RealRocketRacing [Prototype]

edited in Questions and Answers
I love local multiplayer games. I am not a fan of split screen. For many games, split-screen feels like a workaround which could have been avoided with better design.

The problem I'm having is that the game I'm working on sees players having an adversarial relationship AND an incentive to move away from one another. The problem is compounded by the thin lines of the art style.

For some context, this is a racing game in zero gravity with rockets.

The camera works really well when racers are near one another:

But when they are far away, the game becomes almost unplayable:

Any suggestions would be appreciated. :)
640 x 400 - 34K
902 x 506 - 63K


  • What makes it unplayable? Is it the way the line art gets hard to see, or does the control scheme have some camera-relative properties?

    If it's the art, consider changing the art style to use more solid colour areas that will survive aliasing better. Or you could supply different LOD assets for the line-art style if you need to keep it.
    Thanked by 1NickCuthbert
  • edited
    It is largely the aliasing but also that the game requires a fair amount of accuracy and when zoomed out, it becomes a lot harder to steer. I'll test out asset replacement, I hadn't considered that, that could possibly help a lot.
  • Or you can devise a system of actually drawing lines so that they're the same width no matter the level of zoom, but that's probably WAY overenginnering the problem at this level.

    Or split screen cam, but that's non-trivial. Ask @Asbestos about that.
  • It is largely the aliasing but also that the game requires a fair amount of accuracy and when zoomed out, it becomes a lot harder to steer. I'll test out asset replacement, I hadn't considered that, that could possibly help a lot.
    Ah, if steering is an issue, then maybe that can be fixed by adding more visibility to the direction a ship is pointing? Something like a headlight could totally help with that, especially when zoomed out.
  • edited
    Interestingly, this prototype was originally a lot more oppressive, it was focused on navigating through a near dark labyrinth with only spotlights. Increasing visibility seems sensible.
    Again thank you.
    Thanked by 1dislekcia
  • So changing the avatar to a fill sprite does indeed improve it. I'm still not entirely happy though.
    I think I might try using a render texture and a second camera to produce a localized magnification effect.
  • edited
    I've modified the game to use a magnification effect
    Again, I'd appreciate any feedback.

    Windows download link:

    Controllers recommended
  • edited
    Updated Build. Fixed nauseating camera behavior between a player death and respawn.

    Folder include a Windows and OSX Build:
    RealRocketRacing (Google Drive)

    I'm still not entirely sure if the lensing effect is a good solution or merely a poor compromise
  • To illustrate:
    1906 x 996 - 193K
  • Hmm. Lensing looks like it makes it hard to tell what's coming up along the track... Have you thought about doing the sort of splitscreen that Clutchfighters does?
  • I played it. I like the aesthetic of the game and the overall UI looks very clean and polished. The particle effects for the rockets is great and fits the aesthetic very well. Anyway, I really like how the game looks.

    The pace of the game doesn't fit the music. The music is high paced and actiony but the game itself with the tough controls makes it a very delicate and precise game. There was no point I was actually moving really fast to feel insync with the music. I don't know if you should change the pace of the game by not damaging when hitting walls that way you can fly through levels and bump off walls. Or change the music to fit the current pace. Or maybe I just suck at controlling the rockets.

    The lensing like @dislekcia mentioned makes it very hard to see what's coming. I'm assuming you may be opposed to split screen as I also I'm not a fun of that because it kinda feels like it's not thaaaat local anymore. I'm just speculating here. Why not make smaller maps and zoom in when close together and then max out when the cars are far apart. Just that when the cars are far apart they need not to be dots, thus a smaller map. I don't know, just my thoughts.

    Thanked by 1Sash
  • edited
    @dislekcia: Shamefully I actually didn't play clutch racer during Amaze; I watched a few games, but I'm not entirely certain what you're referring to.

    @SUGBOERIE: Thanks for the valuable feedback. It might be a good idea to a/b the game with various configurations and music.

    Ideally I don't want to reduce the map size. I'm not the best at racing games, but I was able to finish a lap in under 3 minutes though, so if I had to make a smaller level, I feel that it might be too short.

    Additionally I'm deliberately scoping down and trying to aim for a short but polished experience, but there are certain takeaways that I'd like to get out of this iteration, one of which is trying to get a better understanding of cameras, particularly local multiplayer cameras. You are probably correct in saying that smaller levels might work for this prototype better, but I feel that it won't ultimately won't be generalisable to what I have in mind for a second iteration.
  • @NickCuthbert: Clutchfighters does this sort of adaptive split screen thing where the cars can move around when they're on the same screen, then when they get too far apart, the game just draws a line between them at whatever angle perfectly separates them and gives them each that half of the screen as a local view. It's very elegant... I hope I'm remembering that correctly... ;P

    So I need to ask, what are you still keen on learning with this? Seems like you've found a weakness in your current art style that performs badly when zooming. What other camera styles and tricks are out there? What have you tried? If the game has certain demands, then that's fine as you see it existing in your head, but you need to decide if it's a learning game or a production game and then either dedicate the time to learning or not.
  • @NickCuthbert You could also adopt the Micro Machines camera/mechanic:
    - Camera gives priority to player in front.
    - Camera only zooms out by a certain amount.
    - If a player falls to far behind and moves off the screen, they are destroyed and the player in front gets a point. Both players start again at the previous checkpoint.
    - Winner is first to x amount of points.

    I think it would fit well with your current rocket steering mechanics, and this way you don't need splitscreen ;)
    The video below shows an 8 player game of Micro Machines V3, all the action on one screen.
  • edited
    @dislekcia: A dynamic splitscreen like clutchfighter seems like it might work.

    For single player, zoom based on speed and a maximum zoom level worked really well. I switched to a combination of the maximum of the maximum speed and distance between players. Then I tried lensing. So I haven't tried terribly many approaches.

    My main interest for this prototype is in trying different camera techniques and attempting to find a balance between simulation and playability. The art style is secondary.

    I want to make a game where players build custom rockets to race in very different physical environments. With physics games, I always find that I verge on the side of over-simulation. So I want to try and compromise in the name of fun in a more minimal sandbox before expanding the scope. I think I'm getting closer in this regard, but I don't feel that I'm there yet. It's mostly little things. At raxter's recommendation, I increased the angular friction, that helped improve the game massively in terms of feel.

    Thanks @pieter. I remember only very vaguely but with some fondness. I'll take a relook at it.
Sign In or Register to comment.