PRO ONE DVD: Answering Questions

[LAMas]

LAMas -- I certainly do admire your persistence in looking for alternatives.

With your no-shift/no-pivot "phase shift" approach, are you really now talking about secondary alignment lines, or just single alignment lines. In other words, are you still sighting CTE plus those other lines, or are you kind of ignoring the CTEL and just sighting the other, single line. Certainly for the last few alignments, I doubt you could simultaneously sight two lines, as the CTE would want to be way off into space past the OB.

So it kind of sounds to me like you're just arriving at a straightforward x-angle aiming system where you aim the edge of the CB at one of 8 reference points on the OB: 1/4, 3/8, 1/2, 5/8, 3/4, 7/8, 15/16, and edge-to-edge. So that would give you 8 cut angles in each direction, plus straight.

If you tried to overlay this on top of, say, Stan's CTE method, producing offsets and pivots for some shots but not others, I think it might be hopelessly difficult to discern, distinguish, and implement all of the menu choices it would present.

I address the pre shot as with other aiming methods like DD by lining the CB in front of the OB and establish that line which can be center to center or center to edge (CTE) or center to GB to establish the cut angle to the pocket/target. Since a user of CTE cannot see the GB :wink:, I am using CTE for it is midway between thick and thin cuts, whereas C to C is more at the thick cuts.

I then proceed to establish the secondary aim line from the edge of the CB to the fractional points on the OB. I use stick aiming oon this line and even close the other eye to verify (which eye is aiming) so that I am at the position/stance directed by my aiming eye. I still do not see the utility of using the CTEL after that. Other CTE experts have directed to roatate the body and eyes at 45 degrees to see both lines, but I cannot do that due to double parallax - I have enough problems with single parallax.:wink:

I stand square to the shot with the cue between both under the center of my chin. My eyes happen to be 2.25" apart (diameter of a pool ball) so my cue is stroking 1.125 (half ball) to the side of the secondary aim line below my aiming eye. This give me a consistent stance and a geometry that I can diagram in ACAD.

"...So that would give you 8 cut angles in each direction, plus straight.", I also add the CTE line that is the ubiquitous 30 degree cut angle along with the straight in C to C like Hal.

The other half of this study, for cuts to the left, is the opposite 1/2 tip offset to the right and pivoting left that yields the following going to aim spots on the OB from left to right:

Edge = 16 degrees
1/8 = 20 deg
1/4 = 24 deg
3/8 = 28 deg
1/2 = 32 deg (same as 1/4 no pivot)
5/8 = 36 deg
3/4 = 41 deg
7/8 = 43 deg ( same as 1/4 pivot right)
8/8 Edge = 52 deg

There is generally, cut angles every 4 degrees apart that may be good enough for OB near the 4.5" wide pocket.

Thanks for your interest.:thumbup:

 

[LAMas]

LAMas -- I certainly do admire your persistence in looking for alternatives....

If you tried to overlay this on top of, say, Stan's CTE method, producing offsets and pivots for some shots but not others, I think it might be hopelessly difficult to discern, distinguish, and implement all of the menu choices it would present.

I concur.
It would be very hard to memorize except for the most persistent or conscientious.

"Your wish is my command"

 

[AtLarge]

... It would be very hard to memorize except for the most persistent or conscientious. ,,,

Eeeks! You're making me dizzy. Just find the intended contact point on the OB and then hit it with the equal but opposite point on the CB. Nothing to memorize.

I get the feeling you like fiddling with that AutoCAD stuff.

Cheers!

 

[LAMas]

Eeeks! You're making me dizzy. Just find the intended contact point on the OB and then hit it with the equal but opposite point on the CB. Nothing to memorize.

I get the feeling you like fiddling with that AutoCAD stuff.

Cheers!

CP to CP is also in the pic as is DD.:wink:

 

[JoeyA]

works perfect! Great Job!!!!!!
tried them all.

Now you need a few more reference points to compensate for throw for example when balls are close together, when you are playing slowly, or when you use a punch stroke.

