DV DVCPRO DVCPRO50

Can someone tell me what the differences are in the above DV file formats?

The most obvious difference is the data rate/size of the files produced per unit measure of time. DV and DVCPRO have video data rates of about 25 Mbps (plus audio). DVCPRO50 has twice that data rate. Higher data rates usually equate to better quality, faster frame rates, and/or larger frames. If your files were originally recorded on a DVCPRO50 device, then you may wish to maintain a theoretically better recording by keeping it in this format. However, there would be little reason to convert from a more compressed format to this format in general use. The differences between DV(25) and DVCPRO are more technical in nature than obvious to the casual user.

If you are not starting with video that is DV saving it as DV will not make it look better and will take a massive amount of space. The only reason you might is to use the DV file in iMovie or Final Cut but those programs can take QT files and make them compatible and essentially do for you what exporting to DV would do anyway.

You cannot make low bitrate video better by increasing the bitrate generally just like a low resolution digital picture cannot be made a lot bigger and look as good.

I know that I can export it in QT as .dv and then import it into iMovie.

Unfortunately, the conversion of MPEG-2 ("muxed" VOBs, MPEG-2, MPEG-1, etc.) using QT Pro, iTunes, iMovie, FCE/FCP and similar applications will result in a loss of audio. On the other hand, third-party MPEG based applications like MPEG Streamclip (free) and ffmpefX (free) can easily handle such problems. (Of the two, I prefer MPEG Streamclip but it requires that Apple's MPEG-2 component be installed for conversions.) Since your goal is the eventual editing of your files in iMovie and iMove works natively in DV, I would recommend converting directly to DV here to avoid any secondary conversions upon importation. While your MPEG-4 source files can be drop/import converted directly to iMovie, I still prefer external conversion in MPEG Streamclip in cases where I wish to crop/scale "overscan" areas devoid of media content or wish to adjust brightness, contrast, saturation, or volume for a file as a whole.

In this case, Guy, or may I call you "new", you have a completely incompatible format for iMovie, so MPEG Streamclip (and you ahve to pay for the Apple MPEG-2 playback component for the MPEG-2 files) is your goal here. And you cannot us QT at all for MPEG-1 or 2 with muxed audio.

Steps:

1. Open file or files (if you number then file1.mpg, file2.mpg, file3..... you can open a string of files in order)
2. Fix time code errors. Critically important.
3. Select export to DV.

Wait for it to finish. Ensure you have 13GB per hour of space for the final product. Use the .DV file in your application. Just understand it does not have the DV quality or even close to it, this is merely to get a compatible file. MPEG-4 if made in QT should work OK. Or yuo can export to DV from QT as it is not muxed.

Am I correct in assuming that mpeg streamclip or QT will convert into .dv (25fps)?

Yes. If using the MPEG Streamclip "Export to DV" File menu option, simply select the PAL "Standard" rather than the default "NTSC" pop-up setting.

Personal recommendation would be to import/convert/edit in the standard of your original file. (E.g., if you have a PAL DVD, convert the VOBs to PAL DV, edit in iMovie as a PAL, and change frame rates if required dependent on your final method of distribution.)

ADDITIONAL NOTES: As I understand it, QT still uses a simple drop/add strategy when it comes to changing frame rates. That is, when going from one frame rate to another, it simply "drops" a frame or "duplicates" a frame where needed to achieve the traget frame rate. MPEG Streamclip, on the other hand, includes an option to "blend frames" when changing frame rates which I prefer. (I.e., the human eye is more sensitive to displacements in motion than optical resolutions so this blending reduces "jerky" motion and replaces it with a "fuzzy" blending of adjacent frames which the eye is less likely to detect,) So I generally prefer using MPEG Streamclip for conversion to a final edited format if frame rates are to be changed at all.

MPEG Streamclip, on the other hand, includes an option to "blend frames" when changing frame rates

Have you compared MPEG Streamclip's PAL<->NTSC conversion with JES Deinterlacer?

I recently tested this: with JES Deinterlacer the motion may be somewhat smoother than in MPEG Streamclip but the difference isn't very big. MPEG Streamclip seems to more blend neighbouring frames than JES Deinterlacer. MPEG Streamclip also completely loses the interlacing lines in some scenes when it blends frames. JES Deinterlacer always keeps interlacing lines visible.

Have you compared MPEG Streamclip's PAL<->NTSC conversion with JES Deinterlacer?

No, I am not familiar with JES at all, (Will seel if I can locate some information and read up on it and/or try it out if time allows.)

MPEG Streamclip seems to more blend neighbouring frames than JES Deinterlacer.

Yes, that is the way it seemed to me. If a frame is to be dropped, it is merged/blended with an adjacent frame which is meing kept. On the other hand, if a frame is to be added, the newly generated intermediate frame is a blinding of the the frame just before and just after the new frames location. Really looks bad (i.e., very blurry compared to other frames) but is much better than simply dropping or duplicating frames. So what sort of strategy does JES use?

MPEG Streamclip also completely loses the interlacing lines in some scenes when it blends frames. JES Deinterlacer always keeps interlacing lines visible.

When deinterlacing, MPEG Streamclip "enables a special motion-adaptive deinterlacer, and it deinterlaces the lower field in the parts that contain motion, and preserves video quality of all parts that do not contain motion; the upper field is left unchanged." Assume that explains the difference here.