[Bernhard G.]

Hello,

please don’t forget to write a Feature Request to Apple!
Discussing here is nice but a FR could actually help.

I also hope Apple will license the open source H.265 HEVC
successor of x264; (it’s not named x265 – no relation to x264)

Best regards,
Bernhard

[Bernhard G.]

[Angelo Lorenzo] “With that said, Adobe is TERRIBLE at documenting the nuances of export.”

Hello,

what I’ve heard, Adobe is listening!

So please write Feature Request!

I’d also appreciate more consistency and clearness in terms of
professional quality concerns, e.g. 100% 32bit float processing and MRQ,
more control how a clip conforms to a sequence, etc.

Best regards,
Bernhard

[Bernhard G.]

Hello,

I absolutely agree applying the ‘Fit so Sequence’ function on all clips;
it could also be turned to default in the preferences.

I would find it helpful, if this was standard, but I would also appreciate
if there was an additional fix assigned category (beside Motion, Opacity, Retime)
in the Effect Controls window, labeled ‘Standards Conform’ or ‘Clip Format’.

Therein I would like to see options for Fit to Sequence, De-Interlacing, Frame-Blending, Color Space, etc.; everything with better control and more professional options like motion analysis;
simply everything about the behavior of a clip to match the sequence.

Wrote this feature request to Adobe. Now hoping for CS7…

Best regards,
Bernhard

[Bernhard G.]

Hello,

I’m a fan of completeness, so being forced to use pug-ins for basic tasks is nothing I appreciate.
The more plugins I need for 90% of my daily tasks, the less attractive an app is to me.

Comparing FCP-X to PP CS6, I have to say that at the moment (!)
I feel Apple has got their homework done:

– FCP-X processes 100% at 32bit float RGB color precision

– FCP-X does optical flow analysis for speed changes
(thought speed ramps can’t be adjusted via bezier splines)

– FCP-X does optical flow analysis for deinterlacing
(at least when doubling frame rate; otherwise compressor does)

– FCP-X stores analysis data into a well structured media management

– FCP-X correctly transforms color space between HD and SD

– FCP-X has integrated color grading (thought it lacks motion tracking and spline masking)

– FCP-X automatically synchronizes clips via audio

– in FCP-X I don’t need to care nor hack my app to get GPU acceleration

– in FCP-X I choose the video standard (professionally…)

while

– PP has an inconsistent quality behavior in terms of 32bit float compatibility of filters/effects

– PP has an inconsistent quality behavior with GPU rendering versus CPU rendering
(different algorithms; I expect pixel for pixel 100% the same value, regardless if I use GPU or CPU)

– speed changes are using frame blending instead of optical flow analysis
(aka motion analysis aka motion compensation)

– Adobe’s understanding of de-interlacing is to throw away the half of my precious image… 😉

– CS6 stores analysis data (image stabilization) into the project file, resulting in hundreds of MB

– in CS6 I have to hack my app to get GPU acceleration. Why?
e.g. Resolve informs the user that performance might not be the best.
The same model of self-responisbility of the user would also work for Adobe.

– in CS6, thought codec-independent (which I appreciate much)
I still have to choose the codec instead of the video standard,
as if Adobe assumes I don’t know which standard I’m working on… 🙂

At the other hand, CS6 offers bells and whistles feature like the absolutely great image stabilization,
or Audition’s automatization for lib-synchronization. But how often a week do I need to lib-synchro a movie,
and how often a week do I need to synchronize two or more clips per audio?

So the only thing I really want to respectfully complain about PP CS6 is it’s contradiction between
absolutely great niche features while having a fundamental lack of some very, very basics.

But Adobe is listening, so this could change quickly…
Therefor don’t forget to write feature requests to Adobe!

Best regards,
Bernhard

P.S.: Some might argue, de-interlacing isn’t a basic task anymore.
But for me it has become an indicator for how serious PP’s take on professional video production really is.

[Bernhard G.]

Hello,

from a cold analytical point of view I think a drawback for M100 is,
that it relies on Quicktime architecture (doesn’t Boris RED either?).
Apple decided to EOL Quicktime architecture and replacing it with A/V Foundation.

So there are two options for Artel to cope with this circumstances:

Option#1: adopt M100 to A/V Foundation, which requires TIME; a lot of it.
[means: follow Apple wherever they are heading for and benefit from it; just like in the past]

Option#2: develop it’s own media architecture.
[means: a lot of work under the hood OR much license fee for 3rd party components]

Another issue is that now FCP-X fills the market of an easy to use but still professional application.

