Why you should use Log and Transfer with Tapeless Media

[Jeremy Garchow]

[Hector berrebi] “i do not trust them…. not after trusting the most expensive DVD’s few years back and realizing it was a mistake,
“

[Hector berrebi] “LTO, with 800 gigs for about 50$ sounds more like it… “

I am totally with you. I cannot tell you how many times I’ve gone back to an archive DVD from 10+ years ago and have it not mount. Even though it was verified, babied, etc. I just don’t like it.

LTO is awesome.

[Jeremy Garchow]

[Wayne Carey] “My experience with tape storage has been that it is slow to work with.”

No too much naymore. LTO4 (and now LTO5) is about fw800 speed. Way faster than disc.

My Cache-A machine does not need to unarchive the whole tape in order to get one file. It’s slick.

[Wayne Carey] “Yes, it does work but tape drives are too expensive for us. “

It really depends on how much your archive is worth. The cost of entry is a bit high, but over it’s lifetime it’s not bad. It’s true, it’s not for everyone I guess.

[Wayne Carey] “That was big lesson learned.”

You didn’t make two? I make two LTOs right away, of the same data set. One’s onsite, one’s offsite.

FWIW, LTO4 tapes are supposed to last 30 years. By that time, all of your storage will be in the cloud or we will be retired on the beach 😉

[Hector Berrebi]

[Jeremy Garchow] “LTO is awesome.”

i know… from my clients and facilities i work with

i’m on the market for one… any recommendation? warnings?

thanks 🙂

hector

Hector Berrebi
Schibber Group
prePost Consulting

[Jeremy Garchow]

I have a Cache-A ProCache. I also had a PrimeCache, but stepped up to a ProCache.

Prime Cache worked great, ProCache has a bit more features and can be upgraded to LTO5 when need be (more storage per tape, faster read/write, raid 0/1 capable hard drives, many more direct connection options (p2/expresscard/etc)).

What is nice about the Cache-A is that it is pretty much plug and play. Hook it up to the network, sign in and you’re off. It is controlled through any web browser, and can be shared out to every computer in your network. It can also push/pull data from network shares as well. There’s a bit of a learning curve on how to handle the files, but it’s easy. If you can dag and drop, you can do it.

There is no other software needed, a simple and text based catalog is stored right on the unit for easy searching, although you might want your own cataloging system, and many support them such as CatDV.

Of course, there are cheaper ways to go about this, but I like the simplicity of Cache-A and they also understand the video business. Also, no extra software or extra hardware connection (SAS/SCSI/etc) is pretty nice. We didn’t need to take up precious PCI slots or dedicate a computer to run it. It is it’s own device and can be controlled/operated from any computer/device with a web browser. Not bad and worth a look.

Jeremy

[Hector Berrebi]

thanks 🙂

i’ll definitely check it out

hector

Hector Berrebi
Schibber Group
prePost Consulting

[David Jahns]

I think can shed a little light on this initial post…

(I’ve tried many different workflows with FCP & Red, XDCAM, P2, Canon 5D, – sometimes using L&T, sometimes using 3rd Party tools for the initial transcode, then importing the QTs into FCP.)

When using L&T, FCP does add additional metadata to the clip – it’s called a UUID (Universally Unique Identifier). Having this UUID tells FCP that this is a tapeless media source, and that it has been transcoded from some other original camera format to what you are now using for offline editing.

When capturing from tape, FCP adds the reel name & timecode info, of course, but does NOT add a UUID.

When you simply import a QT into the project, FCP treats this as if it were a master clip, (not a transcode for offline editing), even if it has a REEL assigned to it. It does NOT add the UUID to the clip.

When you try to recapture media for online, if there is a UUID with the clip, it will launch Log & Transfer – if not, it will launch Log & Capture (assuming it has a reel name).

It would certainly be nice if it gave you the choice of which tool to use, but alas, it does not.

Maybe in the next version of FCP? (How many times have I said over the years?? 🙂

David Jahns
Joint Editorial
Portland, OR

[Jeremy Garchow]

[David Jahns] “it’s called a UUID (Universally Unique Identifier). Having this UUID tells FCP that this is a tapeless media source, and that it has been transcoded from some other original camera format to what you are now using for offline editing.”

It’s a few more lines and a little more complicated than that. All clips either imported, tape captured, or L&T will have UUIDs assigned when importing.

Tapeless clips have info stored in them about the plugin and type of footage in the QT file itself. I can take a log and transferred QT clip, move it to a new machine and a brand new project, and if the log and transfer plugin works properly (meaning it’s not busted like Canon’s E1 plugin seems to be), FCP will know the origin of the clip and be able to recapture it using whatever the original method was.

If this clip was created by a third party or application that doesn’t understand this method, then the clip has no relationship to the camera original media and FCP just assumes it’s tape based as there’s no information telling it otherwise. Third party applications are able to write in the “capture origin” information, but they have to be smart as the application has to know a little about the camera original footage and be able to take that information and format it correctly for FCPs L&T method by writing in to an XML and sending it to FCP. It’s not easy business.

Jeremy