MovieStuff PAL transfer question

[MovieStuff]

Hi, all!

Okay, quite a few people have shown interest in a PAL transfer from the DV8. Unfortunately, the DV8 has a built in NTSC camera and the computer system it transfers to isn't PAL compatible. In other words, I can only transfer to NTSC.

Now, correct me if I'm wrong (and no doubt people are standing in line to do so) but isn't video recorded on miniDV basically just a file that contains a bunch of digital stills? Unfortunately, I'm not totally hip on miniDV technology (the curse of being an old **** ) but it would seem like there would be a way to do the following:

1) I transfer at 30fps with no pulldown

2) That is put onto standard NTSC minDV

3) A PAL client then somehow takes all those individual frames and digitizes them into his/her NLE system for playback at 25fps instead of 30fps.

Surely I'm missing something here, which is why I post this for the current crop of media whiz kids to solve. I'm fairly certain I could do this if I exported all the frames as a number still sequence but burning that to a CD would be numbingly slow and not very cost or time wise. Does anyone know of a way to do this same thing to miniDV tape?

-OR-

Is the firewire data stream universal? Certainly I can't just feed the 30fps NTSC firewire signal into a PAL miniDV camera, right? It can't be that simple, can it?

Roger

[blackangus1]

What type of camera are you digitizing to -- is it DV? If so, you really don't want to do this.

NTSC DV is 720x480, sampled at 4:1:1. PAL DV is 720x576, sampled at 4:2:0. The data streams are not compatible on tape. And the 4:1:1 transcoded to 4:2:0 is going to deliver worse color than a PAL-originated transfer would.

[MovieStuff]

<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">Originally posted by blackangus1:
What type of camera are you digitizing to -- is it DV? </font></td></tr></table></td></tr></table></BLOCKQUOTE>

The DV8 uses a broadcast 700+ line Hitachi analog camera with component outputs going into my RealTime Pinnacle board for frame by frame capture. The RealTime board converts anything going out in real time to all the major types of signals; component, SVHS, composite, seriel digital, firewire, etc. So the output of the system to minDV is going to be standard NTSC firewire (for lack of a better term, here). Despite the differences in color sampling, couldn't the separate images be utilitzed and re-rendered at the proper frame rate somehow for PAL?

Roger

[blackangus1]

First things first -- what you want to do (just transfer at 30fps, send that to DV, and let the recipient convert it to PAL) is likely to produce an unsatisfactory result. The 4:1:1 decimation, then reconstruction, and then 4:2:2 re-decimation, may result in less than satisfactory color reproduction. But maybe it'd work good enough? You can check out a software PAL&lt;-&gt;NTSC DV converter here:
http://www.canopus.com/ppt_dvformatconverter.htm


I'm not clear on what your computer does with the images fed to it by the camera: are you currently capturing a series of still frames? Or is the board converting the input from the camera into a DV AVI?

30fps progressive NTSC, converting to 25fps progressive PAL, would be a nightmare. A sequence of 24fps progressive film stills, converted to 25fps PAL, would obviously be much more palatable (pun intended)... but I'm sure you already know this, so the question is... what am I missing?

NTSC capture will lose 96 lines of resolution as compared to PAL capture, which is certainly not optimal, but if you're sampling in 4:2:2 component, you'll at least be getting a good color sampling from which to make the final tape (assuming DV output?)

How do you output to tape now (NTSC)? Do you capture in a mammoth file, a series of 720x480 stills, and then render that out in Premiere into a compressed DV .AVI file? If so, it would seem a trivial task to instead import the same still picture sequence into a PAL project, apply the "resize" filter, and render out a frame-for-frame transfer, 25fps. The process of creating the AVI file will automatically divide the progressive frames into the proper field order. Many (perhaps most now?) DV boards can handle both PAL and NTSC, but you would still have to invest in a PAL recording deck. You could get something like one of the dual-format Sony decks which record both modes, or you could do it the cheap way and get a cheap $600 PAL DV camcorder and plug it in. All that has to work is the tape transport, the lens & CCD's and all that would be irrelevant in the process, you'd be treating it like a dumb record deck.

The quality loss would be primarily limited to the upsampling from 480 NTSC to 576 PAL lines. Seems simple, technologically, but it would probably add a significant bit of work on your side...

[MovieStuff]

I thought I remember reading that miniDV footage had no "fields" in the classical analog video sense and that the images were just sequential in nature (sorta like a bunch of JPEGS) and that the fields were "applied" during display by the playback system. Am I way off the mark here?

Now ASSUMING that is true, then (color sampling differences notwithstanding)it would seem that the NTSC frames could simply be imported into a PAL NLE program for playback at a different rate. The differences in resolution could then be attenuated by simply selecting the "maintain aspect ratio" function of the NLE program. That would put the image in the middle of the PAL picture with black at the top and bottom; quasi-letter boxed. In that particular case, there wouldn't really be any loss of resolution since the NTSC image would still be displayed by the same number of lines as normal, even though it wouldn't be as high a resolution as full screen PAL would be.

Again, am I just whistling Dixie, here?

Remember, there is no movement from one field to the next within each NTSC original frame since the DV8 transfers frame by frame.

Roger

<font color="#a62a2a" size="1">[ July 07, 2002 01:13 PM: Message edited by: MovieStuff ]</font>

[crimsonson]

"but if you're sampling in 4:2:2 component, you'll at least be getting a good color sampling from which to make the final tape (assuming DV output?)"

