VB2 math - questions

MyRulet visual prediction -VB2
1

Hi Forester

I’ve been trying to get my head around the VB2 method as outlined here and I have one or two little problems.

Firstly, why should it only work for dominant diamond wheels? The formula would seem to be generic.

Secondly (and much more importantly), unless I have read the formula incorrectly (quite possible!) it would seem to be unbalanced. By this I mean the units of measurement involved. If I can try and expand my meaning by describing the formula in uom (units of measurement) and then using the numbers given as examples, we get -

c/sec = [ (ms * c) / ms ] * T sec

or

37 / 4 seconds = [ (0.2 seconds * 37) / 1 second ] * T

Doing some basic algebra on the first above we end up with 1/sec = sec which would indicate an invalid formula. In order for it to be valid we should end up with sec = sec, or 1/sec = 1/sec, or some such.

The second simplifies down to

T = 1 / (0.2 * 4 seconds) or 1.25 (secs**-1)

Note: (secs**-1) means seconds to the power of minus one, or 1/seconds.

Perhaps the T is on the wrong side of the equation? In any case, something is wrong with this.

Even taking the simplified version t=1000 / (4 x 200 ) it doesn’t work out. Let’s put in the units here -

t = 1000 ms / (4 seconds * 200 ms)
Cancelling the ms uom we get
t = 1000/(800 seconds) = 1.25 (sec**-1) NOT 1.25 seconds.

In order to get a mathematically correct answer we need to invert the numbers - i.e. it should be 1/1.25 or 0.8.

I hope I have made myself understood. I just want to try and understand the method so I can learn to use it properly, but unless the math is right I guess there is little point.

Bryan

2

ACC 200ms is not really ms but 200ms/s^2
I assume then your units will match, it is only simplified to get the number.

(v1-v2)/t

But instead of v (ball speed) I just use times of rotations.

Firstly, why should it only work for dominant diamond wheels? The formula would seem to be generic.

Because of left hand side of formula is how much the wheel moves per sec in pockets
On the right side (without t) is how much ball slows down per sec in pockets.
So it is valid only for tilted wheel.
If for example we are 2 rotations earlier in time
We want our “t” to match that ball slows down in that time same amount of pockets as rotor makes for that 2 rotations.

If we want result to be equal and for leveled wheel then we must use not only rotor but and the ball. So on left side of equation it will be how much rotor makes in 2 rotations + 2 x 37. We can’t get that on right hand side because ball can not slow down so much.
That is why E2 system uses multiplication x 3. E2 and Tilt2 have same principles.
With E2 if rotor makes in 2 rotations 16 pockets together with what ball made it would be
16+2x37=90

In about 4.5 sec we get ball slows down by 30 pockets. Multiplied by 3 it brings us back to 90 and we read same reference number as if we did predict 2 rotations later.
2 rotations doesn’t have to be 2 rotations it can be any time where linearity of system applies.

So this is the theory.
In practice the best is adjust time lets say to 2 sec.
Run timer when the ball is about 0.5 sec / rotation
Read number
Then run the timer again and check if result is same or very close.
If it is then is good to play if it is not then try to readjust time.

3

Aha - of course - ACC is an acceleration figure (or deceleration actually, the sign not being relevant). Sometimes I analyse things to death! But I’ve always found that the better I understand something the better I can put it into practice. There are adjustments that can be made when you find the constants are different but still constant - for example, the wheel speed might be reliably 3 or 5 second rotations. A simple recalculation for t and away we go!

Being a musician, amongst my tools I have a wrist metronome which can be set up for the time needed. For example, if you wanted to measure 1.2 seconds then you calculate the metronome speed by simply dividing 60 by 1.2 to give a metronome setting of 50 beats per minute and bingo - you have a perfect method for measuring the time interval.

Is it best, do you think, to use the dominant diamond as your first reference point? I’m just not sure about the actual implementation methodology. Or is it just necessary to use the same reference point each spin and your estimate of the bounce offset handles the difference?

thanks
Bryan

4
There are adjustments that can be made when you find the constants are different but still constant - for example, the wheel speed might be reliably 3 or 5 second rotations. A simple recalculation for t and away we go!

It is partly correct, VB2 doesn’t like much rotor change.
If you readjust time it is one way to go but on small differences.
If rotor is 4 pockets per sec faster to the end it may be 40 pockets
To adjust time for that we may add 1000ms but it wouldn’t match fully with linearity.
So keep play with reasonably constant rotor, and prefer slower one.

Being a musician, amongst my tools I have a wrist metronome which can be set up for the time needed. For example, if you wanted to measure 1.2 seconds then you calculate the metronome speed by simply dividing 60 by 1.2 to give a metronome setting of 50 beats per minute and bingo - you have a perfect method for measuring the time interval.

Is it best, do you think, to use the dominant diamond as your first reference point? I’m just not sure about the actual implementation methodology. Or is it just necessary to use the same reference point each spin and your estimate of the bounce offset handles the difference?

You can use watch for rotor timing to check differences,
For example how much it changed in 2x1.0 sec,
If for example it changes 2 pockets and if your predictions are about 12 sec before then ball drops add 12 pockets.

I start with DD and observe few spins.
Lets say in 2 sec rotor makes 20 pockets
I get predicted zero ball stops to 34 (average jump CW)
Next time I shift my observation point by 9 pockets CW
If rotor this time makes 22 pockets I shift observation point by additional 10 pockets.

Sometimes instead of timer I count fast so I just add to count 1 or 2.

If you are using metrowatch it would be hard to start time at desired moment.

Undestanding VB2.
Let’s say rotor is 10p/s
We want to play VB but we do not know how.
We estimate to read number at same moment in spin.
But how accurate we can be with it?
1,2,3,4… or more rotations when the bal is faster.
Lets say with some practice we can do it most likely within 4 rotations, which in total makes about 3 sec.
We know that in 3 sec rotor shifts 30 pockets so our reference number will float with possibility of + or minus 15 pockets.

We apply VB2 even if time adjusted is not perfect 3 sec time window will not change systems linearity by much. 30 pockets possible error we reduce to only few pockets of error. Few because of linearity and few because of errors when reading number under the ball. So it is much better. Remaining is rotor adjustment but that is same problem no matter what we do.

5

No question about VB2 is silly but what surprises me and a bit disappoints me is this, I found on net discussion about VB2 and E2.

Written by Kelly:

“Bottom line: It MUST work in the casino and not in slow motion at the computer. I don`t wanna be wrong because i added 2 angles up to 96 degrees instead of 100.”

There is no really difference is it 96 or 100.
We can look wheel as 360 deg. Where each 10 represents 1 pocket.
96 or 100 is only mistake of 0.4 pockets. If I see it within 50 deg that would be only 5 pockets error.

I am disappointing because in this forum no matter how many times I explained him he just couldn’t get it.
On the end he finally did, but he said, oh that is the old way, someone long time ago already published it. Now he claims again that he doesn’t understand it. :-\

What I want to say here is : if you do not understand, ask as many questions as you like.

When we had disagreement about traditional VB Kelly decided to not publish here any more.