Myrulet Visual Prediction System VB2

Further more,

Lets say ball of 1000ms per rotation and ball of 900ms per rotation will make same amount of rotations to the end.

When we observe, then the ball of 1000ms in 1.2 sec will make 44.5 pockets
The ball of 900 ms will make 49.3 pockets the difference for our prediction ~5 pockets. If rotor for these 2 spins were in same position 1000 ms may be predicted as zero and 900ms as 21.

If ball of 1000 ms arrives at DD and under diamond is zero,
900 ms ball will arrive a bit earlier because it is faster, so lets say when hits DD there will be number 28 there.

1000 ms ball will hit dd very low and drop instantly on zero.
900 ms ball will hit DD at very top (it has to be higher since stronger ball making same amount of rotations) since it hits top above number 28 it flies about 7-9 pockets and hits in front of 21. Prediction was 21, but and this ball has more chance that it will bounce more pockets then 1000ms ball since it hits rotor at narrower angle.

So the prediction increases by ball speed.
If another spin the ball is a bit faster (for analyses rotor is still same position and speed)

It may be predicted as number 17
It will arrive at our DD and there will be something as number 29 there.
Stronger ball will pass DD and go for another one.
Until ball makes ¼ of rotation in last rotation and rotor will make about 5 pockets. Total is 13 pockets. So the ball will hit next diamond and number 2 will be there, just in front of number 17. This ball will hit diamond low.

And another spin.
If ball is again is a bit faster when we start, we may get predicted 27
The ball after same amount of rotations will come to DD and it will be something as number 9 there. But this ball goes for another rotation, and hits DD (low) as a week ball.
Until ball makes extra last rotation and rotor will move for those 21 pocket difference.

So with suitable rotor speed we can take advantage of 2 diamonds which on tilted wheel could be 70% of spins.

Okey doke. I know what you are saying but i still can`t put an exact modus operandi in how you would apply it from start to the end. Maybe i will get it some day.

Applying it is the easiest part.
Let say you have a timer 1.2 sec and constant rotor.

Every spin all you do is, wait until the ball slows down, start timer and read number under the ball when time finishes.

That is ideal situation; however, as you know real play has many additional factors.
Rotor change, scatter, offset, no hits to DD etc.

With this approach it is harder to make rotor adjustments then with classical approach, but you get other benefits as I was explaining earlier.

To understand how it works the best is if you take one spin, define time then observe what is happening when the time is applied at different moments in spin.

On wheels with dominant drops, I don’t see why you focus so much attention on picking the ball reference revolution via a timing method.

To the average player the following is going to sound impossible and difficult to imagine. However, it’s not nearly as difficult as you might think.

My experience has shown that it’s much easier to just “look” at the ball spinning on the track. Meaning, if you study the wheel you’re playing, then you can rather accurately just “look” (estimate) at how many revs the ball has 7-9 or more revs out. It’s a simple enough skill that many people can develop it. This near instantaneous measurement will enable you to focus more attention on your rotor timings. You will also be able to react more quickly and blend in better at the table.

Try this experiment:

Go to a wheel with a drop, or create an imaginary drop for the test. Pick the dominant vertical deflector for your ball revolution reference point.
When the dealer spins the ball, count each revolution that the ball makes until the ball lands.
Keep counting this way for at least 30 spins.
After a while, try and guess how many revs the ball has left.
Work backwards from five revolutions out. Meaning, guess how many ball revs are left five revs out, then practice guessing 7, 9, and 11 revs out.
Work as early into the spin as you can.
Practice this on a target wheel for a couple hundred spins and record your results. You’ll be pleasantly surprised by your results after you get a couple hundred spins of practice. (If your serious about the game, put some real effort into developing your VB skills by methodically observing several thousand trials.)
If you can, find a wheel that has an older wood track on which to begin your experiment. An older track will have just a little bit of ball chatter to it. If you can’t hear the ball on the track just a little bit, then find another wheel for the experiment. Too much chatter is also bad.

Good Luck

-Snowman

Snowman, myrulet VB2 is superior to everything you described.

It is because it includes it all and in addition it corrects possible errors in it.

