[project] Icebergs
Hi all, Im Carlos Coronado. 3 weeks ago I finished a 2 years Project (Warcelona, a custom campaign for left 4 dead 2), and I think its time to begin a new proyect. Since long time ago, In my head was the idea of making a single player puzzle game has my final university project, I I think now its time to begin with it.
This project wont be like Warcelona, it will be much more personal and experimental, thats why Im going to do it alone, but maybe that will change in the future.
I will do the project in UDK, but Im not going to download UDK until 2 or 3 weeks from now. Thats because I want to begin in UDK with a solid base. Ive done a lot of reference research since now, and now its time to put them in common. Thats why I bought a big cork board and a poket Sketchbook that Im going to take everywhere.
General game idea: The main idea is to have a very modular estructure (programming, gameplay and envirorments). Basically, the game is a first person GAME (not shooter) about puzzle solving a la portal (find the exit with a given gameplay). Every puzzle is an iceberg (100% of the game takes place in the artic) and the main objective is to reach the end (magic light that teleports you to another iceberg). For that, youve got a special artifact/skill; you can play with the climate. You can do global climatic changes like make all colder to freeze water and stuff and create solid parts, warm it to unfreeze them, wind to take impulse, fog to hide from you enemies, and stuff like that. The main idea is that this changes will always be GLOBAl and not local, si if we choose hot all the envirorment will change to hot.
Anyways, thats just the main idea, when playtest time, Im sure a lot of things will change, but for now its ok. The main idea is still to play with clima and weather.
About the story? I cant tell a lot by now, but Thruman show is a big reference for me.
The envirorments have 4 kinds of architectures:
- Rocky structures
- Ice structures
- Roman structures
- And modern architecture inspired by Mies van de Rohe
All these have to achieved with modular sets.
Also, in my new website www.ccoronado.com (still a WIP), I will make some tutorials in English. At the beginning this tutorials will be rubbish (at pronunciation) but the idea is to make them to improve my English pronunciation.
Now Im doing my first test with rocks following an incredible tutorial.
The idea is that all this rocks modular sets have the same diffuse, normal and specular. That why I can have a big texture resolution (2048x2048). With 3 or 4 more models (floor, another rock alone, a rock fence and stuff like that) should be enough to export it to UDK and start playing around. When all is exported, I will do the first gameplay tests and from there we will see.
Well, thats all for now!
Ps: sorry for my bad English!
This project wont be like Warcelona, it will be much more personal and experimental, thats why Im going to do it alone, but maybe that will change in the future.
I will do the project in UDK, but Im not going to download UDK until 2 or 3 weeks from now. Thats because I want to begin in UDK with a solid base. Ive done a lot of reference research since now, and now its time to put them in common. Thats why I bought a big cork board and a poket Sketchbook that Im going to take everywhere.
General game idea: The main idea is to have a very modular estructure (programming, gameplay and envirorments). Basically, the game is a first person GAME (not shooter) about puzzle solving a la portal (find the exit with a given gameplay). Every puzzle is an iceberg (100% of the game takes place in the artic) and the main objective is to reach the end (magic light that teleports you to another iceberg). For that, youve got a special artifact/skill; you can play with the climate. You can do global climatic changes like make all colder to freeze water and stuff and create solid parts, warm it to unfreeze them, wind to take impulse, fog to hide from you enemies, and stuff like that. The main idea is that this changes will always be GLOBAl and not local, si if we choose hot all the envirorment will change to hot.
Anyways, thats just the main idea, when playtest time, Im sure a lot of things will change, but for now its ok. The main idea is still to play with clima and weather.
About the story? I cant tell a lot by now, but Thruman show is a big reference for me.
The envirorments have 4 kinds of architectures:
- Rocky structures
- Ice structures
- Roman structures
- And modern architecture inspired by Mies van de Rohe
All these have to achieved with modular sets.
Also, in my new website www.ccoronado.com (still a WIP), I will make some tutorials in English. At the beginning this tutorials will be rubbish (at pronunciation) but the idea is to make them to improve my English pronunciation.
