Am I going crazy or do these images look like NTSC video?
They are very low resolution, but it's not only that I don't think - there's something more done to them... like being sent through some kind of baroque video signal chain.
We've got crosstalk typical of NTSC video being passed through a 2D adaptive comb filter during decoding: cross-chrominance and cross-luminance are both visible in this sample.
Based on the appearance I'm assuming these images were stored sequentially on U-matic tape.
If they had chosen to repeat each image twice they would have enabled perfect 3D comb filtering (SNR issues aside). Unfortunately, they didn't.
In a comment below it is stated that they were archived on Laserdisc. I can imagine the images were written to a tape master first, then the tape used to write the (probably write-once) Laserdiscs.
The tape could have been U-matic, but maybe it was digital video tape, https://en.wikipedia.org/wiki/D-2_(video) which was released in 1989. If so, the whole project would have been at the very bleeding edge technology wise.
In addition to NTSC artifacts each frame shows a very distinct kind of distortion consistent with storage on an analog videotape medium. D-2 doesn't introduce this kind of distortion.
"...Because the Department of Finance originally recorded each 1980s print as one frame on Laser Video Disks (LVDs), using analog video capture, the low-resolution images were able to be extracted for the Municipal Archives online gallery."
Comments
Am I going crazy or do these images look like NTSC video?
They are very low resolution, but it's not only that I don't think - there's something more done to them... like being sent through some kind of baroque video signal chain.
It's definitely sourced from NTSC video. Checkout: http://files.80s.nyc/photos/3/05338/0023.jpg
We've got crosstalk typical of NTSC video being passed through a 2D adaptive comb filter during decoding: cross-chrominance and cross-luminance are both visible in this sample.
Based on the appearance I'm assuming these images were stored sequentially on U-matic tape.
If they had chosen to repeat each image twice they would have enabled perfect 3D comb filtering (SNR issues aside). Unfortunately, they didn't.
In a comment below it is stated that they were archived on Laserdisc. I can imagine the images were written to a tape master first, then the tape used to write the (probably write-once) Laserdiscs.
The tape could have been U-matic, but maybe it was digital video tape, https://en.wikipedia.org/wiki/D-2_(video) which was released in 1989. If so, the whole project would have been at the very bleeding edge technology wise.
In addition to NTSC artifacts each frame shows a very distinct kind of distortion consistent with storage on an analog videotape medium. D-2 doesn't introduce this kind of distortion.
Cool, thanks.
Where Can I learn more about cross-chrominance and cross-luminance?
https://en.wikipedia.org/wiki/Composite_video#Demodulation_l...
http://www.cockam.com/vidcomb.htm
Yep, NTSC + Laserdisc. From the site:
"...Because the Department of Finance originally recorded each 1980s print as one frame on Laser Video Disks (LVDs), using analog video capture, the low-resolution images were able to be extracted for the Municipal Archives online gallery."