Topic: Exactly why this forum exists

[Fanny Pack]

I think Gary needs to spend a bit more time in LaserBoy land!

I am looking forward to it once you add a Windows interface to it.

[Absolom]

I am looking forward to it once you add a Windows interface to it.

Though I love LaserBoy, I say, Hear hear!!

[Fanny Pack]

The biggest benefit to using wave is that it has already been invented, years ago and there have been millions of man-hours spent designing operating systems, languages, drivers, software applications and hardware specifications that use WAVE!

We live in a world where we can get an 8 channel DAC for $22.50 !!!

WAKE UP PEOPLE!

Wave is an absolute measured information file format. It needs no interpretation. The information stored in the individual tracks needs only to be clock out a DAC!

Oh and BTW, LaserBoy waves CAN contain Z and they can also be stripped down to the bare minimum of un-optomized points and saved that way; NOT ready for scanners, but small.

There are reasons to be interested in developing other file specifications that might be packaged with wave or even ild files, and certainly MP3 or other audio file formats, that go farther to describe the relationships of the collection of files and other abstract concepts that need to be interpreted, like a kind of dance choreography notation that describes the use and functionality of its parts over time.

James. 

Using WAV files, what would you suggest as a method for display laser shows with real time control over positioning (location, skew, rotation, etc) as well as being able to tweak things like blanking and color shift?  The requirement is that these have to be able to be tweaked while playing back the show.  Also, the requirement is that the DAC is not a soundcard but a typical USB DAC such as an EasyLase or Norms $140 USB DAC.  Convince me that a WAV file is the best solution for transporting data into the laser show application and why it is better than a clean structured format that doesn't do things like bitmask different values into single entities.

[meandean]

  As I mentioned earlier, the wave format allows for the inclusion of non-audio information in the header- this is where we specify things like "color shift" (we call them offsets) down to the sample period for each of the 6 or 8 channels, allowing for proper viewing on the VGA or the $22.50 modified DAC via a projector, as well as the ability to re-shuffle the the values for rendered wave on the fly in LaserBoy or LaserWave.

[Fanny Pack]

I understand that but it doesn't solve any of the problems that I would like to be solved. 

[meandean]

  As for a Windows interface in LaserBoy, don't expect anything soon- James is quite the hardcore 'computer purist'; LB was originally written just for Linux in C++. Thankfully, he found the Simple Device Layer (SDL) to allow his source code to be optimally compiled for different platforms including Windows, without the performance limiting 'virtual machine' interpretation overhead of JAVA. At some point, I imagine he'll add code to use a mouse (SDL does support this, I think). Though it has the capability, I wouldn't recommend LaserBoy as a means for generating art from scratch- there are many free utilities that can do this in a more friendly fashion. To me, the value of LaserBoy is in its frame-by-frame viewing, organizing, and editing capabilities; its ability to stand as a 'universal translator' between formats such as ILDA, DXF, and CPP, and let's not forget the all-important task of rendering to a practical wave for scanners that can be sent out to a cheap modified multi-channel audio card via 'formatted' wave complete with subcode that stands as a bridge between our applications.

[James]

When he says "our applications", he means LaserBoy and his Laser Wave app.

James. 

[James]

I think it is important to note that reading a white paper description of a file format and actually making it work in real code is an incredible learning experience. There is NO WAY anyone can possibly understand all of the ramifications of laying out data a certain way until they have worked with it!

There is an amazing amount of information that is not really there in the numbers themselves but rather it is implied by the relationships of the numbers to each other.

Yes, LaserBoy waves can most definitely encapsulate the concept of an "object"; even if it is stored as a consecutive stream of 16 bit samples. It is a matter of examining the data with a set of heuristic tests. LaserBoy's focus and accuracy for discovering this implied information is getting better all the time!

Computer science research is a life long journey. No one can possibly know where a particular path might take them until they get there!

That is THE NUMBER ONE REASON why format 3 pisses me off so much.

No one can possibly know any of the beneficial side effects that it might provide until it is made real in code and fully explored!

AND IT HAS BEEN SINCE 2004 !!!!

ILDA took a perfectly sensible way of extending their file format and RUINED it for everyone else who might have otherwise followed it to its logical implementation and discovered its real values.

