Blade Runner 2049 Voight Kampff Eye Scanner Tutorial / Buildup

[JOATRASH FX]

Another thing I forgot to mention is that you might notice in the video above that I have printed the entire shell, including the separate plates on the back. I made little 'guides' on the front plate so that it kind of 'pops' into place. It will still need glue though. I looked into making it possible to open the unit again at a later date, but there simply isn't room for functional screw fastenings. There IS a type of resin that could be used for 'living hinges' and a more functional 'snap together' assembly, BUT I have not tried it... and I haven't been able to get any solid data on how it holds up over time. (And there's a startup cost of maybe $550 just to be able to start experimenting with it!)

 

[JOATRASH FX]

I just got a 1st animation test from the programmer. This is basically just gathering up a bunch of bitmap graphics and testing out some things in the code.

We'll be aiming for a more 'directed' sequence of events where everything doesn't just pop onscreen at once.

But I really like how Blade-Runner-ish it looks even at this early stage.

 

[joberg]

Speechless

 

[Kylo1]

WOW!

 

[JOATRASH FX]

So, I got some metal printed test parts in of the central arm, done on a Markforged Metal X. Unfortunately, they are not suitable for this. They're too thin, and they expand and contract in different areas, not to mention the surface is pretty rough as you can see. It would cost too much and take too much time to make usable parts for this particular project with this technique. I tried bending the left piece into shape and it snapped, sending parts flying!

That leaves either printing the entire thing in plastic or (if budget allows) laser-cutting the 'flat' part of the arm and pinning it to the 'block' that holds the spring.

 

[joberg]

Oh well, back to the drawing board as they say. Besides that, you're doing a great job on the tuto, the vids and the build

 

[JOATRASH FX]

Oh well, back to the drawing board as they say. Besides that, you're doing a great job on the tuto, the vids and the build

Thanks.

It's not going to be an issue, really. Based on what I had already learned, there was a large probability that printing the entire arm wouldn't work.

And I'm almost certain that laser cutting it will work out great based on the tests I was able to have done. (Laser cutting aluminum will als make it look correct right out of the box.) So as long as I can make it work budget-wise, that's the way it'll go.

Had a good long talk with the programmer yesterday about how to do some of the animations.

 

[JOATRASH FX]

The latest from the programmer. This is looking AWESOME if I do say so. I made a short animatic to give the programmer an idea of how to sequence the animations and this is what he did over the weekend.

Of course, LOTS of tweaking still to come. We're going to try to add some 'static glitches' in a few places, a 'scanning sequence' over the eye photo and some other things. One big part is how to set up the box-animations on the right. Turns out it's not easy discussing how to program abstract patterns! (Looking at the graphics done for the film, there are a lot of nonsensical-but-cool animation patterns on various screens and we'll try to capture the feeling of those.

 

[MangyDog]

This really is looking utterly amazing.

Ive been wanting to make one of these, but truth be told it wont beat this one. Your mechanisms are top notch.

As for metal 3d printing... If thats not worked out, what about metal casting? If you know anyone with a forge that is.

Also your usual domestic laser cutter cant cut metal. You need cutters that are in the kw range. There are services out there that do it, but usually not cheap.

 

[JOATRASH FX]

This really is looking utterly amazing.

Ive been wanting to make one of these, but truth be told it wont beat this one. Your mechanisms are top notch.

As for metal 3d printing... If thats not worked out, what about metal casting? If you know anyone with a forge that is.

Also your usual domestic laser cutter cant cut metal. You need cutters that are in the kw range. There are services out there that do it, but usually not cheap.

Thanks!

Regarding metal casting, it's not viable for this. Lost-wax casting would have been the way to go and it's notoriously difficult to find a foundry that can handle small runs at reasonable cost- I tried last year on another project, with no good results. Then you still have to clean up the parts, which runs the risk of soft edges and wonky mechanisms. I looked into having it lost wax cast in rhodium-plated brass at places like sh*peways, but the inner arm alone would have cost a couple hundred bucks for one piece and then the outer plates still need to be done!

Machining would have been ideal, at least for part of it, but again... getting precision work done on a small run of parts like this would have cost a small fortune.

If you missed it, have a look at post #36 for one of the (professionally) laser-cut samples I had done in steel. Also got a couple in thicker aluminum. They were done at a professional facility which has s cost-range that might be manageable.

 

[MangyDog]

Ahh missed your post #36 sorry, i skim read the earlier posts a bit, my bad was focusing on the amazing photos a bit too much

Assuming a home foundry setup is at all possible, casting aluminium, you can do lost wax with wax resin printing on a msla printer. The only issue is the resins can be a bit pricy.

But having a CNC mill do the piece would be better

 

[Kylo1]

The latest from the programmer. This is looking AWESOME if I do say so. I made a short animatic to give the programmer an idea of how to sequence the animations and this is what he did over the weekend.

Of course, LOTS of tweaking still to come. We're going to try to add some 'static glitches' in a few places, a 'scanning sequence' over the eye photo and some other things. One big part is how to set up the box-animations on the right. Turns out it's not easy discussing how to program abstract patterns! (Looking at the graphics done for the film, there are a lot of nonsensical-but-cool animation patterns on various screens and we'll try to capture the feeling of those.

Absolutely amazing!!!!!!!!!!