Don't you ruin my potential sellings of my aiming system that I want to bring out ....(worked on it for months) :grin-square:
Anyway it's a pivot system....
Best from EKKES

Whatever your pivot system is, I WANT TO KNOW ABOUT IT. Count me in.
JoeyA (UBC seldom ever misses a ball- It must be his "SECRET" pivot aiming system).

 

[jwpretd]

If your robot emulates a human being, then the "visual intelligence" for CTE is built into the robot (and it may even be doing things that a player using CTE could not).

Well, it's not clear to me that we know what a player using CTE (or anything else) actually does.

[Aside: If you were trying to build a robot to pocket balls without further human instruction, I don't think you'd try to program it to do CTE or Pro-One maneuvers. You'd simply have it find the GB position (or the base of it, or the center of it, or ...) for the shot and then shoot the CB into the GB position.]
​

Yes, that's exactly what you do. A couple of weeks ago, there was a thread about pool-playing robots. In that, I posted a link to a video of an "R2D2-like" robot that uses a bridge and cue like a player does (as opposed to the gantry style). In that same post is a link to the source code for the robot.

My robot is ignorant of the precise cut angle needed for a shot (because so is a player)....

So how would your robot find the ghost ball as you advocated above? Would the operator do that for it?

Also, the contention that the player doesn't know the cut angle is questionable, and there's considerable evidence against it. How does any player using any system manage to pocket a ball without knowing the cut angle?

 

[AtLarge]

Well, it's not clear to me that we know what a player using CTE (or anything else) actually does.

Agree!

So how would your robot find the ghost ball as you advocated above? Would the operator do that for it?

No, my robot does not find the GB or know where it is. Nor does the operator. As I said, "it's just a machine that responds to a "call-out" of one of the alignment-menu options, and it then performs the set of instructions for that menu choice. And it does it exactly the same way, with precision, every time that menu item is called. So, yes, a human has to decide which menu item or instruction set my machine should use for each shot." But the human does not know the actual angle, either. He just chooses that instruction set, e.g. "CTEL to right; B secondary alignment; right pivot," based on his sense of the thickness of the cut needed.

Also, the contention that the player doesn't know the cut angle is questionable, and there's considerable evidence against it. How does any player using any system manage to pocket a ball without knowing the cut angle?

It may just be a matter of semantics as to whether we mean the same thing by "know the angle."

A player can sense the cut angle, because he is looking at the two balls and the pocket. But an aiming method need know nothing about the actual angle in numerical terms.

A true "feel" player, for example, just senses the amount of overlap needed to pocket the shot. "A little more, a little more, ... oops, back a tad, that's it -- shoot."

In fractional-ball aiming, the player just senses, or recognizes, which of the alignment points to use by looking at the shot. He may then go a little thinner or thicker on that initial choice depending upon his experience in looking at and shooting such shots. He might conclude "It's a thin 3/4-ball shot," for example, rather than identifying it as "18 degrees."

With contact-point-to-contact-point aiming, the player just finds the intended contact point on the OB and then tries to hit it with the equal-but-opposite point on the CB (or some similar technique). The player need not know the actual angle. Sure, he knows whether it is real thick, thick, medium, thin, real thin, etc., because he sees the balls and the pocket. But operationally, he is just shooting one point at another in a similar fashion for every shot regardless of actual cut angle needed.

I am aware, however, that some people are able to identify the actual cut angle needed to the nearest couple of degrees. There are techniques for doing that as you stand looking at a shot. But I'd have to consider it extremely rare for someone to base his aiming on first determining the necessary cut angle numerically and then having a mental "table-look-up" for exactly how far off the OB's center to aim to produce that cut angle.

 

[Patrick Johnson]

LAMas:
There is generally, cut angles every 4 degrees apart that may be good enough for OB near the 4.5" wide pocket.

Here's a drawing illustrating the methodology for figuring this out. It shows that to make a spot shot from anywhere on the table into a 4.5" pocket takes ~25 discrete cut angles per quarter ball (per cut direction), each ~3.6 degrees wide (contact area ~1/16" on OB's surface).



What does this mean?

