Page 35 of 38 FirstFirst ... 253132333435363738 LastLast
Results 341 to 350 of 379

Thread: CYGN-B

  1. #341
    Join Date
    Sep 2014
    Location
    Colorado USA
    Posts
    521

    Default

    Quote Originally Posted by mixedgas View Post
    There is/was this fellow named Dr. Pierre Brosens who is/was early team member or co-founder of GSI.

    Long ago as an undergrad I was delighted to find one of his papers in the library on optimal open loop control.
    I doubt I can draw the curve correctly but the frequency response graph had three regions as frequency increased.
    That is a 30 year old memory. So take with a grain of salt.

    There was a method in the paper of taking a specified width pulse at 1/2 the amplitude of the command signal, inverting it, and adding it to command to put the brakes on any resonance and thus stop the motion quicker.

    The paper is behind a pay wall now and I really doubt I have a 30 year old photocopy of a two page paper.

    From Left to Right, Linear, Resonant, and Ballistic.

    So besides the integrator you add a very narrow bandpass with a lot of gain (Or invert a sharp notch* ) at where you want your resonance.

    Steve
    Thanks Steve. I may have that paper, I'll check my GSI folder. I can attest from experience that the G-115 open-loop galvos resonant frequency peaks in the 40hz range, and sending a 45hz -90hz square wave to the XY inputs where one axis is slightly off in frequency to the other resulting in the slow, phase cycling image was one of my favorite effects using 1:1 and 1:2 XY ratios. I been intending to make some "modern day" videos of this but can get a roundtoit.
    ________________________________
    Everything depends on everything else

  2. #342
    Join Date
    Mar 2010
    Posts
    277

    Default

    Update: I finally have the parts ordered for the 351 data receiver and the Audio Mod board. Audio Mod is one of the cycloid generating boards from the original Laserium 6b image synthesizer. Among other effects, this board produced something apparently called "triangle audio mod" which I remember as being a very good and unique effect.

    I expect to have data receiver experiment results in a couple of weeks. Hopefully a train of pulses will be present from which the timing of the zero crossings, and hence the 351 data frames, may be extracted.

  3. #343
    Join Date
    Mar 2010
    Posts
    277

    Default

    Photos show the Laserock2 data signal played from .wav file through the first three op amp circuits of the 352 card, which have been built.

    The signal, when captured by the data logger, looks a promising candidate for data extraction based on timing and zero crossing detection.

    Confirmation / correction of my interpretations welcome.

    I hope we haven't lost certain parties from this discussion. The master taught so well the student hasn't needed to ask a question in a while.

    Ron, if you're here, You gave me the audio to Starship on my request a while back. Might it be possible to get Laserock2 audio? That would be helpful as the audio I have has poor levels.
    Attached Thumbnails Attached Thumbnails first_3_op_amp_circuits_of_352_card_1.jpg  

    first_3_op_amp_circuits_of_352_card_2.jpg  

    first_3_op_amp_circuits_of_352_card_4.jpg  

    first_3_op_amp_circuits_of_352_card_5.jpg  

    351Data_captured_1.png  

    351Data_captured_2.png  


  4. #344
    Join Date
    Jun 2009
    Location
    St. Louis, MO
    Posts
    943

    Default

    I'm still here. There just hasn't been anything I needed to say. It looks like you've got things well in hand.

    BTW A friend of mine might be traveling through Buffalo in early December with a truck. Would you be able to come down for some xmas gifts?

    Click image for larger version. 

Name:	351 frame.png 
Views:	5 
Size:	26.7 KB 
ID:	58495

    Here's a 351 frame I cleaned up in Audacity

    I've been playing around with creating stills of classic laser images forever. Now I've started rendering sequences to video with ffmpeg. I reinvented some of the techniques computer animators use to make up for the lack of persistence of vision that laser shows depend on. I asked Jon if he was okay with me doing The Blue Danube for the 50th anniversary - and he was. I'm thinking of putting, "This video was made possible by Moore's Law" in the credits.
    Last edited by laserist; 11-23-2021 at 10:11.
    "There are painters who transform the sun into a yellow spot, but there are others who, with the help of their art and their intelligence, transform a yellow spot into the sun." Pablo Picasso

  5. #345
    Join Date
    Mar 2010
    Posts
    277

    Default

    Splendid! Thank you Ron, for curating and making available your amazing archive. As you mentioned, and I fully agree, The folks at LII put a LOT of work into selecting the music and then putting those pieces together into a show that was PACED properly.

    Wow this sequence of music from all these different bands really hangs together as a unified musical narrative like nothing else I've ever heard. Each song makes the next sound even better. Can't wait to finally get to look at the choreography.

