a while ago about why open gl is better than direct x, then i started to wonder how cool it would be if dark basic instead of direct x used open gl.

Also I'd take a look at the unreal development kit (UDK) which is free to download and has generous licencing terms.

I'd also like to recommend taking a look at UDK. It's free for non-commercial use, and it's quite an impressive engine. It's very flexible and the built-in scripting language is awesome. Check out some of this guys tutorials to get the idea of how powerful the engine is (although to be honest his tutorials only really scratch the surface of what it can do).

Ok, you've confused me. How do create DB game in Ruby? Would that not be a Ruby game then? And I thought Lua was just a scripting language?

It's my belief that a lot of people tend to glorify games from the 8-bit years simply because they were kids when they played them. So I personally write off much of this remake hysteria as nostalgia.



Now, Direct3D and OpenGL can be compared to each other. But OpenGL is ONLY a graphics library. DarkBASIC and the GDK are using so much more from DirectX, including sound, input, network and so on. Stuff that would have to come from somewhere else if they replaced their renderer with OpenGL technology. Or stuff that they'd still have to use, thereby removing one of the reasons to actually use OpenGL in the first place -- cross-platform support.

Anyway, you can find articles about the opposite subject if you just look. It's a very opinionated standpoint to call Direct3D or OpenGL "better" than the other. They're both good, only their top advantages differ them from each other.

Both Irr and Ogre should work fine on older systems (Irr certainly works perfectly with XP). Both of them may fumble if you're using GL on an older machine though (because old, cheap cards have crap support for GL) but DX (and even software in Irr's case) should work fine

Also, make sure you include the correct C runtime DLLs. If you build on Vista and try to run on XP, chances are XP doesn't have whatever CRT the binary is relying on

I would'nt say that, cause when it comes to playing a game, i'm still a kid.
And those days you also had good and bad games in terms of fun to play. There was no difference in graphics quality, cause you had only 8 or 16 different colors to play with.

I remember i had my first computer at the age of 14 (I'm now 41). It was a commodore 64 with floppy disks. Since the first day that i played my first game, i was so interested in programming and making games myself (i still am). I started with basic, assembly and a bit of machinecode.



Alot of company's try to remake those old games, simply because they where already good then. They're just trying to make them better in terms of graphics, sound, etc...
Sure, there's some nostalgy involved aswell, but that's not a bad thing. But it's also about making money, and sometimes that ruins the outcome of a game.

I remember playing Network Q RAC Rally Championship for the first time. The graphics where not that amazing (pixalized), but it had such a fun factor of playing, that 4 years later i still played it. Alot of rally games there after where graphically much better, but did'nt give me the same kick then Network Q RAC did.

Cheers

I made an interpreter that calls DarkGDK functions from Ruby and Lua.
I can make an entire game with Ruby and Lua using the interpreters i made.
I think it would be called a Ruby game. Hehe. But then again im using the dark basic functions so it wouldnt hurt to call it db game still.

It's all nostalgia, but nostalgia for the right reasons. Simple controls and gameplay, fun, engaging characters, unique features etc etc. We may have been kids when we played them, and we remember them above anything else in the last 20 years. Personally I do a few remakes, never exacting ones, but my idea is always to take a proven gameplay mechanic and put my own spin on it. I try to recreate the feel of the original game, as I remember it, not as it actually is thru emulation etc. Know what I mean?, Deathchase on the ZX Spectrum was one of my favorite games, and if you look at it these days you'd see a technically impressive game, but the graphics are bad, the game is repetitive and aiming is very difficult. But the game is awesome because you are riding a bike through a forest and your brain has little time to care about bad graphics that it can't see whizzing past, it's too busy trying to prevent a tree pizza.
Without these games pushing boundaries, we wouldn't have the same games as we do today. Without Turbo Esprit on the Spectrum we wouldn't have GTA. If GTA is a nostalgic remake of Turbo Esprit, then I'm all for nostalgia.

Rose tinted glasses would be a factor for some. However, the notion that something is no longer good, just because it's older, is clearly false..

Yes they have...Unity3D Indie is now free, you only have to pay for Unity3D Pro.

Not what I'm saying. I'm just saying that I write down the latest few years of remake-hysteria to nostalgia.

Some of it is definitely warranted, but making older titles godly and trying to mimic what has already been done to death is nothing more than a desperate struggle against time.

Personally, I'm all for innovation. I'm all for NEW games, even if they're bad. New IPs, new ways of using old concepts, etc.

For example, I think the Metroid Prime game is the best one in the series. It uses a first-person perspective, but does its own things with it, while still being true to a very old franchise. It's some kind of remake/innovation/reimagining hybrid. But that is typically not the case.

A desperate struggle against time?

So how come the DS does so well off it then? - the biggest selling handheld is desperately struggling against time and making Nintendo an absolute fortune in the process. Sit and disagree then ask yourself how many people play truly original 'indi' games? - most indi games are based on something, some old puzzle game from the 8 or 16-bit era. The most original aspect of Metroid Prime, rolling around and and sticking to walls, was in the original NES game. There truly is nothing new in videogames that wasn't in the tens of thousands of games before them.

That's a great attitude to have, abandon convention, lock up your rule book, put down the pen because the Simpsons already did that .

People like retro gaming - there's no need to copy games, but being inspired by a game is something completely different.

I mean, what is the most original aspect of a game you've seen in the last year?

Personally I can't think of a single example of originality this year. Actually I would say Scribblenaughts on the DS was pretty original, you just type the name of something and it appears, then try and solve logic platformer puzzle with them. That was a very original game, although most people probably wouldn't agree with that.

I think I'm not doing so well with these types of sweeping arguments, am I? But really -- all I'm saying is that I'm not too fond of the retro-evangelism that pervades much of today's game industry. There is so much territory left unexplored, I believe, and especially in the types of genre-blending efforts that occasionally crop up. Games that try to step away from the overused abbreviations of our hobby (FPS, RTS, RPG, etc.) and do something that kind of blends them into one cohesive experience.

It just saddens me that games can be complete rip-offs and get credit for it rather than having people saying "it's OK" and move on. Or the fact that a game can have five or six sequels and still be more popular than God.

And the fact that there are hardly any innovations in the industry overall isn't really an argument against me. I'm just saying what I prefer. The fact that I don't GET what I prefer isn't really evidence in itself.

I'll go try it.

Most ideas arent copyrighted. So its fine to imitate some game aspects. Speaking of retrogaming, I've played with good old rpgs lately in different consoles. They made an exact copy for a different console. I wonder why?

If its the same game with the same graphics. Why make another game in another console 3-5 years after?