Pangolin User Meeting Results (and more Free Laser Shows)

[loopee]

Yep, I used the “Projection Zones – Beam Attenuation Map” feature of Pangolin to do that

It’s a real novel feature and very simple to use

Basically set the camera up in place with the lens cap on then go into projection zone and turn on display grid and then draw on the map where the camera is.

You can apply varying levels of attenuation, in this case I used 20%, I could have gone higher so that the beam doesn’t quite disappear but I am unsure as to what level would start to damage the camera so I kept it low.

The other great feature is that you can apply different attenuation maps for different show effects, i.e. Targeted beams, graphics, atmospheric, etc.

[buffo]

OK, I didn't have much time, but nevertheless I performed a short experiment last night. First I created a frame with just two white points connected by a line. It made a horizontal line that was probably 3/4 of the width of the scan field.

When I displayed this file on the projector, Pangolin split the line into two segments with a break in the middle. I believe this is part of the special scanner protection circuitry on the LD-2000 that we've talked about before. It's designed to protect the scanners against bearing failure when a frame consisting of only a few points is displayed. (As to the mechanism of that failure, I have no idea... Need to ask Bill about that sometime.)

Anyway, I then loaded the beam attentuation map and blacked out the middle of the scan field. (I wasn't looking for accuracy; I just wanted to get a black box somewhere near the middle. I think I was pretty close.) This deleted one of the line segments. That's about all the time I had to play, but it does appear that the software is capable of splitting a line even if it's made of only 2 points.

Adam

[carmangary]

Thanks for performing that test. I'm not sure how to interpret the result since you said the protection circuitry split the line before you even applied the mask. Anyway, here is a picture to explain what I am asking about if that helps.

The top and middle line segments are the ones in question. To do it correctly, the top line would need to be shorted. The middle line would need to be chopped off in the middle to create two new segments on either side. A safe, but not so good way to do it would be completely delete the top and middle lines since they intersect the square. A not safe and stupid way to do it would be to delete lines that have points in the square. So, only the top line would be removed while the middle line would still go through the square (BAD NEWS). The bottom case is what you will likely see in frames, though, and the stupid algorithm would do a pretty good job on this one. The top 3 points would be removed leaving a short line on the bottom.

All three of the above are easy to do but checking to see if they cross boundaries during real time display takes time that may not be available. It gets tricky when instead of a box there is a fade. In that case, the laser would have to be appropriately dimmed. But, going back to the discussion about "smart scanners", current scanners provide not feedback to the software so with only a few points, the software has no clue of knowing where the beam is during a midpoint. So correct dimming is impossible unless lots of points are added to the long lines. A smart scanner would have no problem with any of this, though. Give it a frame and a mask and it would know exactly how to do it without guesswork.

I am going to implement a simplified version in Spaghetti. I'll allow one or more blanking squares instead of a bitmap. And first pass I'll only detect if there is an intersection. So, the top two lines would not appear at all. Phase two I would make it smarter to shorten the segments by clipping.

Anyway, sorry to derail the discussion. It's a very interesting subject to me, though.

[Pangolin]

What hardware/software is used for that Standing in Motion show. I noticed that when the laser points near the camera that it is blanked to prevent blinding the person standing there. That is a nice feature that I have been planning on adding to my software for some time. I assume it is some Pangolin variation?

mumble mumble mumble Pending US and International Patents mumble mumble mumble...

[carmangary]

I guess that is one way to prevent competition. Are you sure your last name isn't Gates?

[drlava]

carmangary, consider that a good way to do a safety zone instead of just blanking is to put a limit on the number of points total added to the frame or per line for the boundary gradation zone. At high scan speeds, having a rough gradient by only adding a couple points at the gradient boundary, if necessary, is not so bad. In the example given, as the beam fan slowly scans downward, additional points would not even have to be added to each frame to acheive the desired effect of dimming beam fan as it enters the zone. In your example, adding one or two fade points to the line (always making sure their brightness is determined by the dimmest section that the segment is in) would not unduly slow down the total scan rate, and even though it's only a couple of points it should not look bad.

[Pangolin]

I guess that is one way to prevent competition. Are you sure your last name isn't Gates?

We look at it this way. For literally years and years and years, Pangolin has done the "heavy lifting" in this industry. We think of new concepts, only to have them show up on average around two years later in competitors' software. One partial sampling can be seen on this page:
http://www.pangolin.com/LD2000/software_innovation.htm

Just so you know, we have patents pending on the Beam Attenuation Maps concept, and other extrapolated safety concepts as well. Pangolin has a tremendous reputation in this industry for safety. Really we were among the first to take safety seriously, and then implement features into our system to help ensure safety.

