Aztec:

Well, I do believe that is what all the articles, samples, tutorials and these forums are for.
Compared to having to program and test an engine for each game you make from scratch, xna seriously  allows for short term creation of your game ideas.
Then you also include the fact that whatever you write can run on xbox360, windows and now zune and you really can't beat xna for quick development.

Well, I didn't see any article or tutorial about physics, AI, animations or LOD multitextured terrains. So, what we have are really the samples, but we have to spend some time figuring out how they work and how to make them to working together. Not all samples works together! For example, try to add the chase camera to the tank animated collision sample. Now that sample doesn't have LOD or multitexturing, so try to replace the terrain with a model ( to use a better texture and some LOD ). It won't work also.

So, my point is: XNA is not a quick development tool for making a good game in only a month. Almost any attempt to this, will probably result in a uggly and crapy game, and I say almost any. And this results in bad publicity for XNA, as we all can see. And not even that persons that developed those games, can gain with this, because they almost make some crapy and not organized or optimized code just to enter this kind of contests that don't bring nothing to anyone.

Do you want to see a good contest? What about a tutorial contest? Make a "complete game" tutorial contest! Then release the tutorials. Everyone will gain something about this, don't you think?

Also, I am not saying that XNA is not a very good tool, or that the samples and articles are also not good. I love them all. But look at this tutorial: FPS Tutorial. It teaches you to almost make a decent game. But look at this one: 3D Plataform game. It is even a better tutorial than the first one.

 

XNASorcerer:

Also, I am not saying that XNA is not a very good tool, or that the samples and articles are also not good. I love them all. But look at this tutorial: FPS Tutorial. It teaches you to almost make a decent game. But look at this one: 3D Plataform game. It is even a better tutorial than the first one.

There is a quite important difference between Unity and XNA Game Studio, Unity is a complete engine with a fixed idea of how everything fits together. XNA Game Studio is essentially a framework for building games and game engines, it works at a lower level than Unity; XNA GS doesn't have any say in how you do your terrain or animations or AI. Look at projects like Blade3D which are basically building something like Unity on top of XNA to see how different they are.

XNASorcerer:

Well, I didn't see any article or tutorial about physics, AI, animations or LOD multitextured terrains. So, what we have are really the samples, but we have to spend some time figuring out how they work and how to make them to working together. Not all samples works together! For example, try to add the chase camera to the tank animated collision sample. Now that sample doesn't have LOD or multitexturing, so try to replace the terrain with a model ( to use a better texture and some LOD ). It won't work also.

So, my point is: XNA is not a quick development tool for making a good game in only a month. Almost any attempt to this, will probably result in a uggly and crapy game, and I say almost any. And this results in bad publicity for XNA, as we all can see. And not even that persons that developed those games, can gain with this, because they almost make some crapy and not organized or optimized code just to enter this kind of contests that don't bring nothing to anyone.

Do you want to see a good contest? What about a tutorial contest? Make a "complete game" tutorial contest! Then release the tutorials. Everyone will gain something about this, don't you think?

Also, I am not saying that XNA is not a very good tool, or that the samples and articles are also not good. I love them all. But look at this tutorial: FPS Tutorial. It teaches you to almost make a decent game. But look at this one: 3D Plataform game. It is even a better tutorial than the first one.

 

Indeed, the AI contest was just a month. But it was a warmup. And it had a specific theme: AI. The goal of the contest was to find some interesting applications of A.I. Making a full 4-moths contest on A.I. would not have been a very good idea, because few hobby developers are willing to experiment with AI for 4 months. I think having a 1 month warmup dedicated to this theme was a good idea, because those people that like this area of gamedev had a chance to show a bit of what they can do and gain some recognition from Lionhead, Rare, or MS Research.

But the main competition is 4 months. You don't think very good games will come out in 4 months? Should they just drop the Dream Build Play altogether?

Tutorial contests are hosted by Ziggyware about three times a year. And usually some advanced examples come out of it. Now you can't expect everything to be given to you without work.

If XNA were a "quick development tool", in the way you wish it were, then we would loose lots of flexibility. The XNA framework is not supposed to be a tool. It's supposed to be an API. And compared with other API's (not other tools) it's pretty much the best I've seen. You complain that you can't make a decent game in a month (with which I don't quite agree). Look at other API's. Can you make anything decent in a month using OpenGl, or native DirectX?

The tutorials you link to are tutorials for a tool. Comparing a tutorial that shows how to make a game in Unity with a tutorial that would show how to make a complete game in XNA is like apples and oranges. You can't really compare the two.

And you seem to forget the starter kits. There you have complete games, with all source included, showing how you can apply advanced techniques. And when Fabio is done with his first person starter kit, you'll have : "multiplayer over Xbox Live, skeleton animated player models, advanced dynamic lighting and shadows using deferred shading, depth of field, motion blur and more!"

XNASorcerer:

So, my point is: XNA is not a quick development tool for making a good game in only a month. Almost any attempt to this, will probably result in a uggly and crapy game, and I say almost any. And this results in bad publicity for XNA, as we all can see. And not even that persons that developed those games, can gain with this, because they almost make some crapy and not organized or optimized code just to enter this kind of contests that don't bring nothing to anyone.

Depends on how good you are. :)

Catalin Zima:

And you seem to forget the starter kits. There you have complete games, with all source included, showing how you can apply advanced techniques. And when Fabio is done with his first person starter kit, you'll have : "multiplayer over Xbox Live, skeleton animated player models, advanced dynamic lighting and shadows using deferred shading, depth of field, motion blur and more!"

