Hey guys, unfortunately centre of mass car -> impact location is not the correct vector that the game uses. It's centre of mass car -> centre of ball. In a lot of cases that is almost the same direction and many of the things I said in the video still apply. For the full explanation please watch Episode 6!
You can also balance the ball perfectly on the top of the car. The ball can be stationary or spinning and it's still possible. I've had the ball spinning on my roof for the last half hour.
I have balanced the ball perfectly on the top of my car, I think it's because I matched the speed of the ball perfectly and it was a little bit further to the right for the center of mass. The only way I got it up there was kicking it up when there was back spin on the ball and it was at the perfect position
"Dribbling is pretty much the same as 2 spheres stacked on top of each other" Thank you so much for that quote. I've been visualizing it as a box this whole time and haven't been able to properly wrap my head around how the dribble actually works, but this quote actually makes it all make sense
Great video! Even though some of this is just trivia I found it very interesting. Especially because I had always been wondering how exactly consistent hits were possible with that hitbox. The amount of effort you seem to put into these videos is amazing.
I've just finished watching this video, and I'm literally speechless since I have been wondering how the RL physics work for a long time. What a great work !!
Thank you very much, you have no idea how I appreciate this. I'm currently building a bot in rocket league, and I need a custom built physic state predictor. I wouldn't have been able to implement anything precise without this
Wait, wait wait wait wait! Hold on there for a second. I just can't get my head around the fact that there is 0% effect of the spin while the ball is in the air! Is the Ball camera that confusing to our eyes? I could swear i saw the Ball curving into the goal just so often... Please, i need an answer to this! It's just so important to calculate an aerial hit as well!
Yes. Indeed it can look like it but it never curves. This has even been confirmed by the developers. If you think you saw a ball curve try saving a replay and look at it in top down view. It should look like a straight line from there.
I've noticed that if your car is floating through the air close to the ball where you could touch the ball simply by rotating your car, you can rotate your car to bump the ball, but it doesn't work if the surface that touches the ball is the bottom of your car. In that case it will act like the ball is an immovable object. The ball's movement will not be affected at all and your attempt at rotating your car will be blocked. This doesn't seem like a logical behavior. If it is an unintentional bug, I would love to see it be fixed. If it is intentional, it is just another part of the rules we will have to accept, but I would love to at least hear the Rocket League developers comment on it.
Can you tell, how long the maps are? I think this is important, when we know the speed, lenght of the car etc. And thanks for the great work! Your videos are awsome!
No, they did not. The goal posts have been the same since release. I've been playing for 2050 hours and am Grand Champion. And if you want to check that, just input my Steam profile into a tracker steamcommunity.com/id/HoraryHellfire rocketleague.tracker.network
***** All good I believe you but I don't know... I been playing since release and I am one rank away from being in the purple ranks and I swear the bounces in the goals have seemed very weird lately
Well I don't know what exactly happened because I don't have any access to the code but just from watching the tons of reddit posts where you can see the ball hitting the inside of the net (and not actually the post) and bouncing out for no reason. I'm assuming you're referring to those because otherwise I have no idea what you're talking about. These bounces are just buggy and therefore do not follow the physics in the video.
Even with 2000h the dataset of one person is still rather small compared to the giant community we have on reddit. Are you going to deny how many people have been posting those bounces where the ball hits the inside of the goal and bounces straight back out? We've always had buggy bounces in the past. But I've never seen ones in those positions. If you know of similar posts from 3 months ago I'd love to see them. If you were just talking about bounces that "actually hit the post" then I think Nick probably didn't mean that. It can be hard to see the difference in alot of cases unless you go watch it in a replay.
There's a could elements to power that you didn't mention, like orientation velocity and how hitting it with the corner of the front hitbox's right/left bumper can add more power.
Orientation velocity? If I understand you correctly then you're talking about angular momentum of the car? I said that the "speed of the PART of the car that hits the ball" matters. If the car has angular momentum then some parts of your car can be faster than others which can influence the power. Is that what you meant? About the car getting more power on the corner. Did you do any testing on that because like I said in the video from everything I tested you can hit the ball just as hard with the front or the roof. Now in case you do a diagonal flip then your corner will rotate faster and the part of the car hitting the ball being faster -> harder shot.
Would you ever consider doing a video on dodge physics? I think a lot of people who frequent the subreddit has a general grasp of how it works, but I want to make sure I know with confidence that the my understanding is correct
That’s why I can’t read bounces. I’m champion 2 but I still can’t read some bounces and that’s because it’s unrealistic. I play football a lot so I know my bounces very well but in rocket league I don’t and it screws me up alot
I remember reading somewhere that the top speed of your car gets clamped, and if you accelerate in a different direction, you will be slowed down in the direction you're moving to keep your car going at the same top speed even if there is no other force pushing in that direction. What I wonder is if this would also apply to the wheels of the car when they move up/down, and if this would affect the rest of your car in any way. I have no idea how this would be tested or if it would be significant at all but it's just a thought I had.
