Hi all,

I have been following the CD threads in this forum with an eye toward purchasing a nanoFlash. Hopefully I'm not making the noob mistake going over old ground.

First, am VERY impressed with Mike and Dan's level of communication on this forum and you are to be congratulated for partnering with your customers in such an outstanding way.

A couple questions:
- Does the XDCAM HD422 codec that you use carry metadata? I want to use ProxyMill from Imagine Products to create proxies with timecode but Proxymill needs metadata to do it. Can do?

- How far over the horizon are 64GB 45MB/s cards. Although you say 100MB L-GOP is the "sweet spot", I have never been a fan of L-GOP editing (although your arguments are compelling)... and may go with i-Frame at 160 or higher. The need for larger storage is obvious. I have seen 64GB 60MB/s x400x cards on the Sandisk Business Solutions site: Compact Flash (http://www.sandisk.com/business-solutions/flash-memory-cards/compact-flash-cards.aspx) ... but are they a viable alternative or just a fantasy?

And I have a product suggestion...
- Have you considered consolidating your contiguous files with Reference Movies that are generated by the nanoFlash. FireStore has this feature and it is a great one... cutting down on bin clutter and giving the editor logical files to work with, particularly with long interviews. The FireStore implementation is that, just prior to dumping the disk, you run a utility that creates the Ref Movie and then it downlods with everything else. Possible?

This looks like a terrific product and I am excited by the possibilities of working with it!

Thanks,
Bob
FireDancer Productions, Inc.
http://www.FireDancer.tv

Dear Bob,

Thanks for the kind words. We feel that open communication helps us all.

We have been working on Metadata. We will need to know what specific metadata ProxyMill needs.

We have learned of the SanDisk 64 GB CompactFlash card. As you say, it is posted on their website, but it does not appear to be ready to ship. We are checking it out.

Other 64 GB CompactFlash cards have been announced. We have checked out one, and it was not acceptable, so allow us to fully qualify them before you purchase any.

We are enthused about the 64 GB CompactFlash cards from SanDisk.

Bob, I was not familar with Reference Movies. We will check into this. Thanks for the heads up!

Hi Dan,

Here's a link to the FireStore manual on Reference files:
http://www.videonics.com/dynassets/documents/products/FS-100v4.0_UserGuide(MANL-1023-10).pdf . The section on Reference movies is on page 78.

Pretty much what I figured on the 64GB cards... such is technology. I'll wait for your OK.

As for the metadata, I have no idea what ProxyMill requires but I have sent a note off to a support guy there that I have traded a few emails with. The best deal is to probably connect you guys so you can talk tech. If he agrees, how do you want to get in touch with him? I have his email.

Thanks,
Bob
FireDancer Productions, Inc.

Dear Bob,

You could just send me an email with his email address.

If you left click on my name to the left of this post, you can send me an email.

ProxyMill works with .MOV files from the NanoFlash. I have it and it works with my 1920x1080 HD clips without issue. It doesn't work with NanoFlash MXF's

ProxyMill works with .MOV files from the NanoFlash. I have it and it works with my 1920x1080 HD clips without issue.

Hi Alister,

Been following your exploits with the nanoFlash... thanks for all the great reporting.

My tests with the Moto2.mov file from the CD site shows that that clip does not supply TC to ProxyMill. In QT, the Moto2.mov file starts at 00:00:19:24 NDF. When I ran it through ProxyMill and I got a 00:00:00:00 starting TC. This was done at the "Large" compression setting, no vizcode. Tried a few others too... no luck. Dan K. had suggest that i do this test so I assumed that the file has TC on it.

I really want you to be right about this! Any ideas? Thanks!

Dear Bob,

The motorcycle clips were shot with a POV camera.

As it was a single camera shoot, I do not think timecode was recorded.

The nanoFlash and Flash XDR have the ability to generate timecode internally.

But, unless it was used (and it may not have been available when this was shot), then the timecode will be 0:0:0;0.

We need to get you better test files!

And, we are working with ProxyMill to ensure that our timecode is visible to their software.

But, unless it was used (and it may not have been available when this was shot), then the timecode will be 0:0:0;0.

The 00:00:00;00 code was what I saw on the moto v1.mov file. If anyone has a short clip that they know has TC, I would love to see it.

And, we are working with ProxyMill to ensure that our timecode is visible to their software.

I saw the exchange between Dan Montgomery & Tommy. Cool... please let me know how this shakes out.

Thanks!

Here are some newer samples to try out (I'm not sure how the timecode was handled in those other files) :

(20 Mbytes each)
http://69.15.88.17/downloads/qt_web/new_sample_50i.MOV.zip

http://69.15.88.17/downloads/mxf_web/new_sample_50i.MXF.zip


Tommy

Here are some newer samples to try out, Tommy

I downloaded both but I don't have the mxf player. I processed the .mov and the file worked like a charm! The 22.8MB file compressed all the way down to 40k using the lowest settings in ProxyMill... and kept the TC! Woo-hoo!

So why all the fuss? Here's the workflow plan: I shoot roughly 30 minute interviews in the field with the nanoFlash. I import with ShotPut Pro and ProxyMill. I then send these miniscule compressed files to a writer I work with who also has FCP. Rather than working from text transcripts, he prefers to edit directly in the timeline chopping the interview bites together. He imports them to FCP, cuts these interview files down into the baseline of a story, and sends me an audio file. We go a couple rounds agreeing on the final version, and then he sends me a FCP project that I reconnect to the original media. I add, b-roll, music, etc. and... Voila!... done.

Thanks to Tommy sending these files... and them working... we have now completed testing on this workflow and will use it on the next job.

So Alister, you're entitled to a big "... I told you so!..." and thanks to Tommy, Dan, Mike at CD and Dan at Imagine Products for all their help!

I'm with you Bob, sometime you just have to see it work for yourself! Glad it's all sorted.

The files sizes are working out larger than we hoped because those tiny 40k files are using the AAC encoder for audio... which won't play in real time in FCP without a render. Makes it hard to cut up a file if you constantly have to re-render.

The smallest I could get that file to is 304k using mp4 at 15fps and PCM audio. The better audio accounts for the difference... but it plays in real time.

You wouldn't know of an audio codec that will play in real time in FCP that will offer higher compression rates, would you?

Thanks!