The mathematics behind Bogotas.

[Shifty1]

Yes. Thanks for the post.

[tacit_guardian]

Thanks very much Shifty1 for your input, and your data

I agree that theoretical tomfoolery may seem like just that, but to go back to our OP's point I would be interested to do a bit of analysis on his point.

We should pay attention to the empirical data that's coming from pickers, both professional and recreational. Not the feedback that says "this is great, or that is not" but the feedback that says "I open more locks with this bogo or that bogo, this manufacturer is better than that one (with the all important) because.. " Or even better, "I liked this store-bought one, but when I made my own I opened up the spacing/made the recesses bigger/made the slopes larger/ etc and I found it works better/worse in a particular variety of lock.

Sure it's all statistical when you're raking or rocking or doing anything of that nature. That's a GOOD thing. Because that means that we can use statistical and numerical analysis, as well as some finite element analysis on the lock parts to understand what's going on inside the lock, visualize it, and design tools that give us the edge.

I'm going to try and find some data on spring constants of common manufacturer's follower springs, masses of follower and key pins, and work up some data on how the pins are acting. From there you can start inputting both user impressions of different bogos and dimensions and start weighing them up.

I think this is an interesting and important question. I've seen a lot of "computer designed" items and I'm skeptical because technique and internal components are major compounding factors. I like this question regarding bogos because we're talking about an optimization problem, and that's something we can attack scientifically.

Let's keep it going. Take some measurements, post them up, discuss your impressions. And don't forget to take some time out to kick a locks ass.

Cheers
TG

[GWiens2001]

Only problem with statistical information is that it is averages. Average can be utter BS. Average can be you standing with one foot in a bucket of ice water, and the other foot in boiling hot water. On average, you are comfortable.

Gordon

[femurat]

That's a good one Gordon You reminded me of this short poem:

...da li conti che se fanno
seconno le statistiche d'adesso
risurta che te tocca un pollo all'anno:
e, se nun entra nelle spese tue,
t'entra ne la statistica lo stesso
perch'è c'è un antro che ne magna due.

I could translate it this way: following recent statistics, you eat one chicken every year. If you don't eat it, it's in the statistics anyway, because someone else eats two.

Anyway, just to be somehow on topic, here's a pdf that explains what this thread is about, related to the king and queen picks. A member of another forum called abroxis posted the link recently and said it's not copyrighted material since it was written for the government. I take his word for good.

Cheers

[tacit_guardian]

Only problem with statistical information is that it is averages. Average can be utter BS. Average can be you standing with one foot in a bucket of ice water, and the other foot in boiling hot water. On average, you are comfortable.

Where that might be true for an average, that's why we're not talking about averages (you also can't do statistical inference with only two data points, that's why the bucket analogy falls down, as well as conflating qualitative and quantitative quantities.) An average is an Arithmetic Mean, which is not the way to approach this problem. And statistics is much, much, more than just averages. I would probably try two methods, with a Monte Carlo analysis being one of them. That would put us "in the ballpark" at least, and aid the understanding.

Raking is a stochastic process, so no two attempts are alike. That's not as "scary" for simulation purposes as it seems, as reality is stochastic and we're pretty well versed in modelling these things (guidance, navigation and control systems (GNC) for aircraft and spacecraft are masters of dealing with this). So really, this can be done. And we can learn from the process. I think Steelbacks brought up a good question, and I haven't seen any rigorous treatment of it.

So we break the problem down into parts. The first part, that I'm currently working on is to start modelling the behavior of a lock. Masses of keypins and drivers, spring constants of the driving springs, and friction coefficients. Number of pins and pin spacing is easy to vary in the simulations. I'm still working on how to model binding and tension.

I'd be interested to hear about other's ideas on this, and please submit any information you may have if you can.

Cheers
TG

[tacit_guardian]

Ok team, I think I see where some of the confusion may have come about.

In reading a paper recently on the physics of key bumping (and pirating quite a few of the calculations for my own purposes) I found some parallels in what we're trying to do with bogos and what was being done with bumping and raking.

The confusion arises from this (I think!):

When we think about making a bogo, we're thinking about pin chamber spacing, and maximum ajacent and so forth. We're thinking of a static system. A snapshot in time. So we don't really intuitively think about it as a continuous process, because the bogo/rake is not static, we're moving it in and out and trying different angles.

I admit I got a little crossed-up when thinking about the two as well.

So for the sake of designing bogos we need to optimize the spacing and the angles, of course. But we need to understand that they're in a dynamic system, one that's moving. So if we can't predict how the rake is going to move (the user could use it any old way) then we need to consider that.

I've already started on a parametric model of a bogo based on a Sparrows triple-peak and double-peak that I have. Has anyone else played around with scaling/dimensions?

Cheers
TG

[GWiens2001]

Somebody has been studying Ian Malcolm's Chaos Theory.

Gordon

[Comrade627]

Somebody has been studying Ian Malcolm's Chaos Theory.

Gordon

He must have taken a vacation to Isla Nublar for the inspiration to do so.