    For convenient reference, Here is the list of music from Laserock2:

    Is Anybody Out There? - Pink Floyd
    Turn Me On - The Tubes
    Double Life - The Cars
    The Farther Away I Am - Daryl Hall
    Walking On The Moon - The Police
    The Wait - Pretenders
    I'm Not In Love - 10CC
    The Sahara Of Snow, Part 1 - Bill Bruford
    Cars - Gary Numan
    Duchess - Genesis
    Red - Sammy Hagar
    School - Supertramp
    Run Like Hell - Pink Floyd
    Planet Claire - B-52's
    Rock 'N' Roll - Led Zeppelin

    Attached is a screen capture of the 351 data frame map, provided by Ron early in this thread. Excitingly, and for the first time seen publicly, each of the following hex strings are the 20 bytes described in this document.

    I have built an extraction utility for reading the signal into data. Two snippets are given, one showing various errors that occur in the signal, and a few frames with no errors that probably show a small clip of actual choreography.

    ERROR: invalid frame header
    ERROR: invalid frame header
    ERROR: time value out of range
    ERROR: time value out of range
    ERROR: failed to detect expected 1/4
    000295 000000EA00009BAA7F6FFCFFFEFFFFFF11FBF8
    ERROR: time value out of range
    000296 000000EA00009BAA7F6FFCFFFEFFFFFF11FBF8
    ERROR: failed to detect expected 1/4
    000297 000000EA00009BAA7F6FFCFFFEFFFFFF11FBF8
    ERROR: time value out of range
    ERROR: time value out of range
    ERROR: failed to detect expected 1/4
    000298 000000EA00009BAA7F6FFCFFFEFFFFFF11FBF8
    000299 000000EA00009BAA7F6FFCFFFEFFFFFF11FBF8
    000300 000000EA00009BAA7F6FFCFFFEFFFFFF11FBF8
    000301 000000EA00009BAA7F6FFCFFFEFFFFFF11FBF8

    005528 00FFFFC40000454A80AFFFFFF1FFF2FBFFFBFB
    005529 00FFFFC30000454A80AFFFFFF1FFF2FBFFFBFB
    005530 00FFFFC10000454A80AFFFFFF0FFF2FBFFFBFB
    005531 00FFFFBF0000454A80AFFFFFF1FFF2FBFFFBFB
    005532 00FFFFBE0000454A80AFFFFFF1FFF2FBFFFBFB
    005533 00FFFFBE0000454A80AFFFFFF1FFF2FBFFFBFB
    005534 00FFFFBD0000454A80AFFFFFF1FFF2FBFFFBFB
    005535 00FFFFBD0000454A80AFFFFFF1FFF2FBFFFBFB
    005536 00FFFFBC0000454A80AFFFFFF1FFF2FBFFFBFB

    Brian: What you're doing sounds great. I'd sure like to see The Blue Danube, or any samples of your work as it progresses.

    I'd love to collect the collection you've been trying to get to me. I'll do what I can to make a Buffalo rendezvous possible. Unfortunately I'm heavily booked until the end of the year (weekdays) and getting a day off on a previously agreed on booking might or might not be possible. I'll talk to my employer. Please let me know when you know anything about the day this could happen. Is it stuff that would fit in the trunk and back seat of a car?

    Honestly, I'm scared of the challenges involved in renting a vehicle, navigating to another city and crossing the border and all. I have money, a license, a passport, and vaccines, but I haven't done much driving in the last 20 years, and the last time I crossed the border I was a kid with my parents. It's manly the navigating part I'm worried about. I might be able to bring a friend to help with that if it's any day but Saturday.
    Attached Thumbnails Attached Thumbnails 351DataMap.png  


  6. #346
    Join Date
    Jun 2009
    Location
    St. Louis, MO
    Posts
    943

    Default

    The Laserium data channel had no error correction built in. Originally the decoding was done by 3 boards. The clock, timing, and dac boards were replaced by the 352 board to free up space for an additional dmux board when a whole bunch of stuff was added for the Beatles show. The original set of boards had a trim pot that was part of a phase locked loop circuit that occasionally needed to be adjusted to deal with errors. (In the dark – by touch - during the show) In the best of worlds it was possible to sense bad data and just shut down the output. The digital signals would just hold in whatever state they were in and the analog signals had low pass filters that would hold the signal up with a slow decay. Unfortunately in the worst case the bad data sailed through and was ugly and noisy. But that was infrequent.

    The show tapes were recorded directly from the masters. I believe there was a circuit to clean up the data when the show tapes were made, but in the field we had to record copies backwards to get good data.

    There are three stages for my Danube project. The first is to render the Lissajous figures. I’m still playing with this part. The second step is the star field – almost everybody is doing fulldome starfields in planetariums today – and they’re doing it in real time. The third step is photographing the Lumia. That’s going to require a new camera and about 7200 individual images. Maybe twice that. And almost certainly some kind of post processing. I figure the trick to making this work is figuring out where and how to cheat to get the look I want.

