Video Tape to Laser projector! Image Engineering units

[polishedball]

I am guessing these are tapes that came from Kecked ? That I was going to get then he changed his mind and gave them to Steve.

If you want a svhs deck fully working, pay shipping and I'll send ya one of the two pictured.

*edit see ya got some decks* offer stands though if ya end up needing it.

[mixedgas]

If its encoded as NRZ or Manchester Coding then it can get enough samples onto a video line. It needs to sample at least twice the point rate for a good reconstruction. If you look at one of the raw, unfiltered DAC output waveforms on a scope you should see be able to measure the sample time. Or just measure the latch time for one channel.

Remember its one OR two RGBXY + Three Audio.

Steve

[telmnstr]

Bradfo69 - preserving history is what it's about!

Polishedball - very cool! I picked up some ADATs and stuff just in case! And for testing playback.

mixedgas - the A1/a2/a3 outputs actually aren't audio, they're an auxiliary TTL trigger. Perhaps they switch the analog console in/out with that, or who knows. I suppose it's left up to the end users dreams! The shows with audio have it recorded on the video tape.

I have another video I made with a bit more process and debugging, I just need to upload it to youtubes.

I haven't dumped data to computer yet (I did have a MCU but am going to hunt for a way to detect frame start instead of looking for timing shift in the clock.)

More to come!

Also, I *THINK* that maybe the 14kpps scan rate is cut in half when A/B outputs are both used at once. I think it might go to 7kpps.

[telmnstr]

Okay, after a week and a half or more of it sitting on camera here is the 2nd part to the video series:

https://www.youtube.com/watch?v=75fePEkXRfI

In the video I said I'm not sure about the A and B outputs, but after video I did a little more testing and I'm pretty sure the front edge markers signify which output to write to (A or B.) This means the framerate probably gets split in half when driving both outputs.

I'm going to borrow the analyzer again. I'm not a great programmer but I generally know what needs to be done -- just not efficient at doing it. I've never really messed with interrupts on microcontrollers (just never hit a need for my projects) but this is one. I get setting one to read the clock and snatch the data but the part that trips me up a bit is catching the timing gap for new frame. I thought about it and DMX-512 uses a similar approach I believe so I started looking into code to read DMX on AVRs, but in some of the cases I found I think they're using the hardware UART which I need to leave open for pushing data back to host. I haven't done anything with that part yet but I did try to capture it with Arduino and fast read libs. Not sure the success I had, I will be able to do better measurements when I have a working encoder in hand (not too far off.) Situation 1 I use interrupt and timer to catch the frame start. Situation 2 I find some place where I can catch a signal as to when the frame starts. One or the other will happen.


Here are screen caps from the Logic Analyzer. Image compression took it's toll a bit:
https://users.757.org/~ethan/pics/ge...eeringDecoder/

As soon as I get back on the computer that I have the project session on, I'll dump off the current state of the schematic diagram and post it as well (PDF form.)

[cfavreau]

Ethan,

How fast is the clock? It appears from the analyzer screen shots to be about 3.3uS. You could use the SPI hardware to decode the data. Since there is not an even multiple of 8 worth of clocks (I counted 66 ... You mentioned 68 I think) you would have to do some bit shifting here and there as 68 = 8.5 8 bit chunks and 66 = 8.25 8 bit chunks... however it should not be anything a modern microcontroller cannot handle. Although you maybe able to use the counter hardware to count clock transistions and just sample an IO pin when transition count changes or just put it into a tight loop using direct port reads (the Arduino API is sloooooooww....).

[telmnstr]

Okay. It's been a while. I've got two videos to post. The first is breaking down the data in the actual data stream, the 2nd one is taking a step back and computer capture of video.

This is my way of taking notes :-)

Enjoy:

Looking at the output of an encoder feeding into logic analyzer:
https://www.youtube.com/watch?v=DLWVEss4umk

What happens when you capture and play back the video via firewire<->analog bridge into the decoder:
https://www.youtube.com/watch?v=Kbl9...ature=youtu.be

I'll get down to the USB dump soon enough.Although in a psycho way it might just be possible to convert the video frames to 1 bit then over to number arrays and snatch the bits out of that.

[Laserman532]

Haha.. That's me standing next to Fred. We had a blast (and a few headaches) on that show! Watching him troubleshoot under pressure was inspiring.

hehehe - three of those 171's were mine...the black one was mine for sure...somewhere I have a picture of the entire crew for the 88 Olympics waiting outside the hotel for the van to take us to the river...Not sure why Im not in that picture...I think I arrived after load in...RIP Fred Fenning...a great mind...

[Photonbeam]

That's right Pat. You sent your tech, John ahead with your gear. You arrived the day after the hard work was done Your black 171 was very sexy. You and I had the largest arsenals of SP171's at that time in the US. I sent 4 or 5 of my sytems to Seoul for the olympics. The IEC DVE and DVD units were used to record and play the graphics part of the laser show with audio. Much of the digitizing, programming, recording and dubbing was done at the last minute on site in the front of stage control dugout or in the hotel room. There were some problems with tape dropouts and glitches in the graphics (dirt and dust in the VCR's I guess).

[telmnstr]

@Photonbeam - very nice setup in that pic! At the top of the truss setup, are those fiber launched or is there some sort of platform up there?

[Photonbeam]

The beams from the top of the truss are mirror bounces. I was impressed that they were able to target those small mirrors with scanners without drifting (much). They were using G120PDT scanners with the heaters and Fred Fenning's amps (+/-48VDC supplies) You can see a few beams missing slightly and spilling past the mirrors a little. The mirror alignment was done from a snorkel bucket lift truck.