If the OB is left in place and the CB is moved around it in an arc, the cut angle needed to make the shot changes. If the pocket was exactly as wide as a ball, then the cut angle would have to change with every infinitesimal movement of the CB and it would take an infinite number of cut angles to make shots from all possible CB positions. But with a 2.25" margin of error in the pocket, the cut angle only needs to change with every 3.6-degree movement of the CB (25 times as the CB moves through a 90-degree arc for all the cuts in one direction).

This is why it is often said that any system must define more than a handful of cut angles in order to work "without adjustment". For example, if a system defines only 6 cut angles for each cut direction, then the system by itself can only make 6/25 (~1/4) of all possible spot shots into a 4.5" pocket, and the other 3/4 of all possible cut angles are in the gaps between the 6 system-defined cut angles.

pj
chgo

 

[jwpretd]

It may just be a matter of semantics as to whether we mean the same thing by "know the angle."

A player can sense the cut angle, because he is looking at the two balls and the pocket. But an aiming method need know nothing about the actual angle in numerical terms.

A true "feel" player, for example, just senses the amount of overlap needed to pocket the shot. "A little more, a little more, ... oops, back a tad, that's it -- shoot."

Yes, there is a semantic issue. Also, I have no real objection to the second paragraph. And (didn't quote the last part) I also doubt that anyone thinks of the cut angles numerically.

The third paragraph describes calculation by "successive approximation" (or maybe "estimation" in this case). Such methods are heavily studied, and are used all over the place: economics, ADCs, extracting square roots (known for a couple of thousand years, at least), etc.

But that just makes the question be "What is the meaning of 'senses'?" in this context. Certainly a robot need not calculate exact angles. It could do what I have a feeling people actually do: compare the visual image of the current shot with stored images (memories) of past shots, using whatever information is provided by whatever aiming system they're programmed to use. (In particular, good 3C, one pocket, and banks players seem to memorize a lot of "template" shots that they modify on the fly to fit particular on-table situations.)

I don't think running the robot as you proposed would tell us much. Suppose the robot shoots and fails to pocket the ball. So we record that and put the balls back in place. And then Mikjary comes in, looks at the shot, says "That's trivial.", picks up the cue and casually pocets the OB. Have we learned anything except, maybe, that the robot for some reason didn't align itself the same way Mikjary did?

By the way, I had the robot fail to pocket the ball for a reason. If it actually made the ball and we told anybody, we'd have people beating down the lab door to tell us that no matter what we thought we saw, the ball didn't really go in the pocket. Then they'd stand around cluttering up the lab while they figured out why we thought we saw the ball go in the pocket when it didn't, and they'd drink all the beer and complain because we hadn't ordered pizza.

 

[Neil]

For a robot to test CTE, you first have to figure out how to get the robot to visualize like a human can. Good luck with that one.

 

[AtLarge]

... I don't think running the robot as you proposed would tell us much. ...

John, my use of the word "robotic" in my explanations of manual CTE is simply to say that if you, or I, or a "robot" were able to perfectly perform the discrete number of prescriptions given by Stan for that aiming method -- without "visual intelligence" or adjustments or "feel" -- that you, or I, or the "robot" would produce only the same discrete number of cut angles at a given CB-OB distance. I'm not talking about trying to design a robot to perform CTE either with or without "visual intelligence." It's just a way of saying to carry out the instruction set accurately and repetitively.

 

[AtLarge]

Here's a drawing illustrating the methodology for figuring this out. It shows that, to make a spot shot from anywhere on the table into a 4.5" pocket takes ~25 discrete cut angles per quarter ball (per cut direction), each ~3.6 degrees wide (contact area ~1/16" on OB's surface). ...

Interesting graphic and facts, Pat. And that's for a ball that is only about 35 inches from the pocket! If the ball is considerably farther from the pocket, you need considerably more cut angles.

 

[jwpretd]

... every week my perception of the visuals gets stronger and quicker and more seamlessly blends with my routine, I am trusting the shots more and more, learning new ways to use it for banks, and overall playing better. Can't imagine going back to the way I used to aim.

I know that has nothing to do with how the system works, just saying...

Knowing that your ability to visualize gets stronger with increased use is helpful. It's what's to be expected from something so strongly based on human perception, and confirmation of expectations is always useful. It gives more certainty about what an explanation of a system must account for. And can sometimes be a warning that a cause might have been mistaken for an effect.