One illustration that I will make is this: Before your post in which you admired the ability for our system to prevent damage to the camera, I never saw a single post from you that indicated that you had this idea. In other words, this is an example where we thought of something first, and then, after seeing it you thought -- oh, this is a great idea, I think I will add it to my system. Yes, there are many great ideas in our software! In the past, we interpreted imitataion as a form of flattery, but more recently, we are just not flattered any more...

It's hard to describe why, but this is really a problem for Pangolin users as much as it is for Pangolin. Basically, we got sick of being copied, so around four or so years, we have been protecting our innovations with patents. We REALLY don't like to do this, since it is expensive, and time-consuming, and it delays the release of certain things that we don't want to show until they are far enough along in the patent process. But I feel we have to do this.

The patent system was established to protect people, and also to put the idea into the public domain so that it could be improved upon by others. Anything that we are doing is legal and ethical. And if a system is in place to help protect ourselves and protect our users, why not use it?

Best regards,

William Benner

[carmangary]

For the record, I thought of the idea of adding blanking spots to my output long before I saw that video. The reason I even recognized what was going on was because I had been planning on doing that very same thing so that I could sit in the middle of beam shows in my garage. It was explained that you use bitmaps for the masks. That is a very clever idea but it is not what I plan on doing.

[drlava]

It'a balance I'm sure. On one hand I doubt you would feel comfortable ethically patenting things that are considered 'required' for safety of a professional show but at the same time you must decide what features are considered advanced enough to patent BUT also that they are not required for a show to be considered safe.

It's hard to describe why, but this is really a problem for Pangolin users as much as it is for Pangolin. Basically, we got sick of being copied, so around four or so years, we have been protecting our innovations with patents. We REALLY don't like to do this, since it is expensive, and time-consuming, and it delays the release of certain things that we don't want to show until they are far enough along in the patent process. But I feel we have to do this.

Reading not-so-in-between-the-lines here I guess we can gather this is one of the reasons for delaying Beyond.. software IP and display algorithm protection.

James has told me that LaserBoy has had a method of implementing 'safe zones' via a bitmap colorization method that is usually used for other things, not sure how long it has had this capability.

[Pangolin]

On one hand I doubt you would feel comfortable ethically patenting things that are considered 'required' for safety of a professional show but at the same time you must decide what features are considered advanced enough to patent.

Well we can't just patent anything... In order to receive a patent, a system, method, apparatus, etc. must be useful, novel, and non-obvious.

But to us all of this is real simple math. I will illustrate this with a prime example, and one which has caused some heated debates in the past. AVI patented the Omniscan, which, in it's simplest form can be described as "a fish eye lens for laser projectors". This allowed AVI to cover an entire planetarium dome with laser light.

* Useful? Most certainly.

* Novel? Well, it had never been done like that before. Up until then, people had been either using multiple projectors to cover the entire dome, or simply settling for less than the entire dome being covered.

* Non-obvious? Well, lots of people say "ooooohhhh, well, simply putting a fish eye lens on a laser projector is obvious. That's just obvious...

Well the courts rule (and continue to rule) that, for an industry that has existed for 30 years and nobody did it before, and yet it is very useful, well, if it's so damn obvious why wasn't it done before?

This can be said of most of what we do in fact. We come up with new ways. 6-channel color, beam brush, raster frames, beam attenuation maps, etc. etc. and so on. These are things that we developed first, in an industry that has existed for 30 years or more. If it's so damn obvious, why hasn't anyone done it before?

A lot of what we do is only "obvious" after we do it . Just like putting a fish-eye lens on a laser projector. Seems obvious, but if it's so damn obvious, why hadn't anyone done it before AVI?

James has told me that LaserBoy has had a method of implementing 'safe zones' via a bitmap colorization method that is usually used for other things, not sure how long it has had this capability.

Hehe. More simple math. We released Beam Attenuation Maps in March of 2001. James says he started working on LaserBoy sometime in 2004. (Incidentally, ILDA Format 4/5 existed in 2002...) So.... Simple math.


Bill

(By the way, Omniscan is much more than just a fish-eye lens, but many peope use that simple characterization, when they are seemingly outraged at the fact that AVI got a patent on this...)