The LaserBoy Thread

[dchammonds]

If you export a frame_set as wav, then open it back into the app, every sample of the fully optimized wav is loaded into the app frame_set space. So you can examine it from that perspective and if you save it as-is to an ild file you get all that.

That's a runaround.


A number of laser dac drivers (including EasyAudDAC) have a repeat frame option where the dac repeats the last frame received continuously
until a new frame or close command is sent, so it's a good idea to support generation of an optimized ILD frame that 'loops back' on itself.

What software / hardware are you using to display directly from an ilda file?

Does it matter? The ability to clock out an ild file as-is at a given scan rate is foundational to any legitimate laser playback application.

[james]

That's what the recent conversation in this thread is about.

It seems like a great idea to include the path from the last point in a frame back to the first. But that is only valid if and only if the frame is repeated. Once a frame has been repeated enough times for whatever reason and it's time to move on to another frame, then it's useless and in the way. How do you separate that bit from the frame itself so that you can intelligently use it or not? There is no provision for this in the ilda file format.

Of course it matters! There is still a disconnect between what people expect an ilda file to be and any actual use case information about what that is or how to get it.

What you describe needs to be done in the software that drives the dac.

That's the issue of this conversation.

And by the way, in, out, in, out is not all that different from just in, out. It's just a bit more fun.

Just ask Alex Delarge (A Clockwork Orange).

https://getyarn.io/yarn-clip/877da62...1-7dd0b192d0f1

.
.

[dchammonds]

That's what the recent conversation in this thread is about.

What you describe needs to be done in the software that drives the dac.

That's the issue of this conversation.

.

Then why bother to offer individual ild frame optimization within LaserBoy in the first place?

I've seen numerous requests for a good ild rendering of a logo on this forum over the years- you recently answered one.

[james]

I've answered more than one.

Like I said there is a huge gap in the idea of what an ilda file should be.

The most likely answer is that the format was just for the purpose of getting vector art out of one proprietary app and into another; each having its own way of utilizing that generic information. And that capability was entirely voluntary by the app developer. I don't think it was ever intended to be a feature rich format for direct unprocessed display.

I think you might have some idea about the lack of cooperative attitude with regard to this particular file format.


.

[swamidog]

one cheat i use to remove blanking lines on AM circles is to tie the amplitude to the laser intensity and then just bias the intensity a bit so it draws the circle at all sizes, but is off for the flyback.

Okay, so I've got a laser synth pumping out cycloids from harmonic audio waveforms, with raw frequencies, instead of ILD frames, nor an end of frame reference (unless I can manipulate the waveform's index from the audio library) in order to utilize with flyback blanking between frames of an animated sequence. My current workaround is to have ADSR envelopes to collapse a generator's signal, before attacking the next waveform. But, that doesn't affect the intensity (yet).
Anyway, I'm faced with a similar problem of blanking the beam in between images of a sequence. So, why can't an entire blank frame of n number of defined points be inserted in between every differing image of the ild sequence, during the wav to ild conversion process? Doesn't matter what everyone else position is in this debate. It would simply work, right?
Thanks for your feedback and thinking outside the box of the status quo. It's a hermit kind of thing.
😎

[james]

@TheHermit

It sounds like what you are describing would be to start a display sequence at the origin (the center of space, aka zero volts) and then return to the origin at the end.

That is certainly a good one-size-fits-all solution and would even work with an individual frame repeated many times.

When my app exports a wav file, it manages elements that are not part of a typical ilda frame_set (or at least there is no documentation describing them).

These elements are blanked lines with velocity points that make a smooth transition from the origin to the first vertex of the first frame, a bridge from the last point back to the first for frame repeats, a path from the last point of one frame to the first point of the next and a final path from the last vertex of the whole set back to the origin.

All of this gets written into an optimized wav.

The extensions I added to the wav file format include markers in the least significant bits of the red and green color channels. The red bit marker designates the end of a frame and the green bit marker designates that a frame was repeated in order to achieve a particular frames per second when displayed.

When a formatted optimized wav is imported back into the app, the app can frame it correctly and marks all of the repeated frames. So it is possible to strip all of that out including the repeated frames to get back to the original frame_set that was used to make the wave.

But all of that information is there when a wav is imported and not stripped out and in that form it could be saved as an ilda file (with all the bells and whistles).

[lasermaster1977]

one cheat i use to remove blanking lines on AM circles is to tie the amplitude to the laser intensity and then just bias the intensity a bit so it draws the circle at all sizes, but is off for the flyback.

swamidog's cheat is a good on for when using a ramp wave for AM. When using a sine or triangle wave it would not be as necessary to "hide" the flyback. (besides, even when the flyback of the ramp AM isn't blanked it looks like a straight line diving into or out of a point source.

@TheHermit, for your Teensy setup it should be easy to generate a sync pulse for every AM waveform cycle that has a programmable duration that could be used for flyback blanking cycles.

[swamidog]

swamidog's cheat is a good on for when using a ramp wave for AM. When using a sine or triangle wave it would not be as necessary to "hide" the flyback. (besides, even when the flyback of the ramp AM isn't blanked it looks like a straight line diving into or out of a point source.

@TheHermit, for your Teensy setup it should be easy to generate a sync pulse for every AM waveform cycle that has a programmable duration that could be used for flyback blanking cycles.

yup. good clarification.

[lasermaster1977]

yup. good clarification.

(if I could only spell correctly)

[swamidog]

Thanks for the suggestion, 'dog. ;-)
Yes, I've considered that option either for each QOSC's mixer gain level, it's ADSR, the AM oscillators, or master size level (for an 'auto 3D' effect). Leaning towards the ADSR and/or toggling the auto 3D.
Hey, while I have your attention, I noticed that you're running 60kpps galvos on your DIY projector. Scannermax only sells to Kvant and word has it that CT no longer sells to entertainment customers. Mind if I ask where to find a pair like yours?
Thanks for the chat.
😎

i bought my ct's used about 8 years ago. if i had a source, i'd be happy to point you to them.