Now Im doing my first test with rocks following an incredible tutorial.
The idea is that all this rocks modular sets have the same diffuse, normal and specular. That why I can have a big texture resolution (2048x2048). With 3 or 4 more models (floor, another rock alone, a rock fence and stuff like that) should be enough to export it to UDK and start playing around. When all is exported, I will do the first gameplay tests and from there we will see.
Well, thats all for now!
Ps: sorry for my bad English!
Replies
Yes, that's true, the diffuse and normal of the rocks need some tweaking. That will be fixed in the next version.
Meanwhile, I have finished the Warcelona making of, that are design notes in my portfolio so anybody can see the work that can not be seen in the beauty images.
Also, I asked a friend about doing a prop for practice the high poly modeling and then baking the normal map to a low poly and he recommended me to do a Kabar, one of those cool army knifes.
It's not perfect, but I'm really happy with the result because it is my first "serious" high poly baking into a low poly.
Yes that's true. I was trying to achieve a clean look, but I did it too much I think.
This will be the intro level of the project. Just me playing with udk and falling in love.
The Landscape is a Landscape (lol) with a displacement and color maps generated with World Machine 2. I followed this [ame="
I'm a level designer not an environment artist. The hardest part so far for me is to achieve a good looking natural mountains. This project will be released as a free "game" were the only objective is to walk the way to an old house in the middle of the Valley (walking non stop takes 15 min). Of course, the player will find a lot of environmental narrative while walking and a little story. Think about the film Mr. Nobody, the idea is similar.
Hope you like it! From now on I will post all the process.
not to be a icebreaker.. ( pun intended) but you need level diversity if so !
Take oldschool 2D puzzles as a starting ground. none of em use the same environment the whole game.
- anyway read below -
- this is loosely based on my interpretation of ur starting text ! i might of misunderstood !
Well, icebergs is the concept, but they won't be 100% natural and classy Titanic Icebergs. They will have a lot of variation and 3 sub-environments.
In the other hand, mini-project for learning the base of UDK finished.
Well, as I said, in two weeks that should be all. And that's it. When I begun the project, the main idea was to learn as much Udk as possible, and that's what I did. I've learnt a lot of tools, like particles (flying petals), foliage system or the material editor, but most important, I've learnt a workflow. Lot of time "spent" in UDN and not in the creative process of the project. That's why there's a lack of cohesion between some elements, but well, everything except the skybox and the base textures for the leaves and barks (hello SpeedTree) are custom and made in a record time, so I'm very happy with the result.
Now, I will create the section in my portfolio for this project and do some animation in matinee to make a video.
Here's the final result, hope you like it.
And the video of the environment! Hope you like it.
yeah you're right. There's a lot of stuff that could be improved with more hours of work, but I'm not going to do it. No, I'm not lazy hahaha. The purpose of this project was to learn the base of UDK and not to achieve something visually perfect. The road texture, the shapes and colors of the mountains (not enought sharp), etc could be improved.
Anyways, personally I don't like the Uber dramatic feeling of 99% of the environments in videogames, with toons of bloom, post process, impossible camera shots or bad DOF. They just don't look natural. With a professional Reflex camera with great lens and sensor (big dynamic range) it's impossible to achieve that dramatic lighting that we are used to see in videogames. I'm not saying it's bad, I personally prefer another kind of lighting. Less is more. But yes it is "too less" in this environment:poly142:
Yeah, I know. The menu shouldn't be one of the first things you do when you make a game. But I really needed it. It's because the Betatesters. I must be able to deliver just and executable with the maps to test in a easy way, and this simple but modular menu offered me that.
The menu was done with Scaleform ForecourseUI. Thanks a lot for the tuts Allar!
The background is a UDK level with a plane with the opacity for the silhouette for the boy, then the level rendered behind it and a particle system rendered above it. Then, in the top of the stack the UI is rendered.
For now in the menu you can quit the game, change the resolution or go to a new game (open whatever map I want).
