[Michael Angelo]

Hey Joe,

I appreciate you mentioning that proxy relink technique but it is a bit spicy for us with upwards of 2000 hours of footage, likely too cumbersome for us to manage. These challenges are definitely a FCPX 10.3 upward bug as the proxies disappear and reappear at random, often deciding to be stubborn when we are in delivery mode. Also we have been using the remote duplicate proxy media workflow for a years successfully with XMLs, metadata only media-less transfer libraries and straight up sneaker net hard drive to hard drive library transfers without copying proxies nor optimized. The master library finds the media on the destination drive and yes the different drive names mean the app is finicky about managed media but ultimately we’ve been able to make it work before 10.3

I don’t have the answers but my intuition is that it’s about waking up the hard links to the proxies. Our show is so big, we have a single library for each year and location for example NY14, SF14, LA15, LV12 ie New York 2014, San Francisco 2014, Los Angeles 2015, Las Vegas 2012. We melt material in these libraries, do some rough segment editing then move those sequences to a master library we call EDITS where we complete the final sequences. The EDITS library now has the bug that is causing the proxies to intermittently disappear.

In the past if proxies disappeared (which did happen here and there on a small scale) we would simply drag the clip from the original year location library, say SF14, to the EDITS library with copy proxies selected. This seemingly replaced the old dead proxy link with a proper hard link causing the missing proxies in the project timeline to be restored because we’ve looked inside package contents and no 2nd proxy clip was created. Now when we do this it doesn’t work consistently if at all and multicams rarely wake up. Re-transcoding would add 2TB to the EDITS library and adversely affect our remote drive workflow by losing the space saving hard links.

I’m sure we can improve our workflow over all but perhaps someone out there has an answer on how to reconnect hard links without having to re-transcode footage creating unnecessary duplicate proxy media?

As always thanks to everyone for the dialogue and assistance.

MA

Harvest the compromises…

[Michael Angelo]

Hi Joe,

Sounds very similar though we are still linked to the original/optimized media OK. If there was a way to relink proxies I think it would solve both our problems and since we have the proxies working in other libraries on the same drive we should be able to drag it over to wake up the missing proxies but it’s not working or stops working after quitting and relaunching the app.

Wondering how many other folks are seeing this same behavior in 10.3.2, the software we are using right now as well.

Chees

MA

Harvest the compromises…

[Michael Angelo]

Hmmm disk utility says everything is OK, unable to run “Verify Disk Permissions” assuming because it’s not the boot drive. We use the following settings as far as permissions:

administrator – Read & Write
staff – Read & Write
everyone – Read & Write
CHECKED – Ignore ownership on this volume

I’m thinking maybe the remote drive that the sequence was created on is the issue and corrupted the project? Got talked into buying a few Seagate Backup Plus drives as it was only 5TB portable at the time, nothing but troubles, they keep dying, we reformat, they die. Definitely gonna go back and stick with Lacie rugged in the future.

Any other ideas we are all ears, thanks again

MA

Harvest the compromises…

[Michael Angelo]

BTW share is usually sneaker net direct hard drive to hard drive via GUI libraries or XML. Sometimes we do the transfer library with no media but that often gets spicy and has unfavorable results

Harvest the compromises…

[Michael Angelo]

oh permissions! I should run disk utility (the old useful one), totally forgot about that, thank you as always!

The crazy thing is the proxies are fine in another library for the same clips and on the same drive, they should be linked via hard links. If proxies failed in the past usually you can just drag the needed clips over to the library that has the proxy issue, select copy proxy media and because it’s the same drive it simply creates hard links without taking extra space, then wakes up the proxies.

I could see how permissions could be a deal breaker, gonna look into it, thanks

Harvest the compromises…

[Michael Angelo]

Searching for markers in events still only pulls clips.

Whats worse is that not only does FCPX still not allow you to custom color code markers etc like FCP7 but Apple decided to remove ALL color coding so all icons in 10.3 are white. A maddening design choice to make things pretty but less functional.

Meanwhile we’ll keep trucking, hope color coding and customization comes back.

Harvest the compromises…

[Michael Angelo]

Thanks we’ll start looking into it

Cheers

MA

Harvest the compromises…

[Michael Angelo]

Hmmm, no Media Asset Manager other than our show tracker which we built by hand in google doc. We started small of course so now we just have a ton of 8TB GRAIDs:

– 1x8TB holds libraries and proxies with room to grow
– 3x8TB holds all the optimized media that we import into FCPX
– 2x8TB for all the original source material, graphics, animations, audio, research etc.

Do you think the 10.3 issue opening older large libraries could be resolved with some type of MAM? The idea of a MAM looks intriguing, googled and found this: https://www.keyflowpro.com/ We are pretty bummed as everything worked much better for us before FCPX 10.3 so your ideas are greatly appreciated.

Thanks

MA

Harvest the compromises…

[Michael Angelo]

Hmmm so that would make remote collaboration between proxy only libraries impossible without original media? The project is so big we have multiple libraries at the main office and for each editor remotely so they have access to all the source material. When needed we move things between libraries on the same drive, proxies only, creating sym links to save space. We just have to make sure that what ever we do remotely we do at the main office and vice a versa.

What’s so strange is that it’s an inconsistent challenge, works with some clips but not others, perhaps related to how they were originally imported. Seems like a big limitation especially as we are working with 2,000 hours of footage, would be pretty insane for each remote editor to have to have all the original optimized media just for libraries to communicate with each other (about 20TB)

The inconsistency makes me think perhaps there is a way to make this work, that and having done it for years with other application. Truly appreciate your input, time and energy!

Thanks

MA

Harvest the compromises…

[Michael Angelo]

Hi Noah,

Hmm good idea, just tested the clips in question with all media being available by making sure all drives were mounted, still having the same challenge. Here’s our workflow:

Since it is such a big show, we work with optimized media that we transcode ourselves (not FCPX) on external drives. We then bring this material in with the setting to create proxies only so the optimized material also becomes the original material as far as FCPX is concerned. This allows us to have just proxies, stills and audio clips living inside the library package and we can collaborate sharing “light” libraries.

Because of this we often have the external drives off when working from main office to work faster and lighter. Could that be causing the issue? Turned the external drives on with optimized media and things were still not able to drag between libraries. To make matters worse, ever since 10.3, our larger EDITS library locks up when ever we have the external drives with optimized media mounted! Because of this we had to break out the projects we absolutely need into smaller EDITS libraries to avoid this challenge.

Appreciate you reaching back, thanks so much

MA

Harvest the compromises…