Old Shred Tape Archive Project Blog

[Allan]

Hey,

As some of you may know, I recently invited all the old-school freestylers on the Facebook Group to ship me their boxes of old video tapes so that I can digitize them and backup/upload them to the cloud for safe keeping. This is a significant undertaking, but I'm up to the task I figured that, as I go, it would be a good idea to blog my progress on what I'm capturing. I'm hoping that I can detail the cataloguing process with some interesting enough detail that I think it warrants me creating a separate blog from my players blog. So, I'm just going to dive right in without much of a plan for writing this. Here we go:

To date, Sunny Genz has sent me a box of 80 tapes, spanning from about 2000 - 2005,6? (that's ~160 hours of footage, people!) This is a serious library of some of the best of the scene during that period. Sideline shred from Worlds 2002? SO much good stuff in there.



As stoked as I was about that treasure trove, my intent with this project is actually to reach back in time further than this. My concern is that, as 8mm & VHS tapes age, they start to rapidly loose quality and can actually become unplayable. So, the earlier tapes I get take top priority in the queue. Tapes from 2000 onwards still have, like, 5-10 years before things start to get critical. But 1990? Shit is 25 years old! I need to be careful with these tapes. Tonight I joked that I should actually be receiving the tapes in a static-free environment, with gloved hands and only using clean-room level machines for playback. I would if I could

As it happens, over the weekend I got a box of 23 tapes from Jay Mouldenhour. Dated from 1992. These are the tapes I so desperately want to save before it's too late! There is early Jay & Peter Irish and Paul Munger and Kenny Shults in there!



Kenny commented on Facebook (where I also posted that photo) that the Redwood Spring Jam was his favourite session. Tonight after playing, Matt came over and we popped that one in and watched it. DAMN. Those guys are awesome. There is really something special about watching Kenny doing HUGE PS Whirl to PS Whirl in 1993. Wait until you see the spandex!

Peter Irish is sending me a box of tapes soon. Dennis Jones might be sending me a box too. I want MORE! When I've got a good portion of these completed, I'm going to start soliciting more of the old peoples who were there at the beginning and see just how many tapes I can save

I've got a great start on things and am currently in the process of uploading my first test files from this batch from Jay. My next post is going to be about my process and how I'm going about actually doing this project, from receiving the boxes to capturing them and uploading them to "the cloud."

I'd also like to clarify that, at the moment, I actually don't have a plan as to how we're going to make any of these tapes public. My priority is to simply digitize and backup. Exactly how those tapes make it out to the public remains to be seen. A website of some sort is sure to be setup, but the details of that are mirky at this point.

Anyhow, yeah. Any questions?

[Asmus]

Maybe take a small payment for downloads. To cover the cost of upload/download/hosting and some of the profits could go to the IFPA as well. I would totally be up for paying for this.

Good job man!

[boyle]

This is such a great project, looking forward to seeing how it evolves.

[h0ag3yb3atZ]

There is really something special about watching Kenny doing HUGE PS Whirl to PS Whirl in 1993. Wait until you see the spandex!

Oh man I am ready for 15 year old PS Whirl footage and of course spandex


Great project Allan, props to you for preserving all that history.

[acxel22]

this is christmas in july!

[Allan]

OK, I hope this doesn't come across as too scattered. I'm freestyling this I thought about doing a proper outline and trying to compose something more formal, but fuck it.

There is a whole lot that I could write about my history with capturing video from analog sources like 8mm or VHS tapes. I've been doing it since the late 90s, though it has been a while. The process itself hasn't changed much, but it has become much easier to deal with; computer processors and disk capacity increases means that there's a lot less to worry about.

The typical capture process means hooking your camera (or VCR or other playback machine) to a computer via a Firewire connection and firing up some software like Adobe Premiere or Final Cut Pro; the editing software has an import window where you can see what's playing on the camera, allowing you to control rewind/forward/pause/etc. You record the portions you want into video files on your computer; these files are typically quite large compared to almost every other type of file.

