How to calculate VB2 time from video spins


Let’s say this are the ball times taken from video spin using avidemux
If we use original equation I published for FFA users
T= (0.86 x 1.08) / (2.5 x (1.08-0.86))
T=0.928 / 0.55
I assume 1.7. may not be good enough since FFA was defining ACC more precisely from few rotations clocked and your timings are not so accurate.
So let’s use a full equation and to increase precision use range from 0.6 rotation to 1.08.

T is the function to define reference time for VB2
t is time duration until ball slows down from 0.6 to 1.08s. Adding rotations for this example t=0.63+0.7+0.77+0.76+1.08=3.94s
t1 is time of first ball rotation 0.6
t2 is time of last ball rotation 1.08
R is rotor time let it be around 2.5 seconds / rotation

T=( t1 x t2 x t) / (R x (t2 - t1))

In steps
T=(0.6 x 1.08 x 3.94) / (2.5 x (1.08 - 0.6)
= (2.55) / (2.5 x (0.48))
=2.55 / 1.2
T =2.12s


Nice maths presentation Forester! ;D
I have been reading all the posts from the bottom up working my way to the top. I did watch your video first though. The maths bring it all together nicely.
Thank you for allowing me into this section.


Hello Forester,

why t1 time of first ball rotation is 0.6? because I see that the first ball rotation is 0.43
and t2 time of last ball rotation is 1.08? because I see that the last ball rotation is 2.10

before t1 are 7 ball revoltions, this means that is not the 1st ball rev
and after t2 are 4 revolutions more.


t1 and t2 are not the times of first and last ball rotations of the spin but first and last of a moment during the spin we looking to predict. Anyway on many wheels VB2 losing linearity after the ball is slower than 1.2 seconds/r because after that point the deceleration may drop.


So am i right to say that if you see the dealer is spinning the ball softly (first full rotation of ball already taking 1.1sec for example) vb2 simply will not work?
I ask this qn because on the video spins you uploaded on youtube titled “roulette video spins, test your skills”, no matter what reference time i try ( i tried alot, 0.6s all the way to 4 sec), the predicted number never matched up in the next rotation.


Not sure how did you do it. If you tried 0.6 then increased the time gradually to 4s you should see the point where the number would be the same regardless do you start a few rotations earlier or later.

If I remember correctly on that video the time is ~1.2s.

Apply it say about 9 rotations to the end read number.
Repeat the spin apply it rotation or 2 late , read the number, it should be close to same (it doesnt mean the ball will drop there).

Use another spin and test same .

After you can try to

shift starting point to match where the ball drops or stop also to compensate for rotor speed difference.


Thanks for the reply.
This is the video i was playing back with Avidemux. This is the wheel that i was unable to match up the numbers. Is it a problem with the wheel or Im just stupid? The deceleration of the ball is pretty constant throughout though.


That is not a tilted wheel with a dominant diamond, also if it is the reference time would be longer.


Hmm… Just for the sake of theory of reference timing, am i right to say that this wheel is not suitable for play (even if it is tilted) because the ball slows down too slowly compared to the rotor speed, making the reference time too large for play?


Sure not if counting time but with a timer, maybe.


And if talk about VB2 - all must only remember that it is the only way of measuring ball. So one of several. The main meaning of VB2 is to know wheel position when the ball will have speed which is about equal to reference time per rotation. Many who forget that they also must measure the rotor and do many other things…