Yes that's true. If you have a horizontal speed of 2300uu/s and you start falling due to gravity you will actually lose horizontal speed in order to never go faster than 2300uu/s. I'm not sure what you mean with the wheels. The maximum speed only applies to the center of mass. You can turn your car like you want (should be the same for the wheels) and you will still be at the same speed which means that some parts of your car may actually get faster than max speed at times.
New subscriber 2020 hey question how do you do to look very HD your screen looks so clean my look like shi&÷^×&×& and the other question is which camera settings did you recommend more I saw you video about that but I didn't get it thankyou by the way my bad my English is not perfect like your
Z should not be a concern because of think of it this way in the example he gave where it bounced off the side wall: From far away, yes the ball is accelerating downward. But consider as we concentrate closer and closer on the time of impact with the wall (lets say at t=10 seconds). If you consider the Z position of the ball at 9.9999 seconds and 10.0001 second, intuitively, there shouldn't be any difference in the Z position; therefore, we can say the bounce does not have any effects on the Z component. Alternatively if it DID have an effect on the Z component, then we would see an instantaneous jump of drop in position right at the time of contact with the wall
So since this game has unique values for everything, if I memorize and am able to predict the physics for this game, it wont help me calculate in any other game and would probably make me worse at those other games?
I have a question guys,: If my car in rocket league is driving forward at a speed after i used 50 boost and i hit the ball, and some guy at a standstill just taps it, shouldn't i win the 50/50 due to the force? And wouldn't in reality the force i hit the ball with my car send the other guys car backward because of the force my car exerts on the ball? Thanks for the reply.
Does rotating your car give speed to the part of your car that you hit the ball with? like if I flat hit a ball vs air-rolling/flipping into a ball, does the rotation give extra speed? I don't think the physics works that way but Im not sure
I wish I had a perfect answer to that question. From everything I've seen from reverse engineering and such it does not seem to make difference for the strength of the impulse but there might be some part in how the unreal engine applies the impulse that causes the speed of the ball to be faster. (back when I made this video I didn't have as much info available back then so I was going off observation and setting up some very similar shots with and without dodging (car at pretty much the same speed) and was getting different power. But that might have other causes. There is new stuff I find out every week and I'm hoping that I will just stumble upon some helpful extra stuff so that I can give a satisfying answer. Video about dodges is definitely a planned topic but probably not in the next 2-3 months.
Ah alright, thanks for the quick reply! Man, you really did create a channel over a really interesting topic, rl is such a physic based game that everyone who plays it has to wonder over the exact details at one time or another, and since the premise is so out there, the physics a lot of the time works differently than the intuition from irl. Kudos to you for the idea and then following through really well!
This is so fked up and counter intuitive, that's why every hit seems off, they should have made the hitbox a mesh wich becomes visible (like a force field) when hitting and would be much more clear with no made up physics.
Thanks :) Maybe consider doing your research first before complaining about someone not doing the research. There is a mutator called "Ball Max Speed", which allows you to, well maybe you guessed it, increase the maximum speed of the ball. In a match without that mutator, the maximum speed is 6000 uu/s = 216 km/h.
LMAO "Do your research" says some random guy to someone who literally sank hundreds of hours gathering data experimenting and analysing the game. And not only that, but 5 years after the video came out and was battle-tested by the whole community. I mean, wow, just wow.
Hey guys,
unfortunately centre of mass car -> impact location is not the correct vector that the game uses. It's centre of mass car -> centre of ball. In a lot of cases that is almost the same direction and many of the things I said in the video still apply. For the full explanation please watch Episode 6!
The hitbox is actually tilted slightly: by 0.56 degrees.
You can also balance the ball perfectly on the top of the car. The ball can be stationary or spinning and it's still possible. I've had the ball spinning on my roof for the last half hour.
Guess you've never programer a video game. Probably every 3d one you've ever played uses vector3 to store positions.
I have balanced the ball perfectly on the top of my car, I think it's because I matched the speed of the ball perfectly and it was a little bit further to the right for the center of mass. The only way I got it up there was kicking it up when there was back spin on the ball and it was at the perfect position
Can someone provide me the code which can be used commercially
"Dribbling is pretty much the same as 2 spheres stacked on top of each other"
Thank you so much for that quote. I've been visualizing it as a box this whole time and haven't been able to properly wrap my head around how the dribble actually works, but this quote actually makes it all make sense
Perfect way to sum it up bro
Excellent as always HalfwayDead! Love this series - keep up the good work.
Rewatching your videos and remembering just how ahead of your time you were when these came out. Thanks for the amazing content!