The main thing to note about this is the 1-to-1 time consumption. With digital, the camera records the files directly to the camera's memory (no tape), and copying them to your computer for editing gets faster with time and technology; you can copy an hour-long clip in minutes or less these days. With analogue, you have to capture in real-time; you capture as the video plays off the camera. Got two hours of footage? It takes 2 hours to capture it.

This process is laborious. If I'm going to be capturing a couple hundred hours or more, I need to do something to cut down on the overhead of editing software (I'm not editing any of this, after all) and managing files and file names and backing everything up. I need to be able to ditch the editor entirely and go with software that works from the command line so that I can write a (simple) script around it to take care of basic file management.

This might sound kinda ... complicated? Weird?, but, currently, I've got a working prototype system running on less than 10 lines of bash code:

#!/bin/bash
# ./cap project tapename seconds
# e.g. ./cap jay river-rock-spring-jam 7200
EXTHDD=/Volumes/TeraBack/ff-archives/$1/
FNAME=$1-$2.mov
CAPSEC=$3
echo "Starting Capture at $(date +%T)"
./videosnap -d 'DCR-TRV320' -s 'SD480' -t $CAPSEC $FNAME
mv $FNAME $EXTHDD
echo "Successfully moved the file to the external drive, destined for the cloud."

It will work without even needing a monitor hooked up the computer. Of course, the awesome command line capture utility I found called VideoSnap makes this all possible (with some limitations I'm trying to work around).

I've got an old Digital 8 camcorder hooked up to my Macbook with an external 1TB disk. I fire up my terminal app and type something like:

#./cap jay spring-jam-93-III 7200

"cap" is the script I wrote; the rest are the variables: "jay" is the project name (and the name of the folder the video gets put in); "spring-jam-93-III" is what's written on the label; "7200" is how many seconds of video capture.

I hit 'play' on the camera, 'enter' on my keyboard and walk away (or continue to use my computer for other things as it is with my prototype).

At the end of the 7200 seconds (2 hours), the captured video file gets copied to the external disk (I tried capturing straight to the disk, but it's not reliable enough). The folder on the external that all the files get put into is monitored by another application (currently Arq) that automatically detects new files and uploads them to my Amazon S3 account ($0.03/GB). I'll look at a redundant service for this; I still need to investigate options for public-facing hosting too.

Originally, I had actually planned on using my Raspberry Pi computer to do all of this, but it turned out is wasn't quite up to the task; every possible solution ended being complicated and eventually I gave up on the concept. Perhaps one of the newer generation models of these sorts of mini-computers will enable me to do this (it's not the software, after all, it's the hardware here). I'm currently investigating options, but I'm actually leaning towards just getting an old Mac Mini. It would be trivial, I think, to swap out VideoSnap for ffmpeg ... but I digress.

As I'd really prefer not to use my Macbook for this (I truly hate cables), at some point in the not too distant future, I'll have a headless (no monitor) system that will be connected to my network and the camera, all wrapped up neatly into a box. Push button tape-to-cloud conversion I should run a side business.

There are quite a few details that I've left out here (H.264 encoding etc) that I'll pick up next time. For now, I'm beat. I've got 3 of Jay's tapes captured now and the 4th is in progress

Onwards!

[Tjuggles]

Oh man Allan, you are.

Just you are.

[max]

Hi Allan, great read and great project! Do you send the tapes back to their owners once captured, or are you also creating a physical archive somewhere?

What are the main reasons why you wouldn't upload all of this to youtube? video quality? DRM / DMCA related? I ask as forthe uninitiated like me, Youtube sounds like a good mass-market cloud-based video archival method.. although you are then stuck with only one way of viewing the content.

Also, are you looking into ways of increasing the capture volume either by way of more computers or more capture cars / vcrs (unsure if you could capture 2 sources in parallel from say a VCR and a video camera on the same computer through different interfaces)?

