Liftoff Update Thread, aka "Is my quad too floaty?"

[Empty room with no elephants]

Here's my thoughts: It takes too long for your upwards velocity to slow down and it seems like hangtimes are extended. It's very easy to come to a hover after a freefall, and although I assume it's accurate to the ingame altitude/velocity counters, the acceleration downwards seems to be weaker than it should be. It takes too long after reaching the top of your ascension to start going down. The velocity seems to be correct (9.8meters per second per second), but it doesn't "feel" like it as well as it should. All in all an amazing simulator compared to its old versions, but the gravity feels a bit weak to me. I'd love some developer and other player insight into this and to see if I'm being overly harsh in my judgement, because I'm a relatively new pilot and I know squat. Also it runs totally fine on my system and the loadscreens for maps are minimal and better than velocidrone (!!!) PS - I'm sorry for the terrible video, first time XSplit user just for this recording and I couldn't be bothered to learn

https://www.youtube.com/watch?v=p399FkJ4Nrs&feature=youtu.be

[Empty room with no elephants]

"Thanks for the video! We do plan on still finetuning the drag and videos like this definitely help us tuning it just right. Contrary to what some people think, tuning gravity is never a thing we do, its all about drag.
Currently we also take into account surface area, so you will notice things feel different depending on the angle drop angle.

We are also exploring ways to take turbulence more into account in the future. That's a really complex aspect, aerodynamics, fluid dynamics,... With todays real time technologies, thats really hard to simulate, so we are looking at ways to replicate those effects instead. I suspect these might impact flight behaviour too in some specific cases. "

What do you all think? is the drag too much on a little quad in this sim?

[=W.A.B= Retal]

Still too floaty after latest update

I was asked by someone on the Facebook page Rotor Riot to provide some feeback so here it is:

The ability to "throw" your quad over objects. This is something that is missing from all FPV simulators (not just liftoff). So this is done in real life by punching out over an object, dropping the throttle to 0, then the feel of it dropping with the weight of the quad, it is actually really hard to explain.

BUT in the game, if you Punch out over an object, the moment you drop the throttle, it feels like the quad instantly looses most of its speed and you have to use the quads power to propel yourself over an object, and not inertia.

If I want to do a power loop, to achieve it, you have to unrealistically manipulate the throttle.

How it is supposed to happen:
1 fly forward into the object
2 throttle up and pull back on the pitch stick
3 cut power at the top of the loop (so that no motor power is directed down)
4 continue to move the pitch stick to maintain wanted subject in the camera frame
5 feed in throttle to arrest fall
6 repeat until skill mastered or until all funs have been had

How it happens in game:
1 fly forward into the object
2 throttle up and pull back on the pitch stick
3 cut power at the top of the loop
4 this is where things go wrong, there is supposed to be momentum backwards that allows you to realistically perform the trick
5 the quad doesnt seem to fall at the correct rate after throttle is removed

How to make look like real life in game:
1 fly forward into the object
2 throttle up and pull back on the pitch stick
3 DONT cut power at the top of the loop (maybe half the amount of throttle)
4 manage the amount of throttle on the way back down
5 catch the quad with the throttle to arrest fall.

[=W.A.B= Retal]

Totally agreed with the OP. It is definitely too "floaty" It also feels like the quad is lacking inertia. So, when you remove power, it feels like it doesn't carry the weight of the quad correctly. I dont know if this can be fixed by tweaking drag. But yeah, I think that aerodynamics and fluid dynamics may come into it. And I think that is why noone has got this aspect correct as yet.

[LuGus Studios]

*Merged both topics*

Thanks for posting this feedback. The team will take a dedicated look at it. The 0.10.0 update, while being a huge improvement, is not the last stage in the development of the simulation. Your feedback will be of great help to us.

[TheCodeHippy]

I think I understand what you mean. It's really difficult to get right though. If we give the quad more inertia, it will also be "sliding" much more in the corners. But yea, this is all aerodynamics and fluid dynamics. There is sadly no good physics model for a brick with 4 blenders strapped to it ;)
We're still going to be working on improving the effect.

[m013]

yes. it is like motor idle speed is set too high in betaflight for dshot600. very similar effect

[® 9centwhore]

Is perhaps your calculation treating drag as a constant instead of a variable value? That would explain it accelerating downwards too slowly at low velocities such as shown in op's video