I'm still waiting for the RPG Starter Kit, unless I finish mine first. :)

Machaira:

I'm still waiting for the RPG Starter Kit, unless I finish mine first. :)

There's no reason we can't have both the official one, and yours. :)
I can't wait to see them both. How's your status, btw?

XNASorcerer... look at this thread, for a series of tutorials on making a full game engine in XNA.

:)  I am not saying that what we have so far are bad. And I am also not saying that we can not find tutorials. And I am also not comparing Unity with XNA Game Studio. I showed the Unity tutorials as examples of tutorials. Wouldn't be more helpfull if all the samples or tutorials made by XNA guys were made as progressive tutorials?

1º tutorial - Lets make cameras tutorial ( chase camera for sample )

2º tutorial - Lets make a collision with heightmap tutorial ( Starting with the chase camera project )

3º tutorial - Lets add add some physics. ( Starts with the second project )

4º tutorial - Lets add animation . ( Starts with  the 3º project )

etc.

One more thing before get back to the topic... XNA is a API for make games, but it contains a game Studio, isn't? Well, a game studio without the basic things like animation or GUI or collision libraries, is like a music studio without instruments. Every musician or band can take their own instruments, but that doesn't mean that the studio shouldn't have some instruments. Translating: The XNA game studio should have at least one GUI, one animation and one collision library that the user could choose to use, or not to use if he wants to implente a better of different one. One example of this is the Model importing. There are lots of model formats outhere, but XNA guys decide to provide some default ones like x. and fbx. Why providing that, if the API already allows us to implement our own format importing? Does XNA looses any level of abstraction enought to hurt it, by having a x. and fbx. importing library? No.

Anyway, getting back on topic... Can anyone with honesty really tell what were the gain for the comunity or for xna, that came from this contest? Did XNA showed some good stuff, some good games made fast? No! Did XNA looked good by this? No. Did you guys see the bad feedback about this, or was I the only one to see it? Did the comunity gain anything with this AI bad experience? No! Did the AI evolved enough within XNA? No. Why not? Well, the time was too short, and to the developers that tried to making something had also to think about collisions, animations, GUI. In other words, time wasted that could be used only to developing and implementing AI.

After saying all that, I do agree that XNA is a great API, but it can easelly go to the wrong way. One example of how Microsoft can go the wrong way about this, is the Visual Studio 2008. Why did they decide to release it with another free game api that looks a lot worse than xna but that works with C++? How does that hurts the effort that was made to convince people of changing from C++ and give a chance to C# ? Will XNA have the same future of MDX?... Sorry, getting of topic again.

Catalin Zima:

Machaira:

I'm still waiting for the RPG Starter Kit, unless I finish mine first. :)

There's no reason we can't have both the official one, and yours. :)
I can't wait to see them both. How's your status, btw?

Probably not as far along as the MS one. :( I'm working on the conversation system right now. That's the last of the RPG engine pieces though I think. Part of the problem is that I'm trying to blog about it at the same time.

I'm not really worrying about the graphics engine. I'm even thinking of going old school MUD style (well, maybe not :D). The RPG engine will be in a library so you can plug it into any graphics engine though.

XNASorcerer:

One more thing before get back to the topic... XNA is a API for make games, but it contains a game Studio, isn't? Well, a game studio without the basic things like animation or GUI or collision libraries, is like a music studio without instruments. Every musician or band can take their own instruments, but that doesn't mean that the studio shouldn't have some instruments. Translating: The XNA game studio should have at least one GUI, one animation and one collision library that the user could choose to use, or not to use if he wants to implente a better of different one. One example of this is the Model importing. There are lots of model formats outhere, but XNA guys decide to provide some default ones like x. and fbx. Why providing that, if the API already allows us to implement our own format importing? Does XNA looses any level of abstraction enought to hurt it, by having a x. and fbx. importing library? No.

It's just a name. They just needed a name to differentiate XNA (the brand) from XNA (the framework .dll files) and XNA (the Visual Studio plugins). So they called it XNA Game Studio.

Anyway, getting back on topic... Can anyone with honesty really tell what were the gain for the comunity or for xna, that came from this contest? Did XNA showed some good stuff, some good games made fast? No! Did XNA looked good by this? No. Did you guys see the bad feedback about this, or was I the only one to see it? Did the comunity gain anything with this AI bad experience? No! Did the AI evolved enough within XNA? No. Why not? Well, the time was too short, and to the developers that tried to making something had also to think about collisions, animations, GUI. In other words, time wasted that could be used only to developing and implementing AI.

Publicity for XNA in regards to this competition means nothing. So what if Engadget and others say the games look terrible. The goal was to come up with interesting AI and possibly draw other developers to XNA. Most game developers will look at those games and say "Yes, those look about right for a month of work." It's not as damaging as you might think. I think the contest, both as a warmup and on its own, was a great success. Lots of great games came out of it and it was a good opportunity for people to show what they can do.

After saying all that, I do agree that XNA is a great API, but it can easelly go to the wrong way. One example of how Microsoft can go the wrong way about this, is the Visual Studio 2008. Why did they decide to release it with another free game api that looks a lot worse than xna but that works with C++? How does that hurts the effort that was made to convince people of changing from C++ and give a chance to C# ? Will XNA have the same future of MDX?... Sorry, getting of topic again.

DarkGDK is a totally separate product from a totally separate group of people. You can't pidgeon hole everyone into using a single product, so it makes sense to support more than just XNA Game Studio. It's just another tool for people to consider for game developing.

Ok.  :)