Still, awesome project you're running. Sounds very time-consuming, but I can nearly feel your excitement from reading your blog!

[Allan]

Hi Allan, great read and great project! Do you send the tapes back to their owners once captured, or are you also creating a physical archive somewhere?

I am sending the tapes back when I'm done. I don't have the space or the desire to have a bunch of disintegrating plastic hanging around

What are the main reasons why you wouldn't upload all of this to youtube? video quality? DRM / DMCA related?

First and foremost, I'm capturing raw, full-length tapes with no editing. I don't know about you, but when I'm at a tournament, sometimes I record stuff other than footbag; some of this stuff might be personal. So, I wouldn't want to presume to just upload everything wholesale without the owners having a chance to vet what's actually in the footage. I could make them private and edit from YouTube itself, only publishing the vetted footage, but:

Secondly, I simply don't agree with the direction that YouTube has gone. The bait and switch they pulled has been leaving a very bad taste in my mouth over the last year. I never expected them to remain free forever, but I also didn't expect them to turn into a shittier version of broadcast television. It's bullshit and I won't support it. That being said, some of it may end up on YouTube eventually, but that will be up to the individual owners of the footage. I've uploaded my last videos to that site and plan to delete all of my old videos at some point soonish.

Vimeo is my go-to alternative, but I'm also investigating a self-hosted solution.

Thirdly ... I had another good reason, but it escapes me now

Also, are you looking into ways of increasing the capture volume either by way of more computers or more capture cars / vcrs (unsure if you could capture 2 sources in parallel from say a VCR and a video camera on the same computer through different interfaces)?

This is interesting. Honestly, I'd not really thought of doing this, but it's a pretty damn fine idea. The limiting factor here is that I really only have one camera/playback machine that works; I've another, but it's got something wrong with it where it won't put a picture on the screen. I use that as a rewinder. I could pretty easily pick up another machine for cheap at, like, a pawn shop or something though. I'll keep my eyes open.

After I got another playback device, then I'd need to look at the computer side of things. I've not done _any_ research about this at all, but I've a hunch that there aren't any consumer level systems that would let you capture two video streams at the same time. Even if I had a tower-like PC with two firewire cards in it, I'm betting the software might not exist ... ? Something might done with a batch capture feature, but I don't know.

It's worth investigating more, but in any case, I'm already working on a seriously constrained budget, so I'm not entirely sure if I want to afford all the extra equipment and time it would take to set it all up. If I wanted to run that side business doing this, it would be well worth the investment.

As it is, the capturing process is going faster than I expected I've already got 6 tapes of Jay's 23 that he sent captured and uploaded; doing about 3-4 tapes a day.

I've got another technical post up my sleeve, but it's been a super-long day and I wanna work on other stuff.

Thanks for the questions, Max. Thanks for the support everyone else

[Allan]

Ugh. I'm coming across tapes that have large swaths of borkedness (yes, that's the technical term). I seem to be able to fast-forward through these sections and keep capturing, but there will be files with 5+ minute blank spots in them where I wasn't playing attention. I've got the little LCD screen on the camera itself playing as a heads up display of sorts, so I can see sort of what's going on when I glance over it every now and then; I'm not really watching most of it. I plan on watching it all later from the computer, when I can pause and go back and slow-mo stuff. Watching it while it captures is actually kind of painful.

While it sucks that these sections are borked, it's actually interesting that they're there, as they are the exact reason why I felt an urgency to rescue this stuff.

[Kylescook]

Just read everything, glad you started this blog. Appreciate you taking this project on.

[Allan]

Just tried out one of the few tapes that doesn't have a label on it: Xmas Jam 1996 Pretty sure there's some Torch on there.