For betatesting it's cool enought, but with time, I'm planning to integrate more options and later the load/save functions.
In the other hand, a very first version of the project (without that menu) can be downloaded if you want to test it and give me some feedback.
- 4 levels.
- Stunning Ice Cube texture
- Main objective: take the Ice Cube to the orange hole.
If you want to test it, you can download it here. I need to know if the reflective eater behaves strange in levels 2 and 4, and of course, if you like the "gameplay" and how hard do you find the different levels.
http://dl.dropbox.com/u/4816743/01-Dev_ice_prot.rar
hahah say that in about 2 years or so :P
Well, awesome feedback from Randy Paulk.
It's amazing the toons of mistakes and bad design I detect after watching him playing the four levels.
I must say that I've always wanted to introduce the fortune in gameplay. I like how fortune can make things difficult. I thought about 80% player skills and 20% fortune, but for what I can see, now it's more 60% player skills and 40% fortune (see levels 3 and 4). That's just BAD. And I will fix that.
Problems detected:
So, let's get into work.
[ame="
Right now the number 1 thing I see issues with is the collision objects he gets stuck more than half the time. but that "suicide" jump he made when he did that....uhh naturally because it was a low box. That to me felt like great gameplay right there.
Dev_Ice is no longuer Dev_ice, now its going to be Interactive Landscape. The Ice cube mechanics wont disappear but they wont be the core of the gameplay. Why? Because I dont want a fast gameplay, or to force the player to use precision, plataforming skills or marksmanship skills as do the 99,999999999 % of the games as the difficult curve of the game.
Now the player will have to solve puzzles with macrochanging the landscape (fog, night, rain, etc, and you can also combine them). The other day I had an eureka moment and it helped me a lot to solve a lot of conceptual gameplay mechanics. As Valve would say: we assume stuff that we think its true but its not and that limits us. The core of this games is; the player has to transport objects in order to activate events for macrochanging the scenario. My mind automatically associates do something change with and object > Button pressed, ergo, I need buttons (you can see them in those screens, the blue shapes in the floor). Well, I discovered that this was a huge limitation because the levels are big natural desolated landscapes (not apocalyptic) with diffuse limits of the map. Why? It just doesnt make sence that the player has to puto those interactive cubes or balls in exact places of the scenario. Doing a button means to create a whole infrastructure near the button so the button makes sense there, and that is not cool. My solucion? Goodbye buttons.
Now the players needs to put those interactive cubes or balls in areas (defined by ruins, natural stones suspiciously in circle, etc, but big areas, not buttons. Also, this creates good mechanics because the player also triggers the event when entering in the area and that is awesome to feed the exploration. If the player seens something natural strange him will go there. And even more than that! It also helps me to create new puzzles. Imagine one of this areas in a hill and our interactive object is a ball (ball>hill>balls slides>ball doesnt trigger the event). Time to think!
If you want to make a good game, you better do a good gameplay, but if you want to make an excellent game, you need to offer cohesion between the gameplay, the visuals and the narrative.
In PrIL (Project Interactive Landscape) this is going to be achieved by offering a visual representation of the dynamic macro-changes of the scenario. For example, if you need to activate the rain to make something grow instantly and then been able to climb to that cliff, you will get an instant rainy day, but if you also need the night so you can travel fast, you will get the combo night+rain, and we want to make sure that every state is really symbolic and unique (cohesion between gameplay & visuals).
LUT’s helps a lot in this process; doing a color correction directly in photoshop and the take it back to UDK is… amazing. Unreal matinee is also a good tool to achieve this “modular climatic design”. In the case of the fog, we want full fog when the “fog event” is on, but we want also more fog than usual when the rain is on. How do you achieve that? Unreal Matinee.
When the fog event is on, I animate the fog density parameter; from 0.05 to 2.5. For the fog of the rain, instead of animating the fog density parameter, I increase the Z location of the exponential fog density. In that way, the same effect is acquired but using two variables, so you don’t need to worry about hard transitions or problems between 3 values of fog density. Keeping stuff in a binary form always helps to put some order.