ASCII by Jason Scott

Jason Scott's Weblog

How I Do It, Buffered By Cries I Am Doing It Wrong —

The last entry garnered an awful lot of attention. While I would normally write in such a way as to move around between unrelated subjects, sometimes I’m backed into a corner to do a follow-up or second part; the amount of people who have wandered into my field of vision to ask me for deep details of what they variously call my toolchain, setup, or approach has now overflowed my life’s efficiency.

Here is a quick time-saving suggestion: Skip everything I’m writing down to the nice picture of the videotapes. Then read what’s after that and don’t comment.

Still here? Let’s go.

There are many things which set me off, but few have been as consistent as Nerd Bullying, especially where people who consider themselves kings of very tiny kingdoms take their knowledge and smoosh it in a bowl with their social awkwardness and lack of cue-observing and come out of the gate with a way-too-overtuned neg-throwing style and tone criticism.

By all rights, this should be a minor, almost rounding error of an experience, but part of what I do on the daily is often a process or procedure and Nerd Bullies looooove these linear steps because it’s one dimension away from a flowchart and they feel they need to add one more dimension, one more fix, before it’s either better, or perfect.

Now, of course they have no idea what minefield they’re wandering into with me, or the disproportionate response it often garners when they’re “just helping”, but that’s also because I’m translating their words on the fly:

  • “Not sure if you’re aware” – Salutations, Dilettante Moron
  • “Pro Tip” – I think my opinion matters so much it should be canon
  • “Curious ______” – Everything after the word “curious” is always awful
  • “Why are you choosing to _____” – Everything after the words “choosing to” is similarly awful

And so on.

Their weirdly-framed criticisms or “Not a question, more of a comment” ends up doing nothing to move my needle, and in a few cases, gets a profanity-laden rant they are completely unsure how they signed up for.

Therefore, when people ask/demand how I “do” things, I tend to hesitate, because instead of providing Education, I feel I am providing Ammunition.

This is all, however, my problem, not anyone else’s. Of course people stumble through life, confused and various levels of paranoid, unsure if they’re doing the right thing or oblivious to the fact they aren’t. They found solace in the churning sea of life’s short run on a floating plank of easy-to-master geekery and in their minds it becomes an island from which to send up signals of their prowess before their bad diets send them into the darkness at shockingly high numbers and shockingly low ages.

And, of course, marbled like fine steak among this crowd are some truly generous and thoughtful folks, who honestly do want your opinion on what you do, realizing that someone else’s decisions aren’t edicts or judgements on their lives and choices. They benefit hearing the footsteps of another soul’s journey and taking warmth or warning from what they observe.

It’s to these folks I am now speaking.

Let’s get something out of the way: Depending on how you count it, I have, in my inbox, roughly 100,000 videotapes to process.

With numbers like that, choices are going to be made. Choices of whether to do anything much with it, how far I’ll get along in the expanse and functional reality of time, and what exactly a by-tape approach will take. I’ve spent a lot of time and a lot of consultation to come up with what’s going on currently, and the journey is not over.

Also, the choices are going to shift over time, as new opportunities present themselves, costs or income interfere, and long-term trends with the equipment and materials come to light.

At this point, I’ve digitized roughly 2,000 tapes. Last week I digitized 60. Who knows what next month will bring. My process is a snapshot, not an immutable declaration.

What I Do And How I Do It

One of the most variant choices to make are what Codecs your final video files are going to be in.

For the tapes of value or uniqueness, I am doing Lagarith, a chonky little codec which is lossless, and then compressed. It works out to anywhere between 25-50gb an hour. It is memorably huge. It has been around a long time, and it can be converted into a mass of lossy compression schemes down the line. It leaves the most options open in that direction. It is a pain in the ass to transfer, taking hours to upload and download. I keep them interlaced, again to provide the most options later.

For tapes where I will be dealing with many thousands (think, someone has recorded years of news programs), I use various types of MPEG-4, which is lossy (drops bits in the name of saving space) but gets the information across effectively, and the notably reduced disk space usage means that having a half-million hours of programs will be within the realm of cost availability. Should a tape go by that is so incredible it will need to be ripped via Lagarith, that command will arrive someday and it will be dealt with.

I do not throw out any tapes. They are all stored after digitization.

Currently I use an analog to digital device/card called a BlackMagic Intensity Pro 4k Card, which is about $240 and which does the job very well. It is compatible with a wide range of software, does not get flustered easily, and is not a massive mystery as to how it exactly works. It can work with HDMI, Component, and S-VIDEO cables, and I use the latter two. I currently have five of these cards in use.

I don’t care about brands of cables.

I use whatever cable is generally thought of as the “best” connection between the videotape deck and the outgoing signal. In order, that’s generally HDMI -> S-VIDEO -> Component -> Composite for me. Since none of the decks I work with have HDMI conversion (nor do I want it), they almost all are S-VIDEO right now.

The cables from the BlackMagic Intensity are going either into the deck, or through a “Time Base Corrector” which is located inside of a Panasonic DMR-ES10 DVD Recorder. The cable going into the DMR is either an S-VIDEO or a Component cable, but right now it’s always an S-VIDEO going into the card because why not.

