2

Our company has approximately 27,000 hours of video stored on 3/4" video tapes. We have begun the process of converting the video digitally with the current setup:

  • Sony U-MATIC VO-5850 & Sony UVW-1800 Betacam
  • AJA FS1 & AJA Io XT
  • Macbook Pro

We are going through the process tape by tape and need to choose a vendor to assist us in the digital conversion process.

My questions are

  1. Does anyone have experience with or can recommend a service provider in the New York Metro area?
  2. Are there any recommended steps / equipment we should consider incorporating into our current and future process?
  3. Is there additional information we should incorporate that would help with selecting a Media Asset Management system?
  4. How should we go about storing the digital files inexpensively?

Any help is appreciated.

All the best,

Kevin

3

Assuming DVD quality would be enough, you may consider archiving 720p mpeg-4 files with a bitrate of 2 Mbits, which would equate roughly to 1 hour = 1 GB data.

That would mean 27,000 hours = 27,000 GB = 27 Terabytes. So you need to consider a 50 TB storage system (including redundancy and overhead) and a tape backup system as well (LTO-6 has an uncompressed capacity of 2.4 TB per tape, so that would be relatively easy and cheap, the entire archive would fit into 13-14 tapes).

Higher quality requirements would increase the bitrate, hence the GB/hour ratio, and all values should be adjusted accordingly.

The tough part is the actual acquisition... analog video tapes have to be read at normal speed so 27,000 tapes, assuming 1 hour each, means 27,000 hours = 675 weeks = almost 10 YEARS... !!! To hope finishing this in a reasonable time (say 2 years), your company would need to set up at least 5 acquisition stations each with its own tape unit and computer (and operator). The operators, while saving, should monitor that there are not problems with the cassette and fill the metadata into the DAM system.

DAM systems, there are many, also open source, the choice depends on the use that will be made of the resulting digital video archive.

2

There are lots of things to consider before committing to an operation of this size, but there is also an imperative to act quickly. Given that it's SP Betacam and Umatic the tapes are probably beginning to reach the end of their playable life. In a decade they may all be paperweights.

A good place to start is the U.S. Federal Agencies Digitization Guidelines Initiative in particular this report on Digital File Formats for Videotape Reformatting

In N.Y. I know of a consultancy called Media Matters. We haven't used them, but we have used a report they produced as a guide in our digitisation process.

1

For that kind of quantity, you probably want to look in to a robotic solution that can load and transfer tapes unattended in to a cataloged system. Doing a quick search showed SAMMA as one promising option.

I don't have any direct experience with this particular kind of problem, but I can't imagine that manual tape loading or buying an entire robotic system could be cheaper than hiring someone with a robot to handle it. SAMMA does appear to have U-MATIC support.

0

I would recommend a very old school approach:

  • Buy as many SP and U-Matic play only decks that you can find that work and will play, not record. UMatic is dirt cheap, and SP is right next to it for a play deck.

  • Set up some old Windows XP systems. Think 5 years old CPU speed. DualCore.

  • Get your hands on the same number of Canopus DV Storm2 Cards; hard to find. - WHY? The DV Storm2 Cards allow 1 Computer to capture simultaneously three (3) streams of SD quality video 480i (Same standard as both SP and U-Matic); and will on the fly, compress it to DV Codec. Which is 12 GB per Hour of footage.

  • This was years ago (although the systems are still sitting in a closet); we had 6 of these systems set up, each with 3 MiniDV Players. So we could capture the footage at 6x3= 18 normal speed. Hand us 18 hours of tape, 1 hour to capture. All to a networked hard drive.

  • Now you will need 1 additional part; SP / U Matic will not have firewire out. Which is all the Canopus DVStorm2 will accept if you want 3 capture streams at once... But place in an analog to FW BlackMagic converter box ($300) on each output... and you're in business.

The capture quality on the Canopus cards is fantastic (for SD/DV). But the fact that if you use the above setup; you could get up and running for a very low cost:

(6) Dell XPS Workstations @ $300 Each. XP OS. (6) Canopus DVStorm2 Cards with the Add on Canopus Breakout Bay giving you (3) inputs. Pricing... hard to find. Very. (18) Black Magic Adapters @ $300 Each (18) Decks @ $300 Average Price considering UMatic should be free.

$12,600 total cost not including the DVStorm Cards, which when available can go for a few hundred to a thousand. Depending on if theres a buyer.

27,000 Hours of tape, if you hand feed the decks, keep them running in 2 shifts, 15 hours a day = 15 x 18 = 270 Hours a Day Captured. Add a third Shift: 396 Hours a Day.

Youll need rewinders, or extra decks to keep things moving. But thats what I would do.

Low cost, high output. You could set up 10 systems if you can budget it. But with 6 running 24x7 you could get your capture done 70-80 Days if you stay on it. Double the systems: 30-40 days.

Hope that helps!

  • I just realized I answered a really really old topic. Woops. – McFlySoHigh May 30 '16 at 15:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.