We're starting to look into CatDV having never previously made use of a DAM system before.
I can definitely see the benefits of such a system and it seems it would make our lives so much easier when searching for footage and such.
In terms of workflow however, is it the norm to set up one CatDV directory for all the footage you have? - ideally cataloguing everything, or is it best to set up individual directories per project, client and so on?
I realise there may not be a strict 'norm' for this kind of thing - in terms of background, we're often having to re-use footage for certain clients and in this case, I can see why one large directory may help to make sense of the scale of the media.
I'm worried that having a large directory may cause problems, but I guess CatDV was designed to handle large amounts of media...
I'd love to hear what you think,
the quick way to get a handle on this is to search this forum I have written and others as to best practice in this regard many many times. So check out the group wisdom it should speed up your understanding in this regard significantly.
Agreeing with Kevin and also, engage a reseller/integrator as soon as you can.
DAM is about so much more than software.
Quick Answer: If you need the footage later, catalog it and archive it. If you don't, discard it.
How you do that, is an insanely deep subject. Note that the pros on this forum regularly engage for weeks and months with our clients on these projects before a single bit of software is purchased or installed.
There are several CatDV users with millions of clips and tons with hundreds of thousands of clips, so no matter your size, there is a solution for you.
bryson "at" hidefcowboy.com
Yeah, I totally understand that proper DAM, creating a solid workflow and systems to support it definitely takes time.
We have over 20-odd years worth of footage that at some time or another has been needed to be re-ingested and such. We've become a sort of video library for one of our clients - which both means regular work and the headache of storing it all.
We have an internal cataloguing system which is okay...it's just a little antiquated in a modern world, especially when the client asks for 'images of the ocean' and you have to take a stab in the dark for the relevant tape / carded media.
The prospect of CatDV helping us to carry metadata throughout the production and editing process is something that I think will help to push the 'modern organisation of things' that I'm trying to achieve.
I'm getting into contact with a reseller in order to help us with workflow ideas - this is something I want to do right, I definitely don't plan to rush out and buy the software in the hope it will solve everything without proper planning!
Thanks for the help!
I've been bringing in all our old footage and creating a library in two levels. First all the footage comes in based on the shooting location. This localizes the footage at least to an identifiable area: a business, school or hospital, something alone those lines. We do the same for accounting purposes as well. Shooting is "research." A project (which must be amortized in our little world) doesn't come into being until we actually sit down and design it (based on our 'research') and start to edit. So now I have list of places or areas where we've shot lots of different things. Footage comes in as ProRes422 and I'm storing it on raw external drives for later use. Maybe I get more than one location on a drive, and CatDV keeps track. I'm making proxies of everything at low rez. Sometimes that means M-Jpeg, but lately ProRes Proxy (with the file name and timecode burned in). CatDV proxies are stored on their own drive on the network.
Using Workgroup Server we can do scans and queries across all the footage. Primarily we're using the proxies and doing initial edits with the proxies. I can make copies and consolidate footage onto a machine if I want. Eventually I move closer to a finishing stage — dumping the footage from the timeline and making it "offline." I move the full rez footage from the raw drives onto a machine, and conform it. I have to do a lot of mixing and matching of footage no matter what I do. In the past, in the "traditional" non-linear methodology I was doing a lot of loading, and reloading from HDCam and DigiBeta tapes. I've reduced the time devoted to that part of the workflow a great deal by making my own files as part of my logging/archiving/library process. There are still kinks here and there, but it's working for us. The best part is having huge amounts of source footage available all the time in the form of the proxies. I love being able to hunt up an extra shot in just a couple of minutes and then simply drag and drop it into my timeline. Plus some of our people are capable of creating "sequences" in CatDV and they arrive directly on my desktop where I can open them fully populated. A lot better than trying to read poorly scrawled lists of shots and numbers and spending a couple of days loading them just to start.
Non-linear: all the time and nothing but.
Would you mind linking to one of these best practice outlines? I can't seem to find it, and we are considering implementing CatDV, but I am having a hard time figuring out how to get started with it. I have seen lots of forums and videos pointing out features, but few "tutorial" type instructions.
I'll keep looking in the meantime! Thanks!
Canon 5d MKII
Ben, if you search for catalog size or anything related to the catalog structure, it's there. I linked to a couple the other day.
Let me run this down again here as well.
There is not a lot of documentation for CatDV that is specific to workflows. The reason for this is that Squarebox makes a tool, not a workflow. If you buy a shovel, there is not a manual as they don't know if you are shoveling sh%t or dirt or grain and they don't really mind what you do with it and in each case, the instructions will vary.
Folks like Kevin, JB&A and our company come in, look at what you want to shovel (to continue the metaphor) and lay out a process to get it all into the bin or grain silo or dump, depending on what you are shoveling. We also tell you whether you should get any in your mouth or not. ;)
The reason the pros here are sometimes vague is that we don't want you to go and base thousands of hours of logging or transcoding work on a premise that can't possibly account for all your variables. Storage, use-case, folder structure, server OS, codecs, staff count, deadlines/timelines, editing system, delivery platforms, web infrastructure... I can go on but you get it.
Short answer... and only my opinion, put no more than 1,000 items in a catalog. Some clients put less than 100. I have catalogs currently on systems that have less than 5 clips and some that have over 4,000. All are correct... and all are wrong.
The reasons I say those numbers are based on a 15 minute training/consultation conversation and demonstration that I have with clients. I learned it from Kevin, mostly, and have added to it over the years with my experience. That, added with a few hours of consultation over the course of a few weeks helps us learn your facility and thus deliver a "best-practice" for you. The number we give you takes into account CatDV's structure, the client's infrastructure, the level of sophistication with asset management concepts of the staff and even the depth of metadata, amount of thumbnails and media format.
If you search you will see that others dispute these numbers and we all have our own reasons.
Final thought, according to Squarebox, any number is "correct" it is workflow dependent and that is maddening to a client on the outside and easy to determine for a client who has the software running in their facility.
Engage with someone to evaluate your needs, plan your integration and maybe even give you some training before you even buy a single item. (It's ok to pay them for that too.) That will protect you from spending your company's money on something that may not work in your workflow. I say this not to grow my business but to protect yours.
bryson "at" northshoreautomation.com