Great video! Even though some of this is just trivia I found it very interesting. Especially because I had always been wondering how exactly consistent hits were possible with that hitbox. The amount of effort you seem to put into these videos is amazing.
You must put a lot of work on these videos! Thank you for that because they are amazing!
I've just finished watching this video, and I'm literally speechless since I have been wondering how the RL physics work for a long time. What a great work !!
very interesting and fascinating physics of RL!
Always a great and in-depth analysis. Thanks!
This has more science than I expected
Thank you very much, you have no idea how I appreciate this. I'm currently building a bot in rocket league, and I need a custom built physic state predictor. I wouldn't have been able to implement anything precise without this
Wait, wait wait wait wait! Hold on there for a second.
I just can't get my head around the fact that there is 0% effect of the spin while the ball is in the air!
Is the Ball camera that confusing to our eyes? I could swear i saw the Ball curving into the goal just so often...
Please, i need an answer to this!
It's just so important to calculate an aerial hit as well!
Yes. Indeed it can look like it but it never curves. This has even been confirmed by the developers. If you think you saw a ball curve try saving a replay and look at it in top down view. It should look like a straight line from there.
very well explained! especially surprised by the impact normals thing
I've noticed that if your car is floating through the air close to the ball where you could touch the ball simply by rotating your car, you can rotate your car to bump the ball, but it doesn't work if the surface that touches the ball is the bottom of your car. In that case it will act like the ball is an immovable object. The ball's movement will not be affected at all and your attempt at rotating your car will be blocked. This doesn't seem like a logical behavior. If it is an unintentional bug, I would love to see it be fixed. If it is intentional, it is just another part of the rules we will have to accept, but I would love to at least hear the Rocket League developers comment on it.
5:53 it’s actually possible now
yep it is
Can you tell, how long the maps are? I think this is important, when we know the speed, lenght of the car etc.
And thanks for the great work! Your videos are awsome!
You can balance the ball on your car.
I can't play this game because it fucks with your intuition.
The whole speed at which a ball falls down is so tentative you keep doing it wrong etc...
Look at the white circle under the ball and it tells you when it will bounce.
That's why it takes so long to get good at it.
This explains why I'm so bad at dribbling. Thank You!
my god this is high quality
Did they change how the Goal Posts work cause I swear the ball bounces out of the goal more now if you hit them.
Nick Hooten true
No, they did not. The goal posts have been the same since release. I've been playing for 2050 hours and am Grand Champion. And if you want to check that, just input my Steam profile into a tracker
steamcommunity.com/id/HoraryHellfire
rocketleague.tracker.network
***** All good I believe you but I don't know... I been playing since release and I am one rank away from being in the purple ranks and I swear the bounces in the goals have seemed very weird lately
Well I don't know what exactly happened because I don't have any access to the code but just from watching the tons of reddit posts where you can see the ball hitting the inside of the net (and not actually the post) and bouncing out for no reason. I'm assuming you're referring to those because otherwise I have no idea what you're talking about.
These bounces are just buggy and therefore do not follow the physics in the video.
Even with 2000h the dataset of one person is still rather small compared to the giant community we have on reddit. Are you going to deny how many people have been posting those bounces where the ball hits the inside of the goal and bounces straight back out? We've always had buggy bounces in the past. But I've never seen ones in those positions. If you know of similar posts from 3 months ago I'd love to see them.
If you were just talking about bounces that "actually hit the post" then I think Nick probably didn't mean that. It can be hard to see the difference in alot of cases unless you go watch it in a replay.
Psyonix pls
There's a could elements to power that you didn't mention, like orientation velocity and how hitting it with the corner of the front hitbox's right/left bumper can add more power.
You mean... angular velocity? I'm also interested about this corner hit power.
Orientation velocity? If I understand you correctly then you're talking about angular momentum of the car? I said that the "speed of the PART of the car that hits the ball" matters. If the car has angular momentum then some parts of your car can be faster than others which can influence the power. Is that what you meant?
About the car getting more power on the corner. Did you do any testing on that because like I said in the video from everything I tested you can hit the ball just as hard with the front or the roof. Now in case you do a diagonal flip then your corner will rotate faster and the part of the car hitting the ball being faster -> harder shot.
The 4 people who disliked this probably don’t play rocket league
you deserve more subs/views great work
Would you ever consider doing a video on dodge physics? I think a lot of people who frequent the subreddit has a general grasp of how it works, but I want to make sure I know with confidence that the my understanding is correct
God I need a pinches and redirect video
was really hoping for pinch explanation in this video
more vids plsss :) LOVVEEE YOUR VIDEO!!!