From what you have wrote I can see that instead defining particular ball revolution you suggest practicing to estimate it. So I assume you want to say it will give you better or at least equal results.

I can agree with your approach, since I argue about defining particular ball revolution as not very efficient approach. (One pocket mistake in observation and we get a full rotation error). It is imposable to have better accuracy then defining 2 rotations even if everything is done perfect. I say reality of such prediction will be most of the time within 3 rotations. That is why I can agree with you that someone with some practicing can achieve same with simple ball observation, how it is slowing down and which sound it makes.
But same applies to your described process. There is no way to be more accurate then 2 rotations, and most likely you will be within 3 rotations. (only FF with tilt R2 program can define exact particular rotation with sharp edge at the start and end.)

FFA is amazing tool for training as you described.
When I do predictions in tilt mode FF after prediction will say and A, B, C which indicate ball strength within ball revolution.
When I do spins test after some time most of the time by listening and observing ball speed I can know when system will predict and even will it say A,B, or C. That is something, since difference in between spin marked with A and B is about 60 ms in ball speed. But that is at kitchen table, where there is no time in between spins and where the system always reminds me about particular moment in the spin.

When I am in casino pauses in between spins could be 5 minutes. So it is hard to get tune in to spins. Because of waiting to long in between spin I can’t do it. Even if I can after some time of play it may change.

Myrulet VB system is one step above of what you described, so when you already achieve that it helps you to correct mistake if you made wrong estimation.

So if I can do same as you described and if I can estimate within 3 rotations the system will do the rest. 3 rotations estimation is not good enough, it will make results float 27 pockets if rotor is only 9 p/s. VB2 from that 27 pockets error will reduce it to only few. Even if I am within 5 rotations I will still get good result.

Yes, I explained potential problems with rotor speed changes but neither clasic nor your approach is good enough to handle rotor speed changes especially if prediction is early in time. If you predicting in 7th rotation and instead of that you get it in 8, and if there is rotor speed increase you will get double error. One from wrong rotation the other one from adding rotor speed change to wrong time to the end.

With VB2 is not that there is no enough attention on your rotor, it takes less time for prediction then any other VB, and problem is that system may generate errors as I was describing.
It would be easy to say for me yes , yes I can. But that is not reality.
I play VB2 a lot, when rotor was reasonably constant I manage to get profits, sometimes amazing amounts. When rotor is fast and even worst if it constantly changing I am struggling. If rotor changes from 5 sec. to 2.5 sec. per rotation regardless do I get more errors in prediction, advantage is reduced simply because and ball scatter gets more unpredictable. But of course when everything is faster and prediction errors will increase.

First off: The wheel must have a dominant drop. Level wheels can’t be beaten.

For starters, I can usually pick the rev. If I’m off, I’m off by only one on the “far” end.

Here’s an example: These are just numbers used for the example, not actual.

A 4 second rotor will make 4 revolutions when the ball is 9 revs out from the end.
A 3.5 second rotor will make 4 revolution + 10 minutes. 9 revs out from the end. (Note: A clock is 60 minutes)

With a four second rotor, even if I miss the ball reference revolution, I’m not likely to miss by more than 5 pockets with a perfect drop. The key word is ball reference revolution.

I can also make this ball reference revolution measurement nearly instantaneously.
If I look and see the ball is likely 11 revs out, I can adjust accordingly. I don’t have to wait for it to “spin” down.

I agree that the computer is more accurate, however, it’s slower.

-Snowman.

.

First off: The wheel must have a dominant drop. Level wheels can't be beaten.

Yes for VB, but leveled wheel can be beaten only you do not know how. :smiley:

Well from what you described (4 rotor rotations when rotor is 4 sec) you are having prediction about 16 sec before then the ball drops. Sure if at that ball speed you can define right moment within 2 rotations, because time in between rotations is smaller that your error will be smaller.

It is not important by how many rotations you miss but by how much time you miss.

More time you miss there will be different time to the end that will affect rotor.

What are you saying is “yes I can do it within 2 rotations”.

But you also saying that you can spot particular moment within spin with accuracy of about 0.5 sec. Because that is the difference in between 2 rotations at about 16 sec before then ball drops.