4:2:2 does not apply here since component is an analogue system. 4:2:2/4:1:1 applies to digital. You meant chroma bandwidth.

the 96 lines of res is negligible since S8 aspect ratio is closer to NTSC than PAL. They have the same horizontal res at 720. The vertical is different. the loss is smaller than 96 lines.

1.36:1 of super 8 gives you about 530 of vertical res at 720. So the difference is really only about 50 vertical lines. Using an NTSC camera with 720 it can only see 480 of vertical- missing the 50 more super 8 offers.

DV is field based.

"30fps progressive NTSC, converting to 25fps progressive PAL, would be a nightmare. A sequence of 24fps progressive film stills, converted to 25fps PAL, would obviously be much more palatable (pun intended)... but I'm sure you already know this, so the question is... what am I missing?"

the frame rate of NTSC does not apply since you are getting JPEG (or what ever) stills. You are actually getting 24p.

I still think the color is the x factor. Giving JPEG stills will compromise the quality since the compression will be twice - once by JPEG and the other is the NLE's of the customer.

an NTSC camera and an uncompressed image format (TIFF, PICT, Bitmap, etc) will give you almost the same quality as using a PAL DV camera with DV8. Assuming you are using a compressed NLE (DV or MJPEG).

[MovieStuff]

<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">Originally posted by crimsonson:

an NTSC camera and an uncompressed image format (TIFF, PICT, Bitmap, etc) will give you almost the same quality as using a PAL DV camera with DV8. Assuming you are using a compressed NLE (DV or MJPEG).</font></td></tr></table></td></tr></table></BLOCKQUOTE>

It seems as if we're holding this conversation on both this forum and Andreas'!

Okay, so let me get this straight: If I capture from my Hitachi camera via component, frame by frame (which is how the DV8 works) and then export that footage from the time line as a numbered bitmap sequence, the PAL end user can import that numbered sequence into his/her PAL NLE system and render it out as a 25fps PAL video clip.
Furthermore, if I snug up the super 8 image to the top and bottom of the NTSC frame during transfer, that frame will end up in the middle of the PAL frame with some black top and bottom, assuming that they enable the "maintain aspect ratio" function of their NLE system and the loss in resolution would be minimal.

And finally, since I am avoiding the entire miniDV compression issue, the color of the numbered bitmap sequence should pretty much look like it does over here from the original transfer.

Now, assuming all this works, my only REAL issue is trying to expedite the exporting of the bitmap files, which on my older dual PII system is REALLY slow. Also, I would not be able to fit much on a CD. Is there any way to save this information on a miniDV tape? Not as a video signal but just the raw digital information?

-OR-

I suppose if they wanted to send me a precious 1 gig micro drive, I could save it to that! An expensive proposition but should work, right?

Roger

[crimsonson]

i replied it at Andreas.

I did a test with Avid DV and the quality is the same for both JPEG and TIFF converted to DV.

So, at least with my test, DV8 NTSC JPEG stills should have about the same quality as if you shot it with PAL cam and recorded in PAL DV- minus the 50 lines of res.

24x speed CD Burner can record 650 megs fairly quick - about 4 minutes (maybe less).

You cant use DV tapes to store non DV stream data. There was a guy who wrote a program that did it but it was just beta and you will need every customer to have installed in their computer.

[blackangus1]

<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">Originally posted by crimsonson:

the 96 lines of res is negligible since S8 aspect ratio is closer to NTSC than PAL.... 1.36:1 of super 8 gives you about 530 of vertical res at 720. So the difference is really only about 50 vertical lines. Using an NTSC camera with 720 it can only see 480 of vertical- missing the 50 more super 8 offers."</font></td></tr></table></td></tr></table></BLOCKQUOTE>

Doesn't work that way: PAL pixels are not the same shape as NTSC pixels. NTSC pixels are taller. Both PAL and NTSC are 4:3, but PAL fits 576 lines in the same physical space that NTSC only gets 480. If you did a pixel-for-pixel transfer from NTSC to PAL, without re-sizing, but with adding black at top & bottom, the resulting picture would be significantly vertically squeezed. Again, see the pictures at:
http://www.canopus.com/ppt_dvformatconverter.htm

Their site shows the effects of different styles of converting NTSC to PAL: either leaving the data intact and adding black bars, or scaling/interpolating up to maintain aspect ratio.

<BLOCKQUOTE><font size=2 face="Verdana, Helvetica, sans-serif">quote:</font><table border="0" width="90%" bgcolor="#333333" cellspacing="1" cellpadding="0"><tr><td width="100%"><table border="0" width="100%" cellspacing="0" cellpadding="2" bgcolor="#FF9900"><tr><td width="100%" bgcolor="#DDDDDD"><font size=2 face="Verdana, Helvetica, sans-serif">
an NTSC camera and an uncompressed image format (TIFF, PICT, Bitmap, etc) will give you almost the same quality as using a PAL DV camera with DV8.</font></td></tr></table></td></tr></table></BLOCKQUOTE>

Other than the inherent vertical resizing issue, and as long as you're sampling at 4:2:2. I mean, it's up to the customer to determine whether the vertically-stretched image is okay.

Too bad it's not going the other way -- PAL down to NTSC -- because in that case you could employ a minor anamorphic lens and optically accomplish the aspect ratio squeezing. You'd film a bit too much, then crop the lines and end up with a 480-line master. But going NTSC-&gt;PAL, you're either going to center the image (and end up with the wrong aspect ratio) or scale it up.

[MovieStuff]

But if an NTSC frame is saved as a JPEG or bitmap file, doesn't the original pixel shape become academic? The PAL NLE system won't know whether the pixel came from a digital camera or a video camera or a photo scanned on a flatbed scanner, will it?

Roger