It'd be interesting to know if your perception of other elements becomes stronger over time.

 

[jwpretd]

John, my use of the word "robotic" in my explanations of manual CTE is simply ... a way of saying to carry out the instruction set accurately and repetitively.

Ah... my apologies... I misunderstood. That would probably be useful, if only to experimentally establish that for one person's visualization system (the operator's) there are (or are not) only a finite, small number of successful cut angles available for a given alignment. If several different people pushed the button, then it would probably be relatively safe to make a more general statement. (You need several since if you have shots near alignment boundaries, not all people may perceive them the same way - one person's "thick" might be another's "medium thick". There are other implementation problems, but we can ignore those.)

I don't see anything wrong with that, it's just that the result (whatever it might be) wouldn't tell me much that I care about. I'm interested in the underlying mechanism(s) that allow people who actually work at learning Stan's system get the results they report. I'm also willing to accept that those people state as well as they can what they do, and that they report as accurately as they can the results they achieve. That's all the data I have, and if I'm going to figure out why and how the system works at a low level then it has to be accounted for. I can't simply stick it in a black box labeled "feel" and ignore it.

 

[AtLarge]

... I'm interested in the underlying mechanism(s) that allow people who actually work at learning Stan's system get the results they report. I'm also willing to accept that those people state as well as they can what they do, and that they report as accurately as they can the results they achieve. That's all the data I have, and if I'm going to figure out why and how the system works at a low level then it has to be accounted for. I can't simply stick it in a black box labeled "feel" and ignore it.

I understand what you're saying, and I agree that what you seek is of interest.

 

[Mikjary]

I'm interested in the underlying mechanism(s) that allow people who actually work at learning Stan's system get the results they report. I'm also willing to accept that those people state as well as they can what they do, and that they report as accurately as they can the results they achieve. That's all the data I have, and if I'm going to figure out why and how the system works at a low level then it has to be accounted for. I can't simply stick it in a black box labeled "feel" and ignore it.

Tap, tap, tap!

Best,
Mike

 

[LAMas]

I don't think, while shooting pool, in terms of numerical angles, but I can visualize proportions. I am good at spacial comprehension and employ that gift in my job and pool. There are those that have that gift who don't have to have a college degree or the desire to get one.

There are pool shooters that can beat me that can't comprehend what we are "talkin bout" nor do they care - they just try to make some coin.

I hold that they have high levels of mind motor comprehension - naturals.

Those that can shoot, do and those that can't have to manage (teach).

 

[Patrick Johnson]

...I'm interested in the underlying mechanism(s) that allow people who actually work at learning Stan's system get the results they report. I'm also willing to accept that those people state as well as they can what they do, and that they report as accurately as they can the results they achieve. That's all the data I have, and if I'm going to figure out why and how the system works at a low level then it has to be accounted for. I can't simply stick it in a black box labeled "feel" and ignore it.

Don't forget to account for the fact that what system users say they're doing is impossible.

pj
chgo

 

[JoeyA]

Don't forget to account for the fact that what system users say they're doing is impossible.

pj
chgo

What specifically are you referring to?
JoeyA

 

[jwpretd]

Don't forget to account for the fact that what system users say they're doing is impossible.

So if somebody says "I stand about here and angle my body like this and imagine a line over the top of the cue ball to the right side of the object ball, and shift my eyes a little like this and imagine a line from the left edge of the cue ball to the left quarter of the object ball, and once I'm positioned so I can imagine seeing both lines simultaneously, then I move into shooting position by moving down and directly toward what I see to be the center of the cue ball while sliding my bridge hand in from the left with the tip pointed so it winds up offset 1/2 tip from the cue ball's axis just as I reach my normal shooting position, and I can imagine seeing those two lines remain steady in my vision while I'm doing that."

So if somebody says all that, it can't be true because it's impossible? Well, maybe so, I guess, but there's sure a lot of video out there where it looks like people are doing that. (Except maybe the eye bit - difficult to see on the videos available, though even that doesn't sound impossible.) I think for now I'll accept, just as a hypothesis, mind you, that all that's possible and is what they're doing and see if it gets me anywhere.