Tux emits bubbles while being underwater

[bruhmoent]

This pr adds a visual effect - when Tux is underwater, he will emit bubble particles that will follow a sine pattern until they reach the water surface.

Make sure to test.

[Brockengespenst]

I think this is a pretty nice feature, but is it possible that the air_bubble-*.png files are missing? Maybe some idea: What about using std::array for m_bubble_particles? This would make it easier to iterate over the array or do a check for bounds (e.g. when picking a random bubble).

[bruhmoent]

Oh you're right, I forgot to push the bubbles.

[Brockengespenst]

Looks pretty nice so far. Some comments from my side:

• Sometimes it seems the bubbles are not emitted at the correct place after Tux has turned around. That looks a bit weird when the bubbles are emitted at the wrong side of Tux ;-)
• I am not sure if the bubbles should really leave the water and vanish in the air. Probably they should start to vanish a bit earlier

[bruhmoent]

I like the effect of them "popping" in the air. And for the bubbles not being emitted at the right place, I'm aware - but I don't know what to do about it - sometimes the sprite rotation gets registered improperly - At-least that's what I've observed.

[Brockengespenst]

Some suggestions:

• Use m_swimming_angle as it is already in radians
• I do not fully understand your direction check. If comparing the angle in radians, I thinks something like m_swimming_angle > M_PI_2 && m_swimming_angle < M_PI_2 * 3.0 should check if Tux is facing left, all other cases Tux is facing right
• If the beak was at Tux.height / 2, there would even be no need to check for direction, just rotate it. But if doing so in this case, this might lead to a small offset, maybe you can check if this looks still OK
• I would not use std::vector for m_active_bubbles as it might lead to lots of copy operations when deleting entries as vector must guarantee that all entries are placed in a linear order on the heap. When deleting items in the middle, they have to be copied to a new place in linear order. As you do not need to access the bubles by index, you could simply use a std::list or similar

[bruhmoent]

Thanks for the suggestions!

[Brockengespenst]

No problem. I really like this feature, but I think there are two small tweaks to be done:

• I guess small and big Tux need to be handled differently. Small Tux swimming left looks good, but when small Tux swims right, there is a bit of space between the beak and the point where bubbles are emitted
• Also, kind of related probably, left and right direction of Tux seem to need different offsets. Big Tux swimming right emits bubbles direcly at his beak, but big Tux swimming left emits the bubbles somwhere "behind his head"

It would be great if you could have a look at these two things.

[Rusty-Box]

@bruhmoent Why did you not use the .sprite file I send you?

[Rusty-Box]

Also, I think the bubbles vertical speed could be increased by quite a bit

[bruhmoent]

@bruhmoent Why did you not use the .sprite file I send you?

I just used the images - I found the .sprite file to be unnecessary? Idk how does it matter really

[Brockengespenst]

Sorry, but I have to slightly update my suggestion regarding how to determine the direction. From the swim sprite logic, I saw that the following code is used: std::abs(m_swimming_angle) <= math::PI_2 ? Direction::RIGHT : Direction::LEFT This seems to work more reliable, because it takes also negative angles into account.

[mstoeckl]

While this looks nice and cartoony, I (random passerby) should warn that cartoons are often wrong, and actual penguins probably don't breathe out underwater. (Nor should any species that can't breathe in underwater, except if very close to the surface :-)

Instead of emitting periodic bubbles from their beaks, penguins might produce them through:

1. splashing near water surface
2. possible: some bubbles when opening beak underwater to grab things for the first time
3. releasing trapped air after diving in (looks like many small trailing bubbles from the back/wings of penguin; could take some time for all bubbles to go.)

Therefore: I would suggest emitting the bubbles more irregularly in space and time; from the main body instead of from the beak; and maybe progressively reduce the bubble rate the longer Tux stays underwater.

[bruhmoent]

SuperTux is not aimed to replicate realistic behaviours. This is just a little visual effect. This is out of scope for this PR - and this might be adjusted in the future.

[mstoeckl]

SuperTux is not aimed to replicate realistic behaviours. This is just a little visual effect. This is out of scope for this PR - and this might be adjusted in the future.

Agreed. While I think foundations in reality are nice to have, in the end art and gameplay are what matter.

[Rusty-Box]

The bubbles are still not animated despite the sprite file having 2 action of animation. One for bigger bubbles and one for smaller ones.