That’s why I can’t read bounces. I’m champion 2 but I still can’t read some bounces and that’s because it’s unrealistic. I play football a lot so I know my bounces very well but in rocket league I don’t and it screws me up alot
I remember reading somewhere that the top speed of your car gets clamped, and if you accelerate in a different direction, you will be slowed down in the direction you're moving to keep your car going at the same top speed even if there is no other force pushing in that direction. What I wonder is if this would also apply to the wheels of the car when they move up/down, and if this would affect the rest of your car in any way. I have no idea how this would be tested or if it would be significant at all but it's just a thought I had.
Yes that's true. If you have a horizontal speed of 2300uu/s and you start falling due to gravity you will actually lose horizontal speed in order to never go faster than 2300uu/s.
I'm not sure what you mean with the wheels. The maximum speed only applies to the center of mass. You can turn your car like you want (should be the same for the wheels) and you will still be at the same speed which means that some parts of your car may actually get faster than max speed at times.
That's true I didn't think about how turning/rolling proves that they can still move faster than the car, I'm dumb haha.
lul how the ef do you have 2,000,000. Your videos make Rocket League a science which makes it way easier to understand.
ah, and i thought hitting the corner front of your car will generate the most power. Great analysis by the way! :D
Amazing youtuber keep up the good work, but what screen recorder do you use
Thank you :) I use shadowplay.
New subscriber 2020 hey question how do you do to look very HD your screen looks so clean my look like shi&÷^×&×& and the other question is which camera settings did you recommend more I saw you video about that but I didn't get it thankyou by the way my bad my English is not perfect like your
you mentioned the effects a bounce has on the X and Y components, but not the Z...
Z should not be a concern because of think of it this way in the example he gave where it bounced off the side wall:
From far away, yes the ball is accelerating downward. But consider as we concentrate closer and closer on the time of impact with the wall (lets say at t=10 seconds). If you consider the Z position of the ball at 9.9999 seconds and 10.0001 second, intuitively, there shouldn't be any difference in the Z position; therefore, we can say the bounce does not have any effects on the Z component. Alternatively if it DID have an effect on the Z component, then we would see an instantaneous jump of drop in position right at the time of contact with the wall
Well Z also goes sideways so its the same as X
I clicked like before I watched the video
Amazing :D
So since this game has unique values for everything, if I memorize and am able to predict the physics for this game, it wont help me calculate in any other game and would probably make me worse at those other games?
I wish it did curve that would be cool
Is there a plugin or some setting where the ball height/position is measured in real time? I'm doing a study of it for school.
I have a question guys,: If my car in rocket league is driving forward at a speed after i used 50 boost and i hit the ball, and some guy at a standstill just taps it, shouldn't i win the 50/50 due to the force? And wouldn't in reality the force i hit the ball with my car send the other guys car backward because of the force my car exerts on the ball? Thanks for the reply.
Does rotating your car give speed to the part of your car that you hit the ball with? like if I flat hit a ball vs air-rolling/flipping into a ball, does the rotation give extra speed? I don't think the physics works that way but Im not sure
I wish I had a perfect answer to that question. From everything I've seen from reverse engineering and such it does not seem to make difference for the strength of the impulse but there might be some part in how the unreal engine applies the impulse that causes the speed of the ball to be faster. (back when I made this video I didn't have as much info available back then so I was going off observation and setting up some very similar shots with and without dodging (car at pretty much the same speed) and was getting different power. But that might have other causes. There is new stuff I find out every week and I'm hoping that I will just stumble upon some helpful extra stuff so that I can give a satisfying answer.
Video about dodges is definitely a planned topic but probably not in the next 2-3 months.
Ah alright, thanks for the quick reply! Man, you really did create a channel over a really interesting topic, rl is such a physic based game that everyone who plays it has to wonder over the exact details at one time or another, and since the premise is so out there, the physics a lot of the time works differently than the intuition from irl. Kudos to you for the idea and then following through really well!
This is so fked up and counter intuitive, that's why every hit seems off, they should have made the hitbox a mesh wich becomes visible (like a force field) when hitting and would be much more clear with no made up physics.
I hope they realise this when utilising UE5. Has anyone noticed since EPIC took over, people miss the ball an get that 'strange' hit?
wtf
there is no physics after update.. badest engine @ the moment!
216km/h is not max speed of the boal. There were shots scored on rumble over 500km/h. Do your research well before making this kind of a video.
Thanks :) Maybe consider doing your research first before complaining about someone not doing the research. There is a mutator called "Ball Max Speed", which allows you to, well maybe you guessed it, increase the maximum speed of the ball. In a match without that mutator, the maximum speed is 6000 uu/s = 216 km/h.
LMAO "Do your research" says some random guy to someone who literally sank hundreds of hours gathering data experimenting and analysing the game. And not only that, but 5 years after the video came out and was battle-tested by the whole community. I mean, wow, just wow.