    It sounds like way too much of a hassle for you to get to Detroit. I'll mail them.
    "There are painters who transform the sun into a yellow spot, but there are others who, with the help of their art and their intelligence, transform a yellow spot into the sun." Pablo Picasso

  7. #347
    Join Date
    Feb 2015
    Location
    San Francisco
    Posts
    101

    Default

    Quote Originally Posted by laserist View Post
    The Laserium data channel had no error correction built in. Originally the decoding was done by 3 boards. The clock, timing, and dac boards were replaced by the 352 board to free up space for an additional dmux board when a whole bunch of stuff was added for the Beatles show. The original set of boards had a trim pot that was part of a phase locked loop circuit that occasionally needed to be adjusted to deal with errors. (In the dark – by touch - during the show) In the best of worlds it was possible to sense bad data and just shut down the output. The digital signals would just hold in whatever state they were in and the analog signals had low pass filters that would hold the signal up with a slow decay. Unfortunately in the worst case the bad data sailed through and was ugly and noisy. But that was infrequent.
    Interesting. In the 20 years I performed the show, from the Mark 4 to the CSX, I never once had to touch the 352 card! My only issues were external, with ground loops messing up the data, and THAT was usually when the planetarium was modifying their sound system or some such!

  8. #348
    Join Date
    Jun 2009
    Location
    St. Louis, MO
    Posts
    943

    Default

    The 352 card was a complete redesign, and much more stable in my experience than the dmtr/timing/dac 351 version. That said, if I'm reading the 6b-352_cage_wiring doc correctly the audio transformers were added to the mark IV after St. Louis opened in 75, and there isn't one for the data channel shown on the hybrid card cage diagram for a Mark VI dated 6-10-77, after we got our Mark VI. So, I'm fine blaming it on a ground loop. It's possible Carl had me add the transformers as part of the Beatles upgrade, and that's why I thought the 352 card was superior, but I can't remember.
    "There are painters who transform the sun into a yellow spot, but there are others who, with the help of their art and their intelligence, transform a yellow spot into the sun." Pablo Picasso

  9. #349
    Join Date
    Mar 2010
    Posts
    277

    Default

    Here is a link to a test of my cyc live image generation and rendering software I did a few years ago. The video is what I remember of the choreography of the number Double Life from Laserock2. The main feature is the lick at 1:15 where the third oscillator of the CYGN-A is brought in. I have an audio cassette recording of a live show at the McLaughlin in 1981 where one can hear the audience response to that lick. I'm looking forward to looking at the original 351 data for that.

    https://youtu.be/1F_QQLxN_vY

    Ron: I didn't realize you did Laserium for 20 years, that's amazing. Did the 6b cycloid generators play any role in the CSX projector?

    Brian: Those loops with the four scan pairs and the fixed rotational offsets is such an iconic image. Absolutely love that.

    Yeah, driving to Buffalo or Detroit is not impossible, but kind of problematic. There's the added business of the screening I fill out each day at work that asks if I've been outside Canada in the past 14 days. If you would mail the stuff and let me know the shipping fees, that would be much better for me.

    All: thoughts regarding how to approach inspecting the 351 data. Is there an existing platform that allows audio tracks and voltage outputs to be synchronized that I could write a file conversion utility for? Should I write software that produces ILDA files from the 351 data or renders it to video?

    Speaking of those great looking loops, If anyone at any point is inclined to give me danube board documentation, I'd be happy to build one.
    Attached Thumbnails Attached Thumbnails DoubleLife_experiment_still.png  


  10. #350
    Join Date
    Jun 2009
    Location
    St. Louis, MO
    Posts
    943

    Default

    That number from Laserock II would have been hurt when they replaced the slotted can chopper with the AOM for the Beatles show. They could have preserved the soft edge of the chopper effect and let the laserist turn it on and off, but they didn't. There's no way I could actually reproduce your video recollection with the CYGN-A and chopper. I might be able to come close, but with a 6B you could only bend the string so far.

    I wrote something in VB to read a wav file and output a delimited text file with the 351 data. It's an awful hack - I can get a computer to do whatever I want, but my code is almost unreadable even to me after a little time goes by. Then I import the delimited text into Access and fix the errors. The nice thing about sticking it into a database is it allows selective editing. You can always write a bit of code to regenerate a 351 wav file that syncs to the original audio. (One of the data fields is a time stamp of the frame.)

    Half the fun is figuring out which way works for you. If you want people to be able to run your interpretation of a laserium number frozen in time - then ILDA is a great choice. If you want to do real laserium stuff live - it's a horrible choice. If you want to do art for art's sake - you may need to figure out what that means. It's journey, it's not worth the trip unless you enjoy it.
    "There are painters who transform the sun into a yellow spot, but there are others who, with the help of their art and their intelligence, transform a yellow spot into the sun." Pablo Picasso

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •