HIGH QUALITY TPU DVD Video Released from Jack Durban

[jdurban]

@Jack - I'm assuming the HI-8 deck you used for the source has a TBC? Would it help if you had an external TBC. I would be glad to chip-in to buy a TBC that we could send to you. (I know you said you will at some point upload a deinterlaced version of the DVD, but please let us know if there are any gadgets which will improve the HI-8 capture even further.)

Greetings! Actually I Do have a TBC, a DPS RC-2000. I have a full SD/HD broadcast edit bay in my lab. I also use an Elite video broadcaster video processor. I am editing a pared down video direct from the MPEG2 source.

[wattsup]

Maybe I should get my eyes interlaced. They are killing me from looking to close to the screen. Or buy some special interlacing glasses. lol

The trickiest parts are when the tpu is turned and you are trying to keep track of one part of it while it turns, then you see the wire upside down and you lose perspective and have to start over again. lol

SM made great attention when showing the FTPU to not show to much of the circuit side. He also had added two pieces of tape on the power out wires which helped mask the inner capacitor. You see this when he takes the TPU out of the case until he puts it on the video machines. A big white reflection. He also kep his hands near the device, waving them here and there to create distractions. It's too bad the cut-away scene is not in these videos.

[wattsup]

@JD

With all that equipment, you could start OU TV or the The Overunity Channel.

@All

About the pulsing, what if I had a capacitor, an inductor and a zener diode. once the voltage reached a certain level in the capacitor, would the zener open, discharge, then close and start over. Could this be done fast enough. Could the item E above be a Zener Diode instead of a capacitor?

[orbs]

I'm agreeing with EMdevices regarding the interlacing. The DVD seems to have been de-interlaced already. The quality is probably fine for most purposes, but if you want the best possible still pictures, you probably have to choose the de-interlacing algorithm manually on a case by case basis (for every single picture, like you can do with the plug-ins of VirtualDub). So the original interlaced footage at the maximum DVD bit rate 59.94i would probably be a good thing to have for TPU studies.

Edit: (I should have known better, having written a line 21 decoder...)

The DVD is in fact interlaced (i.e. progressive_sequence = 0 in the stream). However, there are two different ways how this can be encoded in MPEG-2 on a DVD: Each top and bottom frame for itself (called field picture), or both fields together with their lines alternating (called frame picture). The DVD is encoded as the latter (i.e. picture_structure = 3 [Frame picture] in the stream), which means the individual top and bottom frames cannot be properly separated again (because it is compressed in a way similar to a JPEG, which smears the lines)..

So for possibly better results, it could be either encoded as field picture on a DVD with 29.97 fps, or with 59.94 fps with each field in a separate frame in a more advanced video format.

[OzOnE]

@EMdevices, thanks for the screenshots. I see your point about the fields now, by just using them separately we should see more detail instead of blurring them further. I was still a bit confused though, as Jack's video IS indeed interlaced  . I did have a look at www.100fps.com while trying to find the best deinterlacer for the footage, but you're correct in saying that I should really rephrase that to "best field combiner / blurrer" !!

I noticed one of the plugins called "DeInterlace Smooth" converts the interlaced footage into 60Hz by resizing each field as you have done and doubling the frame rate (smooth meaning frame rate NOT image!) - it should be possible to try this sort of thing in VirtualDub itself and this is the main reason for me using it. You can see the effect of the plugins and filters on-the-fly, and it doesn't blur the image further like it sometimes can when playing via Directshow / overlay. It can get very blocky when you "zoom in" on VirtualDub though as it doesn't do any smoothing on the preview images at all. Again, this is my personal preference for using it (plus, it's free, so we can send it to anyone).

@Jack - always good to see you on this forum. I don't think any of us expected anyone to turn up who actually worked with SM directly! Just wanted to say thanks for the video - I know you're very busy and I'm glad you stuck with it after you read some of the "less positive" posts.

Is a TBC supposed to correct (or attempt to correct) the errors already present on the master HI-8 tape? For instance, on the first part of the footage, you can see how wavy the black vertical line on the left is throughout most of the video - I know this is an effect of the head jitter etc., but I thought a TBC was supposed to do it's best to correct this?

Also, if you do run another copy, do you think there is any more info on the tape that can be captured by raising the black level slightly etc? (without clipping the whites). Again, although this footage won't give us all the answers, it's one of the most important steps so far in trying to reverse-engineer the TPU IMHO .In deference to Grumpy's view a few posts back (I should know by now not to be the first fish that "bites"  ) - I'm all for the theory, but seeing that most of it goes over my head, I think I can be of more help in deciphering the wiring / coil configurations and the circuit operation first of all.)

Jack, do you possibly know who Spherics might actually be? (probably been asked before)

@all - Apologies for this, but I made a bit of an error with the files I attached previously - you will need to download a copy of DVD2AVI....

http://www.digital-digest.com/software/download.php?sid=1028&ssid=0&did=2

Then, open your first VOB file from the SM DVD ("VTS_01_1.VOB"), and the rest in the sequence should also load in automatically. Then, just go to File > Save Project, and save the new file as "VTS_01_1.d2v" in the same folder as the other files (SM DVD's "VIDEO_TS" folder). You should then be able to open the .avs file with VirtualDubMod.

The reason it fails to open is because the d2v file points directly to where MY copy of the DVD files are stored. Sorry about that - I knew I should have tested it all fully.