Please add drag and gravity adjustments

[m013]

I have noticed a quad in Velocidrone behaves much more realistic, when you do a throttle blip and drop it into an opening or gate.
Your weight adjustment - is nothing more than a band-aid, since it makes quad not only to fall quicker but adds an excessive feeling of driving a car on ice.

For the drag - i don't know how to explain it. A quad should loose speed quicker and shouldn't have insane acceleration like it has now.

Dont know if this possible now since you claimed the sim was in a final stage of development.
A half-backed overpriced product like velocidrone in not an option for me.

[Tomodaore]

Hello m013,

Thanks a lot for your feedback and thoughts about the simulator!

Liftoff is approaching a final stage, although we are still working on the physics of the game.

Gravity is right. As you know, gravity is really straightforward, a downforce that is equal to mass times a well-known constant. Gravity is as close to Earth gravity as it could get. Although it doesn't mean you're wrong about the "drop effect", just that it's not linked to gravity in the way you think.

We are working on this "drop effect", and our best guess right now is that it is actually your propellers generating negative thrust during a very small amount of time. To give you a little more details, a propeller generates thrust by accelerating air. Now when you are racing at high throttle, you gain speed, then if you cut throttle, the airspeed (approximately the speed of the quad) is higher than the speed the propeller can accelerate air (which depends on RPM), so the propeller is actually decelerating air, generating negative thrust until the speed is low enough, or until you change the throttle value.
All this means that if you fly at an angle <90°, going suddenly from high throttle to low throttle makes you lose speed and push you a little toward the ground. Gravity has only a small impact on this phenomenon.

About the drag, it is another very interesting but complicated topics. The scientific community don’t understand this force very well from a theoretical point of view, outside of really simplistic cases. In other words, it’s very hard to simulate realistically. What you are describing could have a lot of causes we are currently looking into like for example an inaccurate scaling of drag depending on the cross-sectional area of the drone considered in drag calculation or other aerodynamics interactions we don’t know about yet.

What would be a realistic deceleration for you? (given pitch, yaw, roll angles, speed and altitude)

[m013]

Tomodaore ,thanks for the reply.

About the 'drop effect'
I think velocidrone's devs got the idea correctly, more or less. Basically, it is working as motor idle percent in betaflight (if we are taking as an example dshot protocol) - it is a minimum value motors are allowed to rotate:

-Lower it - your quad will drop faster, but you will increase a chance of desyncs and rolls of death after a rapid roll/flip , because motors cannot spin up fast enough from too low rpms.
-Increase it - you will less likely get a roll of death, but your quad will fall at zero throttle like it has a parachute.

So, liftoff feels more like velocidrone at 1100 min throttle. But it s awkward to fly at this value.

About the drag. To be honest, i thought about it as acceleration/. A quad in liftoff accelerates like it is in on/off mode. It feels like a flight in a vacuum.

Dont get me wrong, i understand, different quads with different weight/thrust ratios feel absolutely different. When i fly my high power micro, it feels quite liftofft-ish, since it has lots of thrust and no inertia, deadstops are instant, unlike a heavy one witch weighs 400g w/o battery.

[Tomodaore]

I come back to you after performing some tests :)

Right now, Liftoff's flight controller is using a min throttle of 1070 (the same as Betaflight if I'm not mistaken). As you observed, it might be too high, so I basically measured hang time in free fall depending on min throttle for values between 1100 and 1000.

For this test, I used the mojo (5"), and a slightly different version of drag than the one in the game right now, but it should be ok to assess the importance of the idle throttle value. The free fall time is measured during a fall from a 200m altitude.

- Time before touching the ground (min throttle = 1100) ==> 12.6 sec
- Time before touching the ground (min throttle = 1000) ==> 8.4 sec

It actually makes a bigger difference than I thought! I’m definitely going to work on that !

Although, it is hard to tell if the difference would be noticeable in a race.

The nice thing about simulators is that they are more forgiving with FC going crazy because there is less environmental disturbances to the sensors, so a 1000 value is still fine.

About the acceleration issue, I am still pretty convinced that it is mostly a problem of thrust being too high, but I have yet to come up with a test to make sure of it.

I’m also super interested in the last part of your message. Based on your observation, Liftoff might overestimate the effective thrust to weight ratio resulting on no sensation of momentum being carried. That’s something I’ve been looking into for a while without being certain of that.

Again, thank you very much for your help. This kind of precise, constructive, feedback is really valuable for us, as it gives us real insights on how to make Liftoff a better game and simulator. It is awesome to have people lke you in Liftoff's community ! :)

[m013]

1070 - that means you minimum throttle value for motors is 7% , thats a lot IRL.
by default in betaflight motor idle percent is 4.5% for dshot, and it is a bit on a safe side. usually i set it to 4 and have no issues.