Mechanical TV on Youtube

Forum for discussion of narrow-bandwidth mechanical television

Moderators: Steve Anderson, Dave Moll, Andrew Davie

Mechanical TV on Youtube

Postby aa9dt » Tue May 15, 2007 6:04 am

<h3>We should be on Youtube</h3>
<b>Mechanical TV on Youtube.</b> Check out this mechanical TV post at Youtube...

<b>Pioneer.</b> The poster is a true pioneer! Since this is impressive as a first effort, I apologize for my criticisms below. Anyway, the problems are there.

<b>No genlock.</b> Here's my impression of the Youtube post: You can just see the image of a dancer. The mechanical video isn't genlocked. Periodically, a vertical scan line sweeps across the picture. (That is, the horizontal or frame sync seems out of whack. Because the picture came from a disc scanner, the picture syncs vertically.) The focus could be better, too. Accompanying music <i>almost</i> makes the video watchable.

<b>Annoying Lines.</b> Youtube provides most people's only exposure to mechanical video. They must think that those annoying, flipping lines plague all mechanical video. How can these images impress anyone?

<b>Promotion opportunity.</b> As a recent issue of our journal says, we need to promote our club. For NBTVA, Youtube is an important and overlooked public relations source.

<b>We must improve Youtube images.</b> Club members should think about ways to get better images on Youtube.
<li>The most direct way is to start a new standard. This standard would maintain our scanning method, but increase the scanning speed. The object would be video that we could convert to Youtube format. (I don't know what frame speed Youtube uses. 30fps? 15fps?) </li>

<li>Another way is a Youtube scan-converter program.</li>

Some of you are already experimenting with increased disc speed. Now you have another reason to pursue these experiments.

Laboratory Assistant
Posts: 24
Joined: Wed May 09, 2007 10:09 am

Postby DrZarkov » Tue May 15, 2007 7:04 am

It should be easy to genlock it, we use 12,5 frames, "normal" TV in Europe uses 25 frames. You just take a cheap webcam (or a better Eyecam in your Macintosh), and with many webcams you can freely choice the framerate. I have currently no webcam, because I never had a proper use for it. .. As a Genlock you just have to use your computer for generating the picture, you can do it all in software. Where are the programmers amongst us?
User avatar
I think I've had a cranial implosion.
Posts: 1037
Joined: Mon Feb 19, 2007 11:28 pm
Location: Kamp-Lintfort, Germany

Using a webcam as a picture source.

Postby Steve Anderson » Tue May 15, 2007 2:55 pm

You just take a cheap webcam, with many webcams you can freely choice the can do it all in software. Where are the programmers amongst us?

This was exactly my thinking. I am in the process of creating .wav files that are simple test waveforms, but accurate, no A-D will be used, all generated in software.

However the line-rate will not be exactly right, it'll be 400.909Hz, 0.23% fast. This is due to the CD sample rate of 44100Hz being divided by an integer, 110. It's a shame that CDs don't use the same sample rate as DAT or AES/EBU digital audio data, 48kHz.

From there I am hoping to be able to use some web-cam capture software to create an .avi file (say) then convert it to monochome, adjust the aspect ratio (basically a crop), then convert the format to 32 lines and save as an 8-bit .wav file.

But this won't happen overnight!

I also note that there appears to be very little discussion/posts regarding mechanical scanners/cameras. Not a criticism, just an observation.

Steve A.

Hang on...thinking about it, I'm making .wav files, they can almost be at any sample rate you wish, including 48kHz, as long as your soundcard/motherboard supports the sample rate. Hmmm, let me consider this...It will also make the files slightly larger. Or there's the option for going to 32kHz sample rate.

The advantage with 44.1kHz is that it's easy to create an .iso file (an audio CD image) and burn it to disc. There are sample-rate convertors, both sofware and hardware, but I think they may generate too many artifacts for clean NBTV.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
Posts: 3965
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Postby Klaas Robers » Tue May 15, 2007 9:02 pm

Steve, what I did in the recent past is that I made the file first at a sampling frequency of 4 x 44.100 = 176.400 Hz. Then the NBTV lines fit onto the sampling rate. Then run a low pass filter of 10 kHz over the file (FIR-filter) and the do subsampling (i.e. take one out of 4 samples and discard the other 3) by a factor of 4.

I did this when making the first test with Compatible Colour NBTV for CD and also Gary Miller does this when he makes the NBTV files in the CCNC standard. So the sync frequency is correct with the precision of the cristal of your CD-player. I have an option in my monitor to lock the disc to a cristal and then I can watch NBTV from CD without real sync for tens of minutes.

For the processing of the taken files for the CD's 2 and 3 I made a program that did upsampling by a factor 8 to 352.800 Hz. Then I did all my processing. i.e. clamping, gain control, time base correction and insertion of new sync, and at the end I did subsampling to come back to 44.100 Hz. The advantage was that I could cut and paste the video signal using a standard wave editor (Goldwave) without interruptions, DC jumps and jitter in the sync.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
Posts: 1502
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Return to Mechanical NBTV

Who is online

Users browsing this forum: No registered users and 5 guests