Announcing ILDAC-32 - an ESP-32 based DAC for ILDA projectors

Quote:

---

Originally Posted by HankLloydRight

Thank you for the feedback.

Since the ILDA spec only has one 8-bit value for each RGB color, is there another reason to go with 12-bit DACs for color?

---

You lose resolution through color balancing often.. but it also gives a finer curvature control of color balance.. to top it off.. gradients which actually adress the higher bitrate in software can be a lot nicer when you have more resolution.. ILDA content is great but don't only focus on the ilda content imports.. it only provides you so much :)

I also second mixedgas his idea of using the shutter channel as an extra color channel or for something like timecode.. shutter is a legacy thing for galvo blanking

Quote:

---

Originally Posted by masterpj

You lose resolution through color balancing often.. but it also gives a finer curvature control of color balance.. to top it off.. gradients which actually address the higher bitrate in software can be a lot nicer when you have more resolution.. ILDA content is great but don't only focus on the ilda content imports.. it only provides you so much :)

---

Ok, thanks for the reply. I'll keep the three 12-bit DACs for XYRGB for future expansion possibilities.

Quote:

---

Originally Posted by masterpj

I also second mixedgas his idea of using the shutter channel as an extra color channel or for something like timecode.. shutter is a legacy thing for galvo blanking

---

I'm happy to go down this route, but I need some help. Where would the additional color channel data come from? (Etherdream?) Which ILDA color output pin? I see three additional colors on the ILDA cable spec (violet, yellow, and bright blue). For a timecode, where would that signal come from and where does it go -- an edge connector for external use? What transformation does it need? Thanks.

yes.. etherdream has some user addressable pins. they're designed as alternate color pins, but could be used for other things.

Quote:

---

Originally Posted by HankLloydRight

Ok, thanks for the reply. I'll keep the three 12-bit DACs for XYRGB for future expansion possibilities.



I'm happy to go down this route, but I need some help. Where would the additional color channel data come from? (Etherdream?) Which ILDA color output pin? I see three additional colors on the ILDA cable spec (violet, yellow, and bright blue). For a timecode, where would that signal come from and where does it go -- an edge connector for external use? What transformation does it need? Thanks.

---

Quote:

---

Originally Posted by swamidog

yes.. etherdream has some user addressable pins. they're designed as alternate color pins, but could be used for other things.

---

Thanks. So my first step is to try and get the Etherdream protocol running over Wifi. :)

S = spare, timecode, focus, or depth/Z axis. There are some neat
Tricks with two scan heads and a "Z" Out signal for producing stereo images by adding a few opamps to make depth cues. Z needs to be similar in resolution to the xy dacs.

After making color correction curves for my latest projector I see about 30% laser energy un-used with 8 bit dacs. 10 or 12 bits can really improve that.

The Scaley Anteater company used to include as standard some really neat signals on their boards, now that I have the ability to use those signals, they are not on the standard hardware twenty years later. TTL signals for start of frame and point clock, and odd/even point number let you do some neat things. Having at least one eight bit auxiliary dac would be cool, as would a serial port or 11 ttl lines for beamtable control. [8 position 3 for color]

Don't even get me started on what having a few analog inputs could do.

There were/are some really neat things you can do if you can pass an 8 bit word of outputs on a per frame basis. That alone makes me lament that x29 no longer exists.

Steve

Quote:

---

Originally Posted by mixedgas

S = spare, timecode, focus, or depth/Z axis. There are some neat
Tricks with two scan heads and a "Z" Out signal for producing stereo images by adding a few opamps to make depth cues. Z needs to be similar in resolution to the xy dacs.

After making color correction curves for my latest projector I see about 30% laser energy un-used with 8 bit dacs. 10 or 12 bits can really improve that.

The Scaley Anteater company used to include as standard some really neat signals on their boards, now that I have the ability to use those signals, they are not on the standard hardware twenty years later. TTL signals for start of frame and point clock, and odd/even point number let you do some neat things. Having at least one eight bit auxiliary dac would be cool, as would a serial port or 11 ttl lines for beamtable control. [8 position 3 for color]

Don't even get me started on what having a few analog inputs could do.

There were/are some really neat things you can do if you can pass an 8 bit word of outputs on a per frame basis. That alone makes me lament that x29 no longer exists.

Steve

---

Hi Steve, thanks for the ideas and feedback. But I'm not sure where to start.

Until (and if) I get Etherdream support working, the only source for content is the internal test patterns and ILDA files. Etherdream support is a huge unknown at this point because we don't know just how fast we can stream and display the frames on ILDAC over Wifi. So that might never happen, or if it happens, could be really slow.

