Re: [Avid-L2] Recoding DV tapes #OffTopic

What JBeck said.

Storage cost is irrelevant nowadays, at least when talking DV-PAL (or DV-NTSC for that matter). (Roughly) 25MBit (DV25)+ Audio (uncompressed) vs 1/5 of that in H.264 is not worth the trade-off in quality, imho.

704 is the correct width for DV-PAL, the missing 2*8 to 720 were for left/right digital blanking.

For 768 vs 720 (704) there is a good chart here:

https://en.wikipedia.org/wiki/Display_resolution#Common_display_resolutions

Me, I personally wouldn't deal for now with the specifics as long as I can digitize the original media in a "1:1" manner, that is: with no alterations. Everything else can be done from here, no need to compress or alter an already (from today's perspective) already flawed source material for the sole purpose of posterity. If the posterity will touch it, let them deal with it, in any aspect. But for the estate, you would take a polaroid of your family's pictures, trash them originals and leave the polas, would you?

Jochen "Jo" Hermann

filter Media Postproduction

Salzburg, Austria

On 9. Jan 2022, at 17:05, JBeck <jb30343@windstream.net> wrote:

IMO, hard drives are pretty inexpensive. Do your posterity a favor and capture in the original codec, DV25. Reencoding is always lossy. Ways to upconvert DV (or whatever) will continue to improve. Once you throw data away, it's gone. Keeping a copy in its original codec is the best way to future proof it. A 2 TB drive will hold over 100 hours of DV. A second 2TB drive will give you a backup. JB 

On Jan 9, 2022, at 9:19 AM, crispinmholland <crispinholland@gmail.com> wrote:



Hello all. I'm digitising a load of old family PAL mini DV tapes using iMovie 10.0.5 on High Sierra 10.13.6 on my trusty Mac Pro 5,1 via Firewire, the beauty being that iMovie records continuously, but breaks each original clip into a new file and gives it the title e.g. 2007-08-16 15_52_43.mov - and as a person who loves accuracy and detail, this gives me great joy and delight.

However, as DV is quite old school, I fancy making the files a lot smaller using the newer h.264 codec at around 3 - 5mbps, BUT I am conscious that PAL DV is 720 x 576 with non-square pixels, but only 704? horizontal pixels are visible. BTW these files are just for posterity, not for future editing.

So, my question is, should I simply re-encode using h264 and keep all dimensions & pixel aspect ratio the same as per the DV PAL 601 standard? My only worry with this approach is that with future technology, the ability to interpret the rectangular pixel shape might fall by the wayside as time marches on.

The other option is this: should I try to wrangle the footage into pure 4:3 shape e.g. 768 x 576 with square pixels so that it will always display correctly? But then of course, there is the issue of 704 visible pixels, so presumably this will leave a small black pillarbox?

I have Adobe Media Encoder and I am also just trying out Shutter Encoder, although I can't find any reference to 'pixel aspect ratio' in the latter.

Two more things: Should I map to 0-255 levels? And, should I de-interlace? Oh learned ones, what settings would you use?

Thanks so much for any helpful advice!