Edit: thanks Kyle

[Muffinman]

Ugh. I'm coming across tapes that have large swaths of borkedness (yes, that's the technical term). I seem to be able to fast-forward through these sections and keep capturing, but there will be files with 5+ minute blank spots in them where I wasn't playing attention.

I seem to recall that the last time I used a VCR, it had automatic functionality to fast-forward such borked portions of tapes (in addition to commercials). Maybe you could look into picking up a pawned machine that does that?

[Allan]

I seem to recall that the last time I used a VCR, it had automatic functionality to fast-forward such borked portions of tapes (in addition to commercials). Maybe you could look into picking up a pawned machine that does that?

Weird. Never heard of that. Thing is, these are 8mm tapes, not VHS. I've got some VHS coming from Peter, and Jay has some VHS that he might still send me, but most of the stuff I'll be converting is 8mm/hi8mm.

I'll do a bit of investigating, but budget is a serious limitation. I'm really very poor

To date:

[Allan]

Following a discussion about it with Jay Moldenhauer (owner of the tapes I am currently capturing), I'm happy to report that he is OK with making the raw tapes public He's going to have a highlights video edited from it at some point, but he's given me the go-ahead to do a website that publishes the tapes in their full, raw glory. As I said to Jay: these tapes are history, so every second is a highlight

I'm 15 tapes deep into Jay's 23 tape library. Once I'm done capturing, I'll hopefully have decided on a publishing venue and can publish the whole library at once. In the meantime, as a teaser, here's the first hour from a tape named "Redwood Spring Jam III":



It's 1.6GB, H264 encoded mov file. Your browser should just stream it, and most, if not all, media players should be able to play it back. It was the first thing I captured, and only grabbed an hour as a test...

I'm hosting this file directly off modified's server, which has 2TB/month of bandwidth. I think that this should be sufficient for the amount of downloads this will receive. Realistically, if more than 12 people actually view this, I'd be surprised At the same time, I'd prefer not to tax the server unnecessarily (it could actually slow the forum down if tons of people try to download at once), so please don't post this to the Facebook group or anywhere that's likely to generate significant traffic, OK? This is just a teaser for people reading this thread

For now...

[Asmus]

Maybe use Gumroad? https://gumroad.com/

[F[uns]tylin' Eclectic]

Haven't commented yet, but have been thinking a lot about this. I'm really excited for it. I hope everyone shows them as much appreciation as they're worth.

Thanks for slaving away at this project, Allan.

Sent from my SM-G928T using Tapatalk

[Muffinman]

Niice.
That spandex.
For 8mm, can't you just throw them in a converter tape to treat them like VHS? Isnt that how we used to play them outside of the camera?

[Allan]

For 8mm, can't you just throw them in a converter tape to treat them like VHS? Isnt that how we used to play them outside of the camera?

If we were just watching them, that'd work; but, we're trying to get the best possible quality out of these already crappy tapes. They were crappy when they were new and now it's even worse. We're already up-sampling the video to 480p (720x480 which is actually 640x480 with black bars on the sides) from its native 240p. This is possible because I'm using a Digital 8 camcorder that does the conversion for us. The D8 converts the Video8 signal and outputs it via Firewire. There are a million cheap, shitty USB video capture devices out there, but they all suck; Firewire was designed for video, and for our purposes, it's perfect.

When I get around to doing VHS tapes (Peter has a box), what I'll be doing is a tricky thing where I hook the VCR directly up to the camera, which again does the conversion for us and outputs a nice, up-sampled DV signal. If I were to put the 8mm into a VHS converter box, we'd be down-sampling the video to VHS resolution ("poor"), then trying to up-sample from that to turn into a DV file that isn't teeny-tiny. While it's possible (and might even produce a viable file), I think the fewer conversions we have, the better things will look.

Make sense?

[Allan]

Maybe use Gumroad? https://gumroad.com/

I'm still thinking about ways to pay for some of this stuff. Gumroad seems cool. Thanks for the link I'll certainly keep it in mind.