So I'm not sure where these other signals (spare, timecode, focus, or Z/depth) would originate (or go). I see that there is a Z coordinate in the ILDA file spec, but is anyone using that in ILDA files, or does anyone need that functionality -- for this particular device?

I'm happy to go forward with 12-bit DACs for color - which was in my very original design, I just have been trying to simplify the design and part count for future versions by using the internal 8-bit DAC for one color, and one external dual 8-bit for the other two.

Quote:

---

TTL signals for start of frame and point clock, and odd/even point number let you do some neat things.

---

I have a couple of GPIO left over where I could send these TTL signals, but that would require exposing them to the outside of the enclosure with a connector of some sort, which isn't hard to do, but how many people would actually need or use them (given the limited sources of content)? But also sending multiple signals out to the GPIO would slow down my point clock a few microseconds and slow down the scan rate in non-point optimized mode. So I'm trying to gauge the overall utility of new features like this with the increased parts and assembly cost and some performance downsides.

Quote:

---

Having at least one eight bit auxiliary dac would be cool, as would a serial port or 11 ttl lines for beamtable control. [8 position 3 for color]

---

Again, I'm not sure where or how an auxiliary 8-bit DAC would be used (given the content sources). Same with a serial port or additional TTL lines. ILDAC(ESP-32) isn't a general purpose computer that can do infinite things. It's a very limited microcontroller I've programmed to do a few specific things. Remember, there is no computer connected to ILDAC sending content like all other DACs out there are. It's 100% internal. So I feel trying to continue expanding it for a cool factor would detract from the original purpose and scope of the device.

If someone wants to take the core hardware that I've built, the ESP32 is an open platform and anyone can really do anything they want with it. I can document the specs on how to access the DAC chips, BLE, Wifi, OLED and the SD card, and people can run with it with their own code. But I don't think at this time I can invest the time, cost, and energy to add new features like these that only a few people might ever use.

If there's something I'm missing here that I can easily do, please let me know. You're all the experts, I just built something cool that I needed and thought other people might need or like to have in their toolbox.

That said, if these were flying off the shelves and lots and lots of people were using them for different things and there was real demand to expand the base functionality as you've described, I'd be open to trying to add some or all of these (and I'd then have the time and resources to spend that kind of development). But so far, I've only sold seven ILDAC-32s, and most of those were at SELEM 14 in August. (And I only have four left to sell). I've been working on this since June, 2021 and have invested many hundreds of hours in development (a lot of that was learning as I go).

And just to compare, the ILDA GEM Pocket tester is US$100 and only does a Quadature test pattern, static lines, a static beam, and scan speeds up to 30kpps+/-10%! I think for $150, the ILDAC-32 does so much more - and it can hit any selected scan rate up to 70kpps with <0.04% error. And access via mobile app over Bluetooth or any networked computer over Wifi.

Thanks again for your feedback.

Well, I just found I can get point rate from the settle command on the Mach Dsp scanner amplifier.

That settles that.

You still need an S or Z output for Beambrush.

Steve

Quote:

---

Originally Posted by mixedgas

Well, I just found I can get point rate from the settle command on the Mach Dsp scanner amplifier.

That settles that.

You still need an S or Z output for Beambrush.

---

Ok,cool. I didn't know I was being put up against the Mach DSP amp.

And where would ILDAC-32 get Beambrush content?

I kind of think there may be some confusion about what this device is or, is intended to be. Either on my part or others. This is really just a testing device not unlike an ILDA Gem but, with some additional tricks.

When you say DAC, I think there is a thought that this is going to be used for playing lasers shows not unlike an FB3 or Etherdream or FB4 or Moncha DAC or something of that nature. Unless I'm confused about what the planned future of this is.

Quote:

---

Originally Posted by Bradfo69

I kind of think there may be some confusion about what this device is or, is intended to be. Either on my part or others. This is really just a testing device not unlike an ILDA Gem but, with some additional tricks.

When you say DAC, I think there is a thought that this is going to be used for playing lasers shows not unlike an FB3 or Etherdream or FB4 or Moncha DAC or something of that nature. Unless I'm confused about what the planned future of this is.

---

Hi Brad, yes, I can see where there might be a little confusion about the term "DAC" and existing products, but I attempt to make that clear in the second line in my first post above, as well as the big headline on the website at https://ildac32.com :

"Announcing ILDAC-32 -- a self-contained DAC based on ESP-32 that runs test patterns, lumia cues, static beams, and ILDA files/animations to any ILDA compatible laser projector."

I'm not sure how better I can describe it right up-front.


In terms of future plans, as I'm mentioned above, we're attempting to get an Etherdream emulator over Wifi working, but that's a big unknown at this point and not in the list if current features.