 

[JOATRASH FX]

Ahh missed your post #36 sorry, i skim read the earlier posts a bit, my bad was focusing on the amazing photos a bit too much

Assuming a home foundry setup is at all possible, casting aluminium, you can do lost wax with wax resin printing on a msla printer. The only issue is the resins can be a bit pricy.

But having a CNC mill do the piece would be better

I could print wax masters myself, and you're right, the resin ain't cheap... it's probably a $1000 investment to get started and comfortable with it, working out all the kinks! Then having them cast afterward would still cost more than would be viable. (Likely at least a couple hundred dollars for one arm at the quality level I need it to be at due to the low number volume- and the project isn't big enough to search the globe for a brilliant-but-cheap forge.) And there's still the cleanup problem. I had metal parts cast for my Mandalorian pistol, and even though those parts were 'simple', I noticed where the cleanup had taken off material.

For machining, these are so small and delicate, with precision tolerances, needing several tool setups, that would again probably cost too much. And the outer plates are too thin (0.5mm max) and complex for machining to be the best option anyway. IF metal is going to happen, they will almost certainly be laser cut because I have been able to verify the viability of it and the cost would be somewhat reasonable. There really isn't a lot of wiggle room for much more experimentation at this point and some things need to start getting locked in place.

 

[joberg]

The vid looks amazing and yes; the machining would be quite the main cost of a prop such as this one. Not much wiggle room to start with and the pricing, as you said, would be too much for a small series of props.

 

[MangyDog]

I could print wax masters myself, and you're right, the resin ain't cheap... it's probably a $1000 investment to get started and comfortable with it, working out all the kinks! Then having them cast afterward would still cost more than would be viable. (Likely at least a couple hundred dollars for one arm at the quality level I need it to be at due to the low number volume- and the project isn't big enough to search the globe for a brilliant-but-cheap forge.) And there's still the cleanup problem. I had metal parts cast for my Mandalorian pistol, and even though those parts were 'simple', I noticed where the cleanup had taken off material.

For machining, these are so small and delicate, with precision tolerances, needing several tool setups, that would again probably cost too much. And the outer plates are too thin (0.5mm max) and complex for machining to be the best option anyway. IF metal is going to happen, they will almost certainly be laser cut because I have been able to verify the viability of it and the cost would be somewhat reasonable. There really isn't a lot of wiggle room for much more experimentation at this point and some things need to start getting locked in place.

oh totally, the cost of a full setup has been one of the reasons ive not tried it myself, If i was going to do it, it would have to be something i so a lot and make money back off.... Which i wont. I know some people have made makeshift temp forges with some firebricks, a gas torch air blower and a crucible. That sort of setup could be done for maybe 100 dollars, only extra cost would be a small bottle of gas. But its janky and youll need a good bit of safe garden to do it in.

In any case this project looks amazing, and i cant wait to see more

 

[JOATRASH FX]

An important milestone reached this weekend! Not much coding done, but my programmer friend has been able to check and verify that the voltage regulator works as intended and charges the battery. The Teensy and display both need different voltages (3.7 and 5 respectively). Since there are virtually no 5v batteries out there that are small enough and easily available, a voltage regulator/booster was the solution, but until now we haven't had a chance to test the setup.

This was basically the last "will it work" question we had, and with it, all of the major hurdles are cleared. Looking at it now, I can't see anything on the horizon that will stop the basic functionality. A nice bonus turned out to be that the unit will function even while charging, so one could essentially hook it to a USB charger and display it "always on". (No idea how the screen would fare in the long run, with possible burn-in, but it's doable.)

So... here it is, running on battery for the first time:

 

[joberg]

Excellent news for sure

 

[JOATRASH FX]

OK, it's getting REAL close now. This is a near-final video of the electronics working (sorry about the shaky-cam):

Note the sequence:

1. Screen flashes once as it starts (this will happen when you activate the popup arm)
2. Red static LED lights up
3. Pressing the trigger flashes both the screen and the RBG LED (which alternates BLUE-WHITE. The programmer tried blue-white-red-white but it didn't look good from what I was told.)
4. Releasing the trigger activates the screen animations with all the info, which will repeat in a loop

Still to do:

• Sounds are still to come IF they can be done without killing the frame rate too much. Apparently, the major hurdle right now is processor power and not memory. Every extra feature slows down the entire unit. Around three sounds are planned- the "camera whine" we head in the movie, a "ticking" sound that syncs to the flashing when you scan, and a randomized background noise that will play once the info pops up. One small issue with the whining sound is that the piezo speaker can only handle maybe 200-2KHZ, so it won't be as high-pitched as in the movie. Nothing to really be done about that as it's a hardware limitation. There's also no good way to control volume, beyond sticking some tape over the piezo.
• The eye-scan will only be concentrated in the white of the eye, not the pupil. (Note that it's already randomized!)
• The scanning-wave to the right will have more 'glitches' in it. You can see a couple right now, but I asked the programmer to add as many as the unit can handle. Again, it's a frame rate issue.
• There's a small bug where the portrait of Sapper doesn't disappear if you press the scan button again, so the programmer will try to wipe the screen completely after each press.
• Under "suspect ident" I've asked the programmer to swap the serial number for "Sapper Morton", since the serial is already listed in two other places.
  
  
  
  

 

[joberg]

Wow again Imagine walking in the street with that prop? That would freak people

 

[JOATRASH FX]

Wow again Imagine walking in the street with that prop? That would freak people

I take no responsibility for any shenanigans that might ensue!