Recognized references are the best.

[Syl]

Life... Uh...



Finds a way...

A ha. A ha har har. A har. A har.

Back on topic.
I'd love to contribute. Is there a way I can add to the data pool or contribute anything? I'm not too much of a math guy, but I have some experience writing software in C++ and Python, which may be of use.

[tacit_guardian]

I'd love to contribute. Is there a way I can add to the data pool or contribute anything? I'm not too much of a math guy, but I have some experience writing software in C++ and Python, which may be of use.

This could be very useful in the future if we want to do a "pick designer" program with the results of the computational experiment. You could input the kind of locks you want to tailor a pick design for.. so if you deal with a lot of 6 pin Yales (Y1 keyway) or a lot of Lockwoods or even a certain number of pins or certain kind of pin you could play around with the design. That would be really cool!

Do you have any experience in working in putting together an interface? I'm hopeless at that stuff. I do embedded ANSI(CCI) C for microcontrollers, and scientific computing using MATLAB and Mathematica and not much in between, unfortunately. So your help would be appreciated in the future for certain!

As for

Somebody has been studying Ian Malcolm's Chaos Theory.

there's a lot more to that then what we need. We could work very nicely with Stanislaw Ulam's Monte Carlo Analysis, with maybe some of Rudolph Kalman's regressive-predictive filter

"It's yet another in a long series of diversions in an attempt to avoid responsibility."

[Syl]

Do you have any experience in working in putting together an interface?

No, not really. I messed around with QT for a few days back when I still loved KDE and, other than a brief dive into Mozilla's code to take a look at Gecko, that's the limit of my experience.

Sorry, I should have included more info about my background.

I'm a Linux nerd and I have almost zero experience writing code for Windows. I highly value function at the expense of form. For me, a fancy interface is implemented in ncurses. I stick as much to command line programs as I can; no GUI can possibly match the dynamic power and elegance of a CLI. There are always exceptions, though. Some programs simply require a GUI, such as a browser. I'm not going to attempt to surf the internet using links or lynx. Or, at least not much.

I'd been thinking along the lines of something that could parse the data being gathered and spit out tables of formatted and/or processed data based on any algorithms you math guys come up with. But, it looks like you've got that type of thing covered with your MATLAB experience. I don't know. Maybe there's still a way I can help?

[tacit_guardian]

I'm certain there is, don't worry

I'm a proto-linux guy.. I use it on my work computers as all the tools and modules I need are free and fast, and I get a lot better performance from the hardware. I only have a windows 7 machine around for a few programs I use that don't have a linux variant.

The great thing abot java and python (and properly written C!) is that the code is portable across systems. So don't worry about that. I only mention the GUI as I'm not sure that we would get the breadth of the community with any tools we produce if it were a terminal session interface. There are some people out there who have never even seen that before.. scary, I know.

All of the help you can provide is useful.

Maybe go through and tabulate all of the data from the sites mentioned previously in regards to the dimensions of the inside of the locks and pins and such? I'm sure you could write some Python to format that all into something that's not an ugly awful website? If you could produce that in something like CSV, that would make it MUCH easier for me to import into Mathematica and use directly.

And with that comes the question, where do we store all of this information. I've used GitHub in the past. Does anyone have a preference? Because if we store all the software in the same place, as well as the data files we build (lock mechanical dimension database!) then we don't have versioning problems in the future.

What do you think?
TG

[tacit_guardian]

Apologies to the OP, and the forum. Per moderator I will not be requesting or receiving assistance in this regard, and will not be pursuing it any further. It was made very clear to me that my attitude was not appreciated.

Keep innovating, everyone.

[Squelchtone]

Apologies to the OP, and the forum. Per moderator I will not be requesting or receiving assistance in this regard, and will not be pursuing it any further. It was made very clear to me that my attitude was not appreciated.

Keep innovating, everyone.

Just to clarify, it's nothing against you, so please don't take it personally, or as a personal attack. You seem like smart well written guy who is good at science, design, and engineering, but as you said it yourself in another post "...theoretical tomfoolery..." and to me trying to measure all the locks in the world and weigh all the pins in the world in order to design some ultimate super pick is an exercise in futility. Picking is a simple act, there's no reason to over-think it or to introduce unnecessary variables.

It's like measuring all the golf courses and golf balls on earth in order to make a 1 golf club that will virtually guarantee a hole in one every time you use it no matter who is using it or which hole they are playing. There's a reason people carry around a bag of golf clubs for different occasions, just like pickers have pick sets that are good for different locks. And yes, you can engineer a better balanced golf club that is more comfortable and makes the ball go further, so I do see a place for R&D and engineering and innovation, but to get as granular as massing pins and getting the pin stacks resonant frequencies in order to decode a lock is a dash overboard in my opinion. I just don't want members here to get sucked into some whimsical project that is applying science to a hobby the writer is not very familiar with. If you were already picking for 5 to 10 years or worked for one of the lock companies then sure a project like this would be apropos.

Squelchtone