Are you Superman?

If you can do it then yes your error in between one or the other rotation can be smaller. Because you have to the end of spin 16, or 17 sec for rotor to move.

I think you rotor sample would be more suitable 3 and 3.5 rotations to the end if the ball is making 9 rotations. Because 9 rotations is more 12 sec to the end, but of course that may be different on different wheels.

At about 9 rotations to the end rotations are 0.7s long.
2 rotations would be 0.7 +0.8=1.5 sec
On 10 pockets per sec rotor it means that you will float within 10 x 1.5 = 15 pockets of error even if you have a skill of predicting within that time.
VB2 will at least half it.

Assuming that you counting time with accuracy of 100ms.
Max error that you can get with VB2 will be 6 pockets.

If you use timer then it is reduced to as much as good you can observe.

But that is not all, if you miss to estimate moment in the spin within 1.5 sec, VB2 will again correct error.

No the error on the rotation is approximately 5 pockets at 4 seconds. (Again, ball ref revolution.)

And no, I don’t think you can predict level wheels, even with your computer. Your rotor clicks and reaction times probably aren’t accurate enough to acheive anything but a very tiny edge.

I’m sure your computer is very good on wheels with a dominant drop. There are many that are.

The only way is with a dominant drop.

No the error on the rotation is approximately 5 pockets at 4 seconds. (Again, ball ref revolution.)

Ok, only and only if you can identify when the ball is at your targeted point within 0.5 s.

And no, I don't think you can predict level wheels, even with your computer. Your rotor clicks and reaction times probably aren't accurate enough. That is unless you're Superman.

It creates only few pockets of error. Ask Bago, you know that guy hates me and he has FFA now.

I'm sure your computer is very good on wheels with a dominant drop. There are many that are.

I believe with introduction of “point set” it is the best.

The only way is with a dominant drop.

It is not the only way, with leveled wheel play we may get less accurate predictions, but we may get better ball scatter that matches better with prediction. On tilted wheel the ball is hitting diamond at wider positions.

The difference is that with reasonable scatter on every tilted wheel we can get advantage. But that doesn’t have to be on leveled wheel even if the system predicts 100% according measured parameters and if measuring was accurate.

I guess will just have to agree to disagree.

Again, I’m sure your computer is very good.

I’m not knocking your system in anyway.

Best of Luck.

-Snowman. :slight_smile:

That’s so generous.

If you have single rotation accuracy as FF does, then you can have error within 5 pockets.
If you toggling in between 2 rotations then it is 10 pockets. But to estimate time within 2 rotations on such fast ball where differences in between rotations are only 50 ms is a Superman’s skill. I can’t do it.

With 3 times slower ball where differences in between rotations are 4-5 times greater I can, but it is hard to keep it constant in real play.
Since I am not a Superman, that is why I have VB2 to give me same result as Superman would have.

Lets say that ‘T’ is the time of 1 ball rev.
And based on T you calculate the ‘t’, which is your observing time.

Which T do you take for calculating t? 1s, 1,5s, 1,6s, 2s?

When is the right time T?
How to determine the right time T?

i suppose, you have to catch linearity you mentioned.

But how?

Linearity is everywhere, but it always matters for how long.
If you play when ball is slower then 1.2 sec. The linearity is shorter, because ball deceleration most likely is dropping.

Put it this way.
You estimate when the ball is somewhere 10-12 sec before it drops. Then you start VB2.

VB2 regardless did you start 10 or 12 sec before then ball drops will predict accurately.

If it is calculated properly, probably anything in between 8-14 sec will be reasonably good.

In usual I used only 2 time settings in casino.
1.5s and 2.s, that’s how FFZ was programmed. Only recently I made it adjustable. But most of the time I do not use timer so I just count fast from 1-10 to get reference time.

If you have rotor about 4 sec.
If you use 1.5 s as a [t] you will be ok.
If you can manage to estimate when the ball is about lets say 9-13 sec before it drops.
That may be 5-6 rotations wide window.
If you just estimate in between 9-13s and at your estimated moment read number under the ball you can be wrong by 37 pockets if you do not apply VB2.
If you do, even if time [t] is not perfect you will be only few pockets wrong.

