MathLib.Lerp Bug

[ahmad.karami]

Hi Unigine Team,

 

I verified  MathLib.Lerp , this function have a bad bug.

 

If the distance from the origin to the destination increases, The shooting point to the destination happens with a fast time interval! this is not correct!
While this process should be uniform as a third parameter(speed*Game.Ifps).
For example :
The starting distance to the destination is 10 meters and the particle travels this distance in 2 seconds

For example :The starting distance to the destination is 100 meters and the particle travels this distance in 20 seconds

 

I think this function needs to be reviewed , or at least create another function, according to me, so that missiles and rockets can be designed.

 

Thank you

 

[christian.wolf2]

Hi ahmad,

actually your pointed out behavior from lerp is the expected behavior and not a bug. Have a look at the actual implementation:

v₀ + (v ₁ - v ₀) * k

 

With increasing distance between v0 and v1 becomes greater even for the same k-value. So to get your proper result you have two choices: Either modifying k-value resprecting the distance (maybe something like speed / length(v1-v0) * iFPS) or using a normalized direction vector that will be used for movement calculation.

 

 

 

 

[ahmad.karami]

17 hours ago, christian.wolf2 said:

Hi ahmad,

actually your pointed out behavior from lerp is the expected behavior and not a bug. Have a look at the actual implementation:

v₀ + (v ₁ - v ₀) * k

 

With increasing distance between v0 and v1 becomes greater even for the same k-value. So to get your proper result you have two choices: Either modifying k-value resprecting the distance (maybe something like speed / length(v1-v0) * iFPS) or using a normalized direction vector that will be used for movement calculation.

Thank you so much , I know this your tip, I tested LOT ,  I did Lerp in unity for 5 years , this issue is correct in unity but for unigine is not correct (MathLib.Lerp)

[silent]

Is there any chance to take a look at the minimal test scene where lerp is not working as you expected? Are you sure that you are not mixing lerp and MoveTowards (from Unity)?

Thanks!

[ahmad.karami]

1 hour ago, silent said:

Is there any chance to take a look at the minimal test scene where lerp is not working as you expected? Are you sure that you are not mixing lerp and MoveTowards (from Unity)?

Thanks!

Hello Mr @silent

MoveTowards function is exist for unigine in c#?

I tested lerp function a lot for unigine , but the result is strange, I'm working on it.

[ahmad.karami]

18 hours ago, christian.wolf2 said:

Hi ahmad,

actually your pointed out behavior from lerp is the expected behavior and not a bug. Have a look at the actual implementation:

v₀ + (v ₁ - v ₀) * k

 

With increasing distance between v0 and v1 becomes greater even for the same k-value. So to get your proper result you have two choices: Either modifying k-value resprecting the distance (maybe something like speed / length(v1-v0) * iFPS) or using a normalized direction vector that will be used for movement calculation.

 

 

 

 

Thank you @christian.wolf2 , I test your solution soon , I report result that here

[silent]

There is no MoveTowards in Unigine API. I'm wondering if maybe you somehow mistaken lerp and MoveTowrads? Anyway, a sample with a reproduction is required for sure to get the idea what is going on.

Thanks!