-
iMac Pro or next year’s Mac Pro?
-
[Oliver Peters] “But if my requirement is an MPEG2 or Avid DNxHD in an MXF wrapper, I don’t believe the service will generate that from the ProRes master. Is your experience different?”
It’s the party requesting that deliverable I’m questioning.
If Telestream Episode can fix this on your desktop – shouldn’t a virtual Episode-like system be able to fix it better, faster and more directly in the cloud?
Why should it be your job? It’s “automatable.” Right now.
Asking EVERY editor to become a transcoding expert is silly.
It’s work better suited to the machines.
My 2 cents.
Creator of XinTwo – https://www.xintwo.com
The shortest path to FCP X mastery. -
[Bill Davis] “It’s the party requesting that deliverable I’m questioning.
If Telestream Episode can fix this on your desktop – shouldn’t a virtual Episode-like system be able to fix it better, faster and more directly in the cloud?
Why should it be your job? It’s “automatable.” Right now.
Asking EVERY editor to become a transcoding expert is silly.”Sure, I completely agree. But, that’s not living in the real world. I can barely get clients to give me the real specs of what they want/need. Therefore, expecting them to do it correctly is a real long shot. And that includes a lot of broadcasters/networks/studios. The bottom line is that what you envision doesn’t currently exits via Vimeo, Frame.io, Wipster.io, etc.
– Oliver
Oliver Peters – oliverpeters.com
-
[Bill Davis] “Check again.”
Will do. Thanks Bill.
[Bill Davis] “If Telestream Episode can fix this on your desktop – shouldn’t a virtual Episode-like system be able to fix it better, faster and more directly in the cloud?
Why should it be your job? It’s “automatable.” Right now.
Asking EVERY editor to become a transcoding expert is silly.”
But what file do you send to Telestream-in-the-cloud to begin with? That’s the crux of this. In this scenario the editor isn’t tasked with creating a bunch of different transcodes, just one file out of the NLE that will then be sent to the client so they can generate a bunch of different transcodes.
If the client wants a 23.976 4K ProRes 4444 and I export my ‘master’ as 720p60 MPEG2 then upload it to Telestream so it can make the 4K ProRes file the client’s probably not going to be too happy with the results. Unless clients start supplying export presets that can be imported into NLEs (or Compressor/AME) then editors will still need to have enough technical knowledge to generate a master file out of their NLE that adheres to client specifications (or pay someone w/that knowledge).
-
Diminished gains to be sure. My interest is in seeing what the gains/benefit are of the eGPU/AMD/$1000 kludge vs. $5-$10K iMac Pro.
We’ll need the new FCPX with official eGPU support on the next High Sierra version to really see if there’s any value there or not.
-
[Bill Davis] “This is what we’ve done in the commercials area for a years now. Master to Extreme Reach or equivalent – they do the rest”
Extreme Reach does not supply the same services as Vimeo or Frame. They are an actual spot delivery service. In addition, they dictate to the broadcaster a very limited number of formats and broadcasters either accept the service or don’t. That’s not the same thing as sending a 23.976fps ProRes file and having 60i (NTSC) and 50i (PAL), with and without captions, supplied by the service, based on 100 different end user specs.
So, yes, in theory your concept could work, IF and only IF, the end users’ specs are standardized and limited. Which they aren’t.
– Oliver
Oliver Peters – oliverpeters.com
-
[Andrew Kimery] “But what file do you send to Telestream-in-the-cloud to begin with? “
I see this as a temporary issue that will be solved pretty quickly.
The clear trend is two-fold.
The first is metadata instructions against mirrored footage pools. Both X and Premiere have gone there via their Proxy plumbing. It’s trivial and expected now.
That sets up the key factor as the nature of the Original Media pool – wherever that may reside. Right now I have the most experience with how Frame.io handles this – but basically, I work initially with h.264 encodes for ease in the early stage content creation and review. Then I upload full Rez out of X – which just replaces my source pool.
But “full res” can vary, at will, as well.
The “pixel density” of that source media pool is determined by me when I determine my Project settings. But that can easily be revised later.
The clear trend is that cameras are capturing both larger pixel rasters AND improving their color depth capture abilities rapidly – so this seems like a process built for an easily foreseeable future.
So the simple answer is that you send to your distribution service a “digital master” created from the largest, pixel densest, least destructively encoded footage pool you can – and then let the big processing guns in the cloud handle the rest.
FWIW.
Creator of XinTwo – https://www.xintwo.com
The shortest path to FCP X mastery. -
[Bill Davis] “The clear trend is that cameras are capturing both larger pixel rasters AND improving their color depth capture abilities rapidly – so this seems like a process built for an easily foreseeable future. “
I agree that the trend towards higher spatial resolutions is trending upward, and I expect 6k and 8k capture devices for consumers to be ‘the thing’ by 2025 (maybe), but I don’t expect that highly compressed, in-camera, 8 bit capture from mainstream camera manufacturers to go away anytime soon… a lot people just don’t see the difference between 8bit and 12bit images… including (unfortunately) a lot of professional media creators.
Shawn
-
[Bill Davis] “So the simple answer is that you send to your distribution service a “digital master” created from the largest, pixel densest, least destructively encoded footage pool you can – and then let the big processing guns in the cloud handle the rest. “
Nice in theory. Too bad it’s only theoretical for the near-term, unless you can work within very narrow guidelines.
– Oliver
Oliver Peters – oliverpeters.com
-
[Shawn Miller] ” I expect 6k and 8k capture devices for consumers to be ‘the thing’ by 2025 (maybe)”
A big maybe. So far there’s no viable argument for any resolution higher than 4k in regards to image quality and the human eye being able to detect a difference and even 4k is a long way from being adopted as the default broadcast format.
I Hate Television. I Hate It As Much As Peanuts. But I Can’t Stop Eating Peanuts.
– Orson Welles -
[Bill Davis] “So the simple answer is that you send to your distribution service a “digital master” created from the largest, pixel densest, least destructively encoded footage pool you can – and then let the big processing guns in the cloud handle the rest. “
For :30 spots this seems viable (aside from Oliver’s concerns). For long form content you are talking about uploads that can span the range from hundreds of gigs to over a TB depending on the settings (ex. a 90min 23.98 UHD movie in ProRes XQ is 1.2 TB). Most people will need faster Internet pipes for that. ????
Of course the genesis of this branch of the thread wasn’t that file delivery is inherently complicated, it’s that FCPX/Compressor currently can’t export a common format w/closed captions like the other major NLEs can which makes Don’s current workflow unnecessarily complicated.
I was half kidding when I said it, but I’m digging my idea about companies sending export presets for editors to use. Obviously this wouldn’t be viable in all cases (like Don’s or where the client doesn’t have the tech knowhow to do it), but if Netflix can write up a 53 page deliverables guide I’m sure they can make some export presets to go with it. 😉
Reply to this Discussion! Login or Sign Up