I've seen this high-speed 'pseudo-stall' problem (and the oscillation issues) occur in one or two other recent videos of Scramble. It's definitely concerning. My (very tentative) current theory is that what we're seeing is due to floating-point precision limitations beginning to creep in, the further we get from the center of the game world. I've seen this kind of creeping inaccuracy error in another game which (like Scramble) needed to track positions of multiple objects with great accuracy across a large volume of space. The further away from 'center', the worse the positioning errors became. In said other game, the devs had to change from 16-bit to 32-bit floating point precision to make things behave properly. In a game like Scramble, which has to track even individual bullets and trajectories with great precision (to within a few centimetres, I'm guessing), not to mention the positions of (relatively) tiny parts of the various aircraft and the pilot too, I imagine the game already uses 32-bit FP precision, but at the edges of the game world, this may not be enough. They might need to switch to 64-bit precision before things start to behave as they should, when at the boundaries of the game world. Of course, I might be totally barking up the wrong tree here!
First off, let me be clear. The oscillations are 100% pilot error. It's NOT a bug. The issue is that I'm trying to achieve an (almost) exact speed. On one turn, I'm flying level and fast and so I pitch up. The next turn, I'm flying up and loosing speed, so I then pitch down to almost level. And the cycle repeats. In most dogfights, you won't notice or even care if your plain wiggles a little. Now the "pseudo-stall" IS a bug. A very annoying bug too. While I don't know what causes it, (it might be floating point error) it seem like there might be some directionality to it. Like if you fly one heading you stall, the other way, you "over lift" like the second Stuka. 24:47 More testing needs to be done.
You are right that there is a bug in the current build that becomes very pronounced over many turns of stable flight. It *is* a bug related to small angles, and I first thought it was a floating point issue, but floating point error only became a problem because of another issue I had in the physics. It will be fixed in our first big update!
@@JonCoughlin-x2o "Hey, I love that you're doing some real flight test analysis here! My background is aerospace engineering so this is the part of the game I nerd out about." - JonCoughlin-x2o Sorry, I know I caught your comment mid edit, but thanks. Truth be told, I do a lot of this kind of thing. Testing game mechanics and doing science within videogames. (And yes, I do wright it down. (thanks Adam)) Looking forward to any bug fixes that come along. I love the game, but it's hard to give a solid recommendation in it's current state. I'll be sure to make an update video when the physics gets fixed. Good luck and fly safe. 👍
I've seen this high-speed 'pseudo-stall' problem (and the oscillation issues) occur in one or two other recent videos of Scramble. It's definitely concerning.
My (very tentative) current theory is that what we're seeing is due to floating-point precision limitations beginning to creep in, the further we get from the center of the game world. I've seen this kind of creeping inaccuracy error in another game which (like Scramble) needed to track positions of multiple objects with great accuracy across a large volume of space. The further away from 'center', the worse the positioning errors became. In said other game, the devs had to change from 16-bit to 32-bit floating point precision to make things behave properly.
In a game like Scramble, which has to track even individual bullets and trajectories with great precision (to within a few centimetres, I'm guessing), not to mention the positions of (relatively) tiny parts of the various aircraft and the pilot too, I imagine the game already uses 32-bit FP precision, but at the edges of the game world, this may not be enough. They might need to switch to 64-bit precision before things start to behave as they should, when at the boundaries of the game world.
Of course, I might be totally barking up the wrong tree here!
First off, let me be clear. The oscillations are 100% pilot error. It's NOT a bug. The issue is that I'm trying to achieve an (almost) exact speed.
On one turn, I'm flying level and fast and so I pitch up. The next turn, I'm flying up and loosing speed, so I then pitch down to almost level. And the cycle repeats.
In most dogfights, you won't notice or even care if your plain wiggles a little.
Now the "pseudo-stall" IS a bug. A very annoying bug too. While I don't know what causes it, (it might be floating point error) it seem like there might be some directionality to it. Like if you fly one heading you stall, the other way, you "over lift" like the second Stuka. 24:47
More testing needs to be done.
You are right that there is a bug in the current build that becomes very pronounced over many turns of stable flight. It *is* a bug related to small angles, and I first thought it was a floating point issue, but floating point error only became a problem because of another issue I had in the physics.
It will be fixed in our first big update!
@@JonCoughlin-x2o
"Hey, I love that you're doing some real flight test analysis here! My background is aerospace engineering so this is the part of the game I nerd out about." - JonCoughlin-x2o
Sorry, I know I caught your comment mid edit, but thanks. Truth be told, I do a lot of this kind of thing. Testing game mechanics and doing science within videogames. (And yes, I do wright it down. (thanks Adam))
Looking forward to any bug fixes that come along. I love the game, but it's hard to give a solid recommendation in it's current state.
I'll be sure to make an update video when the physics gets fixed.
Good luck and fly safe. 👍
@@JonCoughlin-x2o So I see the high speed "stall" bug has been fixed. 👍
Guess it's time to update these videos.