CYGN-B

[Greg]

Speaking of monsters, I forgot to post evidence of this coincidence. A few weeks ago I was readying Mercury Blues for export, and a real one appeared briefly a few doors down.

[lasermaster1977]

I've had the privilege of doing some more 'digging' into the dark recesses of Laserium's DOGN interface that Greg shared with me. I started this back in Feb. 2022 but soon got sidetracked about 3 months later, but returned to the hunt recently. One thing that caught my eye was a 234 byte Hex listing described in the DOGN documentation as the "World". This block of 8-bit Bytes was probably at some point stored on the 2nd of two 2K ROMs installed on the DOGN (actually, the DOGN card had up to four 2K ROM or EPROM chips installed with a SPDT switch that selected either of two ROM chips at a time.) Greg has more familiarization with these ROM sets than I do

I knew from the side-notes of the DOGN documentation of this Hex listing that it was the North American continent plus Alaska and a bit of Mexico down to the Yucatan peninsula.
Last night I finally typed in the 234 X,Y Hex values into my Apple IIe Windows emulator memory and saved this block of memory as an Apple binary file in the same manner as I did with my own laser image files. I loaded a BASIC program I wrote years ago to plot an image's XY points on the Apple II/IIe high resolution graphics screen, loaded the "World" binary image file and plotted it.

When the "World" point plotting first appeared on the screen, the continent was split in half with some of it on the far left side of the screen, the middle of the screen blank and the rest of the "World on the far right side of the screen...and upside down. I knew the likely reason for this anomaly and adjusted my program to deal with 2's complement values.

And that did the trick as shown here, an emulation of the Apple IIe monochrome screen:

[Greg]

When the "World" point plotting first appeared on the screen, the continent was split in half with some of it on the far left side of the screen, the middle of the screen blank and the rest of the "World on the far right side of the screen...and upside down. I knew the likely reason for this anomaly and adjusted my program to deal with 2's complement values.

A round of applause, everyone please!

[Greg]

I suggest what is needed is a utility which either of us could write for the bela that will contain the points data from the ROM files as arrays. The utility can repeatedly loop through the points, converting two's complement to decimal, repeating each point a certain number of times to slow the pps down, and output a stereo wav file which can then be used as an image source in the 6b Emulator.

[lasermaster1977]

I suggest what is needed is a utility which either of us could write for the bela that will contain the points data from the ROM files as arrays. The utility can repeatedly loop through the points, converting two's complement to decimal, repeating each point a certain number of times to slow the pps down, and output a stereo wav file which can then be used as an image source in the 6b Emulator.

Nice! Might I suggest, converting two complement values to the bela audio wav format values whereby there is an intra-XY point delay that achieves the desired image framerate. We're probably saying the same thing.

[Greg]

Courtesy of lasermaster1977 and his AppleIIe wielding prowess, we have a visualization of some points data recovered from the DOGN EPROMs from The Police show. The imagery, however, seems unrelated to a The Police show. lasermaster1977 succeeded in locating the appearance of the rainbow cadenza graphic at the end of this video:

https://www.youtube.com/watch?v=-Jc2RQIWh0I

[lasermaster1977]

Thanks Roj.

Here is the Rainbow Cadenza image plot with the missing end segment of 'NZA' restored to its rightful place.




For some odd reason this end segment was at the beginning of one 2K ROM that immediately preceded a string of ASCII symbol characters shown previously and here again:




I still working on unscrambling a set of images that was on one of the Pink Floyd 2K ROMs.

I just sent Greg text files of each of these image's XY point list, with hopes of soon seeing them rendered in laser scans.

[Greg]

Hey, I found that bottle graphic, that I thought was from 'Message in the Bottle.'😎

Neat! I keep forgetting to search for numbers on youtube before diving into restoration. I have that XYZ number from the Rush show from a 1/4" reel. The underlying choreography in that video looks similar or the same as what I have, but I think what is going on is that an image generator called the ADIG supplied much of the imagery that we are seeing in the video. I have some docs from Ron that give a catalog of the ADIG built in graphics, and most of those graphics are in there. I don't know why there is stuff on the DOGN EPROMs that are unrelated to the shows they are labeled for. As lasermaster1977 and I discussed, perhaps the chips were partially re-written without being cleared at some point, or received a dump of something that contained obsolete points behind the intended content.

lasermaster1977: I see why you think some of the plots look like control voltage envelopes, but I'm pretty sure the DOGN never provided control voltages to anything. Just image offsets, dogloids and graphic frames. I looked through my notes and found a page with some code that might be a points reading routine (attached pic) but please be forewarned, as I don't want to send you on wild goose chases, that I wrapped up my DOGN work about four years ago, and I don't remember exactly what is what from back then, except that the ported code works great for producing the dogloids and dynamic offsets on the C64.

[Greg]

lasermaster1977, or anyone else interested, I suggest anything you want to do involving the bela just ask me first, that will probably save you time. I watched those videos and asked questions on the forum so you won't have to. As time permits, I'll look into what is needed to get those sensors you purchased working with the AD signal IO and file read / write codebase I sent you.

[lasermaster1977]

Neat! I keep forgetting to search for numbers on youtube before diving into restoration. I have that XYZ number from the Rush show from a 1/4" reel. The underlying choreography in that video looks similar or the same as what I have, but I think what is going on is that an image generator called the ADIG supplied much of the imagery that we are seeing in the video. I have some docs from Ron that give a catalog of the ADIG built in graphics, and most of those graphics are in there. I don't know why there is stuff on the DOGN EPROMs that are unrelated to the shows they are labeled for. As lasermaster1977 and I discussed, perhaps the chips were partially re-written without being cleared at some point, or received a dump of something that contained obsolete points behind the intended content.

lasermaster1977: I see why you think some of the plots look like control voltage envelopes, but I'm pretty sure the DOGN never provided control voltages to anything. Just image offsets, dogloids and graphic frames. I looked through my notes and found a page with some code that might be a points reading routine (attached pic) but please be forewarned, as I don't want to send you on wild goose chases, that I wrapped up my DOGN work about four years ago, and I don't remember exactly what is what from back then, except that the ported code works great for producing the dogloids and dynamic offsets on the C64.

Great investigative sleuthing, Roj! It cools to see the image pieces fall into their choreographed place.

Greg, when I was referring to CVs, I was extrapolating the use of 8-bit values being used as inputs to the DOGN 8-bit DACs for "image offset control voltages". To me a CV is a very broad term with many applications. (CVs in music synth terms seem to have a very narrow meaning...to me.)

Another point of clarification on why "there's stuff on DOGN EPROMS that are unrelated to shows they are labeled for...", and I was thinking about this just the other night.
To "reburn" an EPROM back then the clear window on top of the chip had to be exposed to UV light for a period of time resulting in total erasure of data. Once an EPROM chip was erased the clear window would be covered with an opaque material or sticker prior to having new data written to it to keep any new programming to the EPROM from "fading away". The data bytes, however long, must be written to the EPROM burner in one action into a blank EPROM. It would seem plausible and highly likely to me that the manner in which the raw 2K data was created and stored on some type of computer and/or its floppy storage disks back then in 2K chunks is where the image data juxta positioning must have happened, intentionally or unintentionally.

I hope we can find the PF YT video of the "sleeping and awake head side profile" with the puzzle pieces overlays.