So listening for particular sound when ball is spinning to estimate right moment will never match VB2 with accuracy.

Rotor.
In real play I do not calculate rotor as it was done in example.
I observe.
I like to use same time that I use for ball and for rotor observation. Other way it may be confusing.

But using 1.5 sec for rotor is not much.
So if we observe rotor in 1.5 sec. and if it makes 3 pocket more. That means about 2 pockets per sec. if remaining time is 10 that makes 10 pockets.
But remaining time may be 13s or anything.
So I observe, if with 3 pockets faster rotor the ball drops 15 pockets further then expected, in my play I just add 5 pockets for every extra pocket that rotor makes.

I do not get prediction then add lets say 15 pockets.
I simply shift my starting point by 15 pockets.
Sometimes instead of that I simply count for longer.

Avoid playing fast rotors.

Recenlty I played Cammegh Connoisseur tilted; I was amazed with result of prediction. When dealer changed he was really ace with rotor. I tried but I could not progress so I stopped playing. Lucky that in first hour I made very nice profit.

Next time I played VB on Huxley wheel, it wasn’t good at all.
I wasn’t playing big only ~10-15 chips trying to tune.
But for 30 spins I had only one proper hit on sector, few small wins and many spins the ball stopped 1-2 pockets from my played sector.
One spin I have had 5 units on zero and 5 split 0-3, the ball stopped on 26 where I have had nothing. Very frustrating. I stopped playing and waited for friend.

Next 15 spins I watched the wheel.
I have had 5 straight hits on single number, 3 missed by 1 pocket and one missed by 2 pockets. Only one spin was really bad predicted. Unreal, that was better then 1:6 hits.

If you playing very different wheel where you want to target ball of 1.5 s.
Then you need to get time difference in between rotations when the ball is 1.5 s.
With FFA that would be Accuracy 4, and system will tell you.
But that is the difference in 1.5s so whatever it is it needs to be divided by 1.5.

37/ (rotor rime in sec) =(ACC x 37) / (targeted ball speed) ] x t

37 / 4 = (0.200 x 37 ) / 1.5 x t

I took for acc 200ms =0.200s

9.25=(7.4 / 1.5) x t

t = 9.25 / 4.93 = 1.86 s

If you want to target when the ball is 1.5 s and if you still have enough time, your ACC might be smaller then 200 , more likely 150-160. So 2 -2.5 sec would be good.

On Cammegh Connoisseur 4 sec rotor I used 2 sec.

This is a curve how ball decelerates on Huxley. I got this data from FFA.
Notice how after ball speed is 1.3 sec the ball deceleration is dropping.
So with VB2 we need to have prediction earlier then when the ball is 1.3 sec.

Let me give an example:

Save this video http://www.youtube.com/watch?v=Zf0ev4h-zfk in HQ.
Do it with help of this page: http://keepvid.com/?url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DZf0ev4h-zfk

Lets focus on the first spin.

Ball drops at 54s. Lets go 11 sec back to 43s. Exatcly 43,120s when the ball is at 12o clock.
We have 3,5s rotor.
We have ACC = 200ms
And our 1 rev ball time is 1520ms = 1,5s (from 43,120 to 44,640)

That means our reference time is t= 1500/(3,5x200)= 2,15s

At 44,640+2,15= 46,8s under the ball is number 4.

It is hard to see, where the ball drops because it is behind the edge of the wheel at that moment.

Am I doing this correct?

I still think, this wheel is strange…

On that wheel FFA says remaining time 9 sec , ACC 158

t= 1550/(3,5x158)= 2,8s

2.8 sec would be hard to count, have to use timer.

Yes on the wheel ACC is low. Slower the ball decelerates it is harder to notice differences within ball rotations.
158 on 1.55s long ball rotation is only about 100ms per sec.

