infinite scrolling background inconsistency

[hadan24]

edit: I can't recreate the issue consistently but don't know what could possibly be the cause. Any pointers or ideas would be greatly appreciated!

I'm making an endless runner for my first bevy game, and am using 2 images as my infinitely scrolling background. Sometimes when I recompile my game, the images aren't lined up as they should be, leaving several pixels of empty space between them. I've noticed it tends to happen when the game takes longer to initialize, but not exclusively then.

My current code to handle this is below. Any ideas what could be going on or how to solve this consistently? I've tried several different conditions for "resetting" the images once they're off-screen, different offsets, adding the dt*spd at different points in the function but none seemed to fix it consistently.

pub fn scroll_bgs(
    mut bgs: Query<(&mut Transform, &BgDimensions), With<Background>>,
    time: Res<Time>
) {
    let scroll_spd = 350.0;
    // w/o this offset, the images are *always* 4 px apart. With it, they're only *sometimes* a few px apart
    let x_align_offset = 4.0;

    for (mut bg, dims) in &mut bgs {
        let real_width = dims.width * dims.scale;

        // offset the scroll reset coord to keep this "is off-screen" check simple
        if bg.translation.x < -real_width {
            bg.translation.x = real_width - x_align_offset
        }
        bg.translation.x -= time.delta_secs()*scroll_spd;
    }
}

[Carson1234561254]

I don't know much about scrolling in bevy (depends if its side to side, up and down, ect) but maybe your not deleting the level behind player/user?

[hadan24]

The backgrounds go from right to left (it's just a simple 2d game). I'm not sure if this matters, but I'm not deleting the images when they go off-screen, I instead set their x-coordinate back toward the right so that the image is placed just after the last one in the line.

You did prompt me to think more carefully about how I'm doing that and to double-check my math though. I'm not sure yet whether the changes have definitively fixed the issue, but thank you!

[Carson1234561254]

No problem I can think about it more if you want me to, I'm currently debugging our own game me and the team making (used to be 124 errors a few days ago now about 26, and I'm only coder, they handle models side) want me to think about it more?

[hadan24]

I'd totally appreciate it if you have the time! I'll also keep monitoring it and will post here when I'm mostly sure it's fixed