next »

[Deraj]

I'm sure you've thought about this, but how are you going to avoid infringing  on copyrights/trademarks?

I assume the graphics can be redrawn, but do you have a backup title?

[J-Snake]

It's a free game for the fans. If DoctorM64 doesn't have a problem with his Metroid 2 remake I think there won't be a problem with this one aswell.

The commercial version will get its original skin and it has an original title.

[J-Snake]

In this video jumping and some important things related to it are demoed.





Feel free to ask if you want more information.

[J-Snake]

I want to elaborate on the fundamental mobility mechanics presented so far to reflect the thought process. The granular aspects in this stage can fundamentally affect the whole game. Thus it is important to know what the game is actually aiming for. It is not possible to understand everything at this stage but the big picture has to be there, that is strategic gunplay utilizing sophisticated environment traversal. That demand already dictates how the granular level of mechanics should work, in many occasions. For example a jump should not be slippery like in a mario game since the challenge is in the higher mind-level of play. Basic navigation and positioning shouldn't involve skills at all so that you can focus on strategic qualities of gunplay and map traversal. However more sophisticated chains of mobility actions can involve a lot of skill (and look very cool when performed well;)), but they are fundamentally tied to strategic repercussions which are dependent on positioning of the players, their load out, and current suit-energy. So you still have to know what you are doing no matter how cool you are at it. That's the guideline to design for.

Because of the above explanation it came very straight forward to design instant canceling of horizontal movement. It means when you release the button you will stop horizontal movement instantly in the same frame of registered input (60hz). That makes precise landing on ground a child's play. On the other hand starting horizontal movement (like running) has to be accelerated to provide up to pixel perfect positioning control, like shown in the running video. Luckily, by taking all this into account the mobility model feels overall still more dynamic and fluid than in the original Super Metroid. There are only few instances where I had to balance direct control vs feel: for instance it looks overall ok to cancel horizontal movement instantly, but it looks very bad (weightless) when you cancel the rise of a jump instantly because the sign-change in vertical velocity is like a "cut". That is why when you release the jump button your vertical velocity is not
zero but a small number which has to be decelerated until you start falling down. This compromise will let you rise only negligibly very few pixels higher than instant canceling
and feel a lot more natural at the same time. Another reason why the mobility model looks more fluid and dynamic than in Super Metroid is because of the more sophisticated and precise animation control.

All in all the creation process is a series of strictly considering every mechanical addition to the game, and the decision process is like climbing up safety ropes provided by the big picture so you don't fall down.

Hope it was a valuable read. May be some of you feel the same here and there.

[James Edward Smith]

Are you still developing this in XNA? If so, do you know what the standard practice is now for creating installers for people so they can run a game made in XNA if they don't already have the framework?

[J-Snake]

Yes, that is still xna but I am considering to expand to java later on.

I use the inno installer. You can find help on using it for xna here:
http://damnigames.com/2012/06/28/creating-a-pc-executable-for-xna-4-0/

Feel free to install the TrapThem-Demo to see the result. No internet connection required;)

https://www.dropbox.com/s/hjg91izqbzu6hit/TrapThem-Demo%20Setup.exe

[dancing_dead]

Just watched the original brainstorm videos and the latest one about jumping, too, I am very impressed by the movement achieved in them. Good job!

I'm not entirely sold on the idea that you get real, tangible benefits with not using floats for coordinates, tho. They do lose precision, yes, but for any non-massive space I'd wager that they're very much good enough, especially given the size of the areas in the videos so far demonstrated.
I applaud the technical dedication, though! I just don't think it's really necessary.

also, can't resist...

That sounds much easier than the calculations that would go into 3D AI development. I highly doubt that any amount of suit-energy, difficulty-spline, or cross-contained micro-actuation could make this "more difficult" than what goes into an average FPS's AI.

haha, nah. average FPS AI is A* + raycasts for aiming. this game requires the same + something else to manage the suit energy and navigate the environment tactically. these requirements alone make this game's AI require more algorithms and calculations than your "average FPS AI". if you want the AI to be good, too, then these requirements are only rising.

[J-Snake]

It is all running in Java and Opengl now. So far everything fine but I am still not sure it is preferable to the C++ + SDL2 + OpenGL option.

In the following the wall jump is explained and why it differs from Super Metroid:

[J-Snake]

I have spent a hard time evaluating aiming mechanics: Free aim vs 8-way aim. I initially wanted to go with free aim but I feel overall a greater value is to be gained by going with 8-way aim.

To better understand the implications of different aiming mechanics let's put two extremes side by side: free aim vs 2-way aim (megaman-like). One can quickly see that a megaman-like aim enforces a strong focus on positional placement, for example if an enemy is over your head you have to jump and then shoot in order to kill it. So potentially you have to string together movement actions with shooting actions, which can add a puzzly layer to the action gameplay. And this is potentially good, however it shouldn't feel too restrictive. It seems there is a sweet spot between the freedom of aim and the action depth I can get out of it for SMT. And 8-way aim turns out to be a good solution. Free aim, on the other hand, eliminates a lot of beautiful gameplay action relying on positional placement.

It turns out a snes-like gamepad is just about right to be properly maxed out for this game, which is nice.

[J-Snake]

So here is an overview of the alternative control design (or a close approximation of it) if I decide to go with the 8-way aim.

Two things deserve an explanation at this point.

- Locking x-position and locking aim direction is both possible by one single button. Btw. you can see in Metroid Zero Mission
  that it aimed for the same thing but the results aren't satisfying. I believe we can do better. Fortunately we can exploit  
  the following fact: Whenever you need to lock the aim you need to move. So when a direction is pressed before you
  press the "R-button" the aim will be locked and you can move. But when you first press the R-button the x-position will be
  locked and you can aim. Mechanically, there is no conflict.

- The second thing is the weapon selection. It works like in Crysis. That means you hold the weapon selection button down, a
  pop up menu appears and you select a slot with the corresponding direction. That way you can switch between up to 8 weapons
  instantly, literally on the fly;)

[gimymblert]

subscribing will read later

[Overman]

Aren't you worried about copyright? 

Anyway I'm not a big fan of clones or fangames, sorry. Good luck anyway! 

[J-Snake]

This project is paused for now in order to allow other projects to complete. It remains rather a secret long term side project but whenever there is something valuable to share without spoiling much updates will follow.