I did not always use Time-Base Correction at the beginning of my process, but I use it all the time now. When you’re dealing with terrible, terrible, terrible tapes, you want to run them through this device, which helps with the signal to make it synchronize with the sound and not drop frames in a way that the result is a broken, fuzzy, off-kilter mess. If I don’t use it, it’s simply because I am dealing with the best of the best tapes, or I recently got a deck and I haven’t purchased one. In general, I use them all the time.

Finally (on the hardware side), I use a variety of VHS and UMATIC decks, and will be adding BETAMAX and BETACAM decks as well in the future. In general, I go for brands that are either well-regarded, or passable with my being given them for free playing a part.

Right now, the weirdest decks I use are SONY SVO-9500MD decks, which normally go inside MRI machines. They have very, very, very capability and they’re built like tanks. Someone who donated one to me had gotten it, and his insight caused me to love this model type very much.

I also use some variation of JVC S9600U SsVHS machines, because they have a lot of circuitry inside to take absolutely terrible tapes and make them look better. When I was dealing with bootleg rock tapes that are on their 3rd-5th generation, this makes a difference.

For the UMATIC tapes, I’m currently working with a SONY VP-5000 machine. This machine is hell on earth to work with, but when it all comes together, the image and output is very nice. The JVC and 9500MD decks put out SVIDEO, while this VP-5000 puts out a BNC connection I convert to component into the Time-Base Corrector.

Now you know the “Hardware Chain”.

For software, I am currently using Virtualdub, which is a rather old piece of video processing software, but which works very well for the straightforward “take in what’s coming on the BlackMagic card, and turn it into a file”. I can run two of it on a single Windows machine, with two BlackMagic cards in it. When I’m done with the files, I can use a third instance of Virtualdub to crop either side of the recording so I don’t have minutes (or hours) of blank space being stored in the file. So, in this way, Virtualdub is both a capture, and a finisher.

I do not process the files further. I do not de-interlace. I do not convert them to another codec beyond Lagarith. I put them on USB drives and I upload them as is, with the filenames as the metadata that was written on the tapes, if any.

Again, for the more “Tape to File” mindsets:

  • VHS or UMATIC Tape
  • Sony SVO-9500MD or JVC S9600U SVHS, or SONY VP-5000 UMATIC
  • Connected via SVIDEO or Component Cable to Time Base Corrector
  • Panasonic DMR-ES10 DVD Recorder being used as a pass-through Time Base Corrector
  • S-VIDEO and the Audio Cables into a BlackMagic Intensity Pro 4k Card for each Deck
  • Capturing video via Virtualdub for each card, running in Windows 10 or 11
  • Lagarith Codec, Nothing Special Added,
  • Cropped via Virtualdub and uploaded with Filename as Metadata

This is how I do, and this is how I do it.


Come back in a year, and I may have changed any part of this – this is not a final decision with no negotiation, although I have talked to more people than most might expect, and the choices here come from a real place and aren’t random.

The tapes are not thrown out, and can be revisited. I do not let myself worry that I get one shot with the tapes – in almost all cases, I’m going after tapes that would never have seen any treatment at all, so what I’m doing in the aggregate is already better than expected.

This is all very hard on the decks. I expect to go through many along the way. I do not throw them out when they break, yet, and will see what goes on.

My experience is USB digitizers (you plug the video into a little device, and the little device goes into a USB port) are not dependable.

Everything I digitize, I share immediately. People find errors and I have fixed them.

I am focused on doing the same thing decently enough 100,000 times, not doing a tiny handful of things perfectly. If you have 20 tapes to work with, no doubt you’ll be different. You may be different all the way down, in fact. Have a ball. Don’t tell me.

See you on the stream.

Categorised as: Uncategorized

Comments are disabled on this post


  1. grimmtooth says:

    Nice to see activity here again, even though I hate the reason behind it.

    I love how you’ve managed to so succinctly describe the act of “Nerd Bullying”. I have been using the phrase “Small Fish / Smaller Pond” to describe this kind of person (we have a lot of them at work since we often buy smaller companies and and the integration process is fraught) but your description has so much better thinking in it. Imma steal it but will attribute where appropriate šŸ˜€

    I don’t have a horse in the video capture space but I do like your choices, especially that of fidelity over efficiency. That seems to be the mission, here. From your discussions of this process in the podcast it’s clear the short feedback loop has helped catch mistakes before they get buried too far in the outbox, sort of like CI in software development šŸ™‚

    Keep up the good work.

  2. Keith says:

    Hi, very interesting. Thanks for sharing.

    I can only afford one between the SVO-9500MD or JVC S9600U.
    Which one would you suggest?

    I shouldn’t have more than 50 VHS tapes to process.
    So, since long term reliability isn’t paramount, maybe the JVC would be a better choice?
    Even if the Sony unit really looks badass šŸ™‚

    • Jason Scott says:

      Absolutely the JVC. The SVO-9500MD can’t handle anything that has any issues (or last more than 2 hours recording) so the family of tapes that will work in it is much larger.