There are also two options for Artel to cope with this situation:

Option#1: make M100 easier than FCP-X
[can’t imagine how]

Option#2: make M100 more powerful than FCP-X
[Boris filters and effects are some of the very best. Using them in a unified, well designed editing interface that basically assembles a finishing app – *cough* *844/X* *cough* – would be a powerful combination.
And sorry, NO! Boris RED’s interface is the opposite of well designed]

Furthermore I’m the opinion that in M100’s long history two fatal mistakes were done:

Mistake#1: EOL 844/X and reactivate the Media100 app instead of
getting rid of hardware development and push development of 844/X.

Mistake#2: restrict video-I/O to AJA Kona boards. !NOTHING against AJA nor Kona! Great products!
But this decision simply was a shot in the knee! Lightyears way off-road the general trends in the market.
Simply a STUPIDITY!!! Sorry for this hard word. It is not that analytical.

Nevertheless I don’t see M100 to be dead – yet.
At next NAB we will need to see how Artel copes with the new situation
an evaluate again.

But while Artel copes with the current situation today, Adobe prepares Adobe Anywhere for the market of tomorrow and I would expect Apple to bring something similar, making us editing FCP-X on iPads while
heavy processing is done elsewhere…

Best regards,
Bernhard

[Bernhard G.]

Hello,

currently, PremierePro does Smart Rendering
as FCP7/X or MediaComposer does, only for a handful of formats
which means, only those portions of the sequence are rendered
that need to be in case, the sequence-codec/format and the clip codec/format are the very same.
See here:
https://blogs.adobe.com/kevinmonahan/2012/10/11/smart-rendering-in-premiere-pro-cs6-6-0-1-and-later/

When exporting via AME this means, every time You output the sequence in PremierePro, 100% of the images are processed during export again.
If You export a sequence into 3 different
delivery formats, the whole sequence is rendered 3 times.

Therefor, to render a sequence doesn’t make much sense.

You COULD set the preview codec to e.g. ProResHQ
and check ‘use preview files’ in AME.
BUT: this would mean, if the render-files are ProResHQ and You export the sequence into a ProResHQ master, the ProResHQ render files are decoded and re-encoded into the very same format again…
So: DON’T DO THAT!

When You need to export into several formats:
PremierePro’s rendering workflow delivers a better quality
but it also takes longer.
(But as long as PP’s image processing pipeline is not 100% 32bit-float,
combined with CPU-CUDA differences in quality, this advantage is very theoretical…)

One thing I don’t understand:
Why does Adobe offer so many different sequence presets?

Since PremierePro’s sequences are codec-independent,
there seems to be no technical reason for that!
It appears to be a help for those who don’t know in which
video standard they are working…

Best regards,
Bernhard

[Bernhard G.]

Hello,

I hope for

– a colaborative workflow:

the disk image workaround for network storage being standardized
and configurable from within FCP-X; giving the new FCP Disk Images a nice icon 😉

those FCP Disk Images to have rudimentary access permission handling comparable
to SNS SANmp: Read/Write (others can read); Write Exclusive (others can’t read); Read Only (for the others)

FCP Disk Images to be also the new standard procedure on local storage

all ProApps to access Events

– Compressor X:

with high-end scaling algorithms on GPU to compete with dedicated hardwares like Kona3

furthermore a native x264 implementation (but as far as there is no sophisticated scaling algorithm in
Compressor, x264 would make no sense to me…)

– FCP-X to get:

tracking for CC (by optical flow analysis; means without setting tracking points per hand);

Bezier and BSpline mattes; separate feathering masks;

temporal adaptive Noise Reduction analysis that could be set in import dialogue;

again better scaling algorithms with Super Resolution technology for up-rezing footage
(similar technology to image stabilization; details are gained from several images);

better keyframe handling with bezier splines

stereo 3D 😉

I’m aware that some of these features are available via plugin – but the more plugins an
other apps I need to purchase, the less attractive FCP-X gets with it’s long-term promise
of ‘completeness’.

Best regards,
Bernhard

[Bernhard G.]

Hello,

the role of consumer demands and innovations is scientifically well researched.

FCP-X is what Clayton M. Christensen has called a Disruptive Innovation.

See his book The Innovator’s Dilemma.

A Disruptive Innovation not only introduces an innovation –
every successful company need to do so – but also shifts
the valuation criteria of the market itself.

One interesting point of Disruptive Innovations is, that at their introduction
they can’t compete with established premium products, but within a niche they
evolve much more faster that the established competition that is already over-engineered.

I would recommend to read this book to everyone who wonders about Apple’s product strategy – for Steve Jobs it was one of the most influential books he ever read!

Clayton M. Christensen: The Innovator’s Dilemma

Best regards,
Bernhard

[Bernhard G.]

Hello,

I feed my Dreamcolor with the HDLinkPro3D;
what additional format options does HDP2 offer?

Best regards,
Bernhard

[Bernhard G.]

What’s about a

HP Dreamcolor 2480zx
+ BMD HDLinkPro3D

both together under USD 3000.-
and 10bit calibrated.

Best regards,
Bernhard