I set the FFA to accuracy 2
Then I got remaining time 15.2 s and ACC 95

It means that ball deceleration when the ball is 1 sec or 1.5 sec is same.
Problem is that deceleration is only about 100ms/s.
It means that whatever we do we will be ½ accurate as if deceleration was 200ms/s

It also means that on this wheel we can have linearity and when ball is 1 sec and when it is 1.5 sec. On Huxley we can not, as you could see what is happening after the ball is 1.3 s. Rotations are longer but difference in between them is smaller.

With FFZ I did not bother to have prediction where ACC was smaller then 120 or higher then 255.

You will be better of if you download stefano’s spins or download this

http://rapidshare.com/files/213534308/Spin_2.avi.html

http://rapidshare.com/files/213560349/Spin_2.avi

start … end … prediction
2220 … 4220 … 29
2780 … 4780 … 29
3360 … 5360 … 29
4000 … 6000 … 29
4640 … 6640 … 29
5360 … 7360 … 7
6120 … 8120 … 29
6960 … 8960 … 18
7920 … 9920 … 7
9040 … 11040 … 26 >>> start losing linearity
10280 … 12280 … 25

Times when the ball starts at V diamond at 11 o’clock.
First column is when we start, next one after 2 sec, and last one is predicted number.

You can see that accurate predictions are possible from start of the spin.
And how last 2 results lose linearity.
So we really have 7 sec or more time window to estimate for prediction to start.

If we estimating within 7 sec. that is 2 rotor rotations and 74 pockets of error.
With VB2 we can be 7 sec wrong and still have results within 2 pockets of accuracy.
If we look that in rotations we can have prediction in 9 different rotations.

If we had prediction 9 sec to the and or 16 sec it doesn’t matter. Of course if rotor changes it is harder to adjust it for 16 sec.
Player should be able to estimate ball within 3-5 sec in spin, then even if time [t] is not so perfect he will have reasonable results.

to view times in slow motion download free program avdemux
http://rapidshare.com/files/213557475/avidemux_2.1.zip.html

http://rapidshare.com/files/213557475/avidemux_2.1.zip

start … end … prediction
2220 … 3720 … 12
2780 … 4280 … 35
3360 … 4860 … 0
4000 … 5500 … 15
4640 … 6140 … 21
5360 … 6860 … 2
6120 … 7620 … 2
6960 … 8460 … 25
7920 … 9420 … 6
9040 … 10540 … 10
10280 … 12280 … 14

And this is when I used 1.5 sec, so you can see possible errors.
If we estimate within 3 rotations error is about 5 pockets.
Still can’t complain.

start … end … prediction
2220 … 4720 … 9
2780 … 5280 … 31
3360 … 5860 … 20
4000 … 6500 … 33
4640 … 7140 … 5
5360 … 7860 … 23
6120 … 8620 … 30
6960 … 9460 … 36
7920 … 10420 … 11
9040 … 11540 … 10
10280 … 11780 … 31

This is with 2.5 s
I marked in red where most likely we would play.
That is 5 rotations to estimate and error is still within 6 pockets.

Also notice how is important to have prediction earlier then the ball is 1.1s/r
On this wheel that is where about is “knee point” where the ball deceleration start dropping.

So how to get ACC if you have FFA.

Select accuracy 2 (iqe6 or tilt system it doesn’t matter)
Run few spins and system will tell you.

It might be even better to select accuracy 1
Get information from FFA then divide it with 0.75

This would be exact calculation

T1=(745 + ACC) / [{rotor in sec.} x ACC] ; for system set in Accuracy 1
T2=(1000 + ACC) / [{rotor in sec.} x ACC] ; for system set in Accuracy 2
T3=(1255 + ACC) / [{rotor in sec.} x ACC] ; for system set in Accuracy 3
T4=(1510 + ACC) / [{rotor in sec.} x ACC] ; for system set in Accuracy 4

Other way is to record video spin and to observe it in slow motion.

When i run this spin in accuracy 1 i get system tell me ACC=125 which makes time 1.98s
In accuracy 2 ACC=170 and calculated time 1.96 (rotor3.5s)

That is why 2 sec, applied time suits so well.

This is how the ball deccelerates per rotation for that spin, measured by FFA.

It is very simmilar to previous graph but it shows ball decceleration when it is 750 - 1650 ms

[attachimg=1]