If you're thinking to yourself, there are probably only about 5 people in the whole world who know or care anything about this issue... consider the fact that I am one of them!

James. 

[meandean]

  I can't say that laser-formatted wave is the solution to everything... I don't have any laser equipment, nor do I know ANYTHING about ILDA; I wrote an abstract function generator program that commits directly to wave, so I don't have the insight to answer many questions (I live in a small bubble, you know). I'd like to think that the unused low bits in the wave color channels would be enough to encompass everything you want, but if you have a proposal for a new format, put it out (that's what this forum is for!!!). In the meantime, consider the fact that James and I already have the satisfaction of something that TOTALLY bypasses Pangolin and ILDA.

[James]

LaserBoy wave files can contain EVERY bit of information that might be found in an ILDA file; except for the concept of a palette of colors separated from the drawing by palette index or point order.

However these things can be mathematically deduced and restored by examining the data and distilling the RGB information into palettes.

The LaserBoy wave format has the added advantage of being something that can be quantized to a time constant and clocked out of a DAC as a working set of laser control signals.

Plus, LaserBoy goes beyond the ILDA file format by having a global wave polarity, signal time offsets, individual channel signal names and signal polarities, sub-code information to designate end-of-frame and unique-frames, adaptable signal bit resolutions with proper identification per signal per channel in the header and information about how the wave was optimized!

Not to mention the fact that the whole thing has been designed to be infinitely extensible into the future! It contains a version string!

It might not have anything to do with solving the specific issues that you (Gary) might be interested in for Spaghetti, but it is a fantastic solution for many other important issues of making the transition from descriptive laser art information to real working streamable wave data... and back again!

I completely understand what you are talking about and I agree that there needs to be a totally new open standard file format for laser art that is capable of containing much more abstract concepts; like VRML or Flash.

If you are interested, I'm sure it would be trivial for you to write a LaserBoy wave reader for Spaghetti.

James. 

[Fanny Pack]

It would be relatively easy to come up with a file format that does everything we want.  I'd probably start with XML to begin with because it is object oriented and easily parsed directly into objects.

But, that's just the easy part.  The hard part is developing an engine that can take those objects and manipulate them in intelligent ways and then finally render and optimized 2D vector sequence that can be handled by the scanner.  One of these days I might like to take a stab at that but I am hoping that someone else like Zoof or DrLava will do it first and make it public (either free or for sale).

The thing that I don't like about your WAV file solution is that it seems like you are playing tetris with data by turning it this way and that way to make it all fit within the combines of a file type that was never designed to hold that data.  I know that you can make a 100 channel WAV file and put all kinds of stuff into the extra channels but to me that's like putting all of your stuff into a bunch of shoe boxes and hauling it around with you. 

[BlinkenLights]

DOCs use wav... james just worked backward from there..

I agree that the wav has limitations that its hard to get james to see.
but as a final format it is completely valid

you should write the wave header into spaghetti, that would be cool..


i think that XML is the right way to go.. i was thinking that compressed XML would be awesome

[Fanny Pack]

XML is a nice starting point because it is easy to look at the contents of the file and see what it is representing.  And, if you need to change something, you just plug in some new attributes or elements where they logically belong and let the parser deal with it.  No need to worry about reading a single byte here while ANDing and ORing this and that.  At the end of the day, I am not opposed to tightening it up to a binary file if performance dicates that it needs to be done.  It's the content that I am concerned with more than the format.  So, if everything fit nicely into a WAV file I wouldn't be all that opposed to it, but as it is, it seems like there has to be a lot of massaging and masking in order to keep it a valid WAV file so that it can be sent to and understood by a media player.  But, since I don't use a media player to send laser data I don't want to be tied to a format that is intended for that.  It seems to me that the file format should be the least common denominator instead of the least common multiple.  Meaning, I don't think it should be based on some hardware that not everyone uses.  Instead, it should be based only on what everyone can use.  I guess I have said enough about that. 

[BlinkenLights]

great ideas all around but i think you need to realize that the sound card DAC is the driving force behind cheap laser control...

And when sound card dac users look for control software, where do you think they are gonna look?

[Fanny Pack]

I'd be willing to bet that there are more "cheapies" buying this

http://tinyurl.com/d2zot6

than going to the trouble of building and getting sound cards to work.