Capture from VHS to 59.94p or 50p: best practices & goals 7
Support ProVideo Coalition
Shop with Filmtools Logo

What Do You Think? Let Us Know.

Subscribe
Notify of
guest
11 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
David

This could not come at a more opportune moment. I’m about to embark on a journey of digitizing over 600 VHS tapes. Just purchased an S-VHS player with S-Video out. Eagerly awaiting your recommendations on analog to digital conversion methods.

Eric

You’ll want to check out Digital FAQ as well. They’re very hardcode about quality and their components over there. https://www.digitalfaq.com/guides/video.htm

Jon Worthen

Need some advise please. I have a few VHS tapes from Argentina (PAL) format. I am in the US and am looking to convert to video file. What equipment will I need to go from PAL to NTSC?

Jon Worthen

Thank you for your answer. I do not think I have a player that can play the PAL VHS. How can I tell if I do. If I do not have one that will work what would be your suggestion to convert to MP4 compatible to play in the U.S. – Thank you in advance for your guidance.

David P.

Thank you for the excellent article, Allan.
I have several video tapes that I would like to transfer to digital files, including VHS (NTSC), 3/4″ SP (NTSC), Betacam SP (NTSC) and DV (PAL). My only intended purpose for the material is preservation at the highest possible quality, for archival and to maintain future flexibility. My knowledge of video production predates the digital and non-linear eras. I have never digitized video footage before, and want to make sure it is done correctly.

After reading your article, I think I want: optimal frame rate (59.94P (NTSC) and 50P (PAL)); and output resolution (960×720). Is that correct? Your logic in selecting these specs makes perfect sense to me, so I’m surprised I have not seen them listed on transfer facility websites.

Given my archival aim, I’m also uncertain about two other things: 1) the best option (from your list) for head-switching artifacts; 2) the desired digital file format/codec and color sampling.

Transfer facilities list many file formats/codecs (which are only vaguely familiar to me): Prores 422 HQ, MOV, XDCAM, DCP, Quicktime uncompressed 8- or 10-bit 4:2:2, etc. Given my desire for maximum quality, should I choose uncompressed 10-bit, or are the large file sizes simply not worth it? And is 4:2:2 color sampling only appropriate for a true component format, like Betacam SP? If so, what is best for (Y/C-outputted) VHS and 3/4″? (I’m Mac-centric, if that is relevant.)

My final question relates to DV. As the information on tape is already encoded in digital form, what is the appropriate computer file format?

David P.

Many thanks for your reply, Allan.

My remaining questions relate to workflow, especially whether to deinterlace, upscale, etc., during capture (via hardware) or post capture (using software). I know you’re planning to cover this topic in upcoming articles, but I hope you will feel able to share some of your insights in advance.

Postings I’ve seen on Reddit forums mostly recommend the following workflow:
– Capture the original SD interlaced analog source format in D1 resolution (NTSC: 720×480; PAL: 720×576), still interlaced, to an uncompressed (or lossless codec) file. This is your digital archival master file.
– Do not upscale SD video during capture. Upscaling/(oversampling?) the frame resolution does not capture additional information.
– After capture: post process (using software) to deinterlace (NTSC: 59.94p; PAL: 50p); resize to square pixels; upscale and remove head-switching artifacts (if desired).

A highly regarded transfer facility here in Los Angeles (that I will likely use) says it digitizes tapes “to their native line/frame rates or we can up-convert to HD,” with each step priced accordingly. So, it’s best to know what I want in advance. [I’m not sure if this would all be done in hardware or not.]

With all this in mind, here are my questions:
— Question #1: Do you agree with the “Reddit” comments and 480i/576i workflow outlined above? Or is there merit to upscaling and deinterlacing during capture, say to 960x720p, while also removing head-switching noise?
[One Reddit post said upscaling to 960x720p (not sure if referring to during capture or later using software) would result in “less noticeable video encoding artifacts.” But another comment warned this upscaling risked losing resolution by upscaling less than 200% (from 720×480), per the Nyquist sampling theorem. They recommended 1440×1080, instead.
Any thoughts?]

— Question # 2: Should I specify capture to an uncompressed 10-bit file, or is this too unwieldy/overkill? Do you have a preferred lossless codec for archival purposes?

Thank you, again, for sharing your knowledge.

You Might Also Like