[Bernhard G.]

Hello Hector,

Thank You, this is really interesting.
I think psychology is the main criterion at many decision makers.

To be honest, I believe for much of professional work FCP-X could be sufficient,
but the next generation of editors needs to (dis)prove this, since also the
general context of media production changes these days.

Psychology also caused the marked dominance of FCP – it was the
feeling to use exact the same tools like hollywood professionals
(which was also mere PR than fact).

Best regards,
Bernhard

[Bernhard G.]

Hello,

the BBC has developed a professional codec and took care
that it’s components aren’t bound to licenses:

Dirac Pro

It utilizes wavelet-compresson like JPEG2000 and Cineform
and was developed to be robust for post production.

There is even hardware, ready to use:

https://www.bbc.co.uk/rd/projects/dirac/
https://www.bbc.co.uk/rd/projects/dirac/diracpro.shtml

I seems the development has been finished,
but I don’t know why it is not commonly used, e.g. by Adobe.

What I very appreciate about ProRes is that it caused a paradigm shift
in who defines the recording codec in field.

In my opinion, camera vendors should give up codec development at all!

In the 21st century there is simply no demand for e.g.
an overpriced format like HDCamSRLite inside a strange container
passed over an inpractical interface any more!

Best regards,
Bernhard

[Bernhard G.]

Hello,

Digital Cinema is and was always RGB.
Also PremierePro, Motion (also the old one!!!) and AfterFX uses RGB rendering.
And Smoke works only in uncompressed 10bit RGB at all!

AJA needed to implement a WYSIWYG (WhatYouSeeIsWhatYouGet) method
to get video properly out of PP. The same with Matrox. So RGB is not that odd.

When feeding our hp dreamcolor we send out a HDSDI signal to a BMD HDLink Display Port,
to convert the YCbCr (not YUV bdw) signal into high-quality RGB.

Nothing here convinced me there will be no possibility for proper high-quality video output in FCP-X!
And if Apple tends to introduce also a new paradigm to video monitoring at delivering
high quality DisplayPort signals up to 16bit color deph per channel, than it’s more than ok.
https://en.wikipedia.org/wiki/DisplayPort

Personally I would expect a very high quality output class delivered with AVFoundation
with the lion update.

Best regards,
Bernhard

[Bernhard G.]

Thanks for the tipp!

Yes, I need to downscale primary for DVD production.

For web and BluRay I use Matrox MAX which uses the Matrox HW-scaler and delivers excellent quality,
but only outputs progressive H.264 files; therefor unusable for further DVD encoding
(for DVD encoding I use CinemaCraft CCEmp with usually 30 passes – so I really need the maximum SD quality I can get)

BitVice does indeed have one of the best SW-scalers and it’s the same as VideoPurifier.
My tests have shown it is nearly indistinguishable from Final Cut Pro’s scaling with nesting HD into a SD timeline with motion estimation set to best. I believe Apple and Innobit are using 100% the same method (if not the same algorithm)! But the images are blur compared to a HW-scaler.
Good results I also got from MPEG Streamclip, but it seems to process 4:2:0 only at 8bit.

My hope is that Matrox expands MAX to output uncompressed 10bit interlaced,
or that AJA delivers a small application for file-based scaling.
I’m also interested in the CUDA scaling of Premiere CS5, but I don’t have access to an appropriate CUDA card for testing yet. PP uses Lanczos2 low-pass filtering with bicubic resampling – in theory an absolutely excellent quality!
(CS5 uses SW only a different algorithm than with CUDA and the results suffer the same problems as any other SW scaling)

Currently there is only one solution for arbitrary file based HW-processing out there:
https://www.teranex.com/products/vc400
And I don’t want to know the price …

Best regards,
Bernhard

[Bernhard G.]

Thanks for listening 😉

And absolutely agreed!
Backup is a huge problem.
And some steps in the chain of file-based workflow still aren’t sufficient yet.

e.g. Scaling HD -> SD:
My current workaround for scaling is to put and playback my ProResHQ HD masters with KiPro,
and recapture them with my MXO2 in SD.
I HATE THIS WORKAROUND! – but I need to use it to get acceptable results.
Unbelievable but true – there’s not a single software capable in good scaling on the market.

Best regards,
Bernhard

[Bernhard G.]

You’re welcome!

I was in the same situation years ago.
My personal tip is not to ask any video-guy
(ironically I’m a video-guy myself 😉 ) for things like this.

Ask instead a true expert on network storage for whom videos are nothing special
but data like any other.
Your part is to describe him/her exactly the demands of your workflow.

(I believe video guys generally tend to be too conservative to benefit consequently
from latest IT-technologies;
or why else would someone e.g. pay €70.000 for a HDCam-SR Videorecorder –
A VIDEORECORDER !?!?!?!
in the 21st century – instead of getting a modern IT solution like a KiPro… )

Best regards,
Bernhard

P.S.: Sorry for propaganda – but you’re listening 😉

[Bernhard G.]

Hello,

in the past M100 had a “user-permission” error at soft-importing
video files from an ethernet storage. But this seems to be solved since M100 v1.6!

– For max performance I would recommend the most expensive Cat7 cabling.
– If you have a cheap ethernet switch then throw it out of the window!!!
I made good experience with hp procurve 2910 and 1810G; for up to 6 users
the ProCurve 1810G with its 8 ports is very affordable and FAST!
– You need to use bonding/link-aggregation
(2 ethernet-connections acting as one) between storage and switch
– Use jumbo frames
– Use RAID5 at your NAS
– Only use AFP to connect your OSX workstations
– keep internet and other nw-connections on
a separate network and switch!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
– personally I would prefer a NAS with Intel Atom processor like a Qnap TurboNAS +
(“+” means the version with the even faster CPU)
but this depends on how many workstations should be connected and how many HDDs
the NAS has to manage;
If you need to connect 2 workstations, then 4 HDDs in the iomega might be fine.
If not, I would reconsider my choice!

If the setup is correct then this configuration will work up to ProResHQ 1080i and even better with DVCProHD!

Best Regards

[Bernhard G.]

Hello,

scaling is always a compromise.
A ‘good’ downscaling is not defined as ‘the sharpest image’.
A ‘good’ downscaling is defined as the maximum sharpness at the minimum aliasing.

So every hardware scaler out there doesn’t simply have a good scaling algorithm,
but performs tons of pre processing – take a look at this:
https://www.teranex.com/technology

There is not a single software scaler out there which could compete with hardware scalers.
I’ve tested them ALL. Either you get too smooth images every SD-cam would have delivered better,
or you get very sharp images with heavy aliasing so you get sick.

The best SW-job does MPEG Streamclip, but it is 8-bit only and the 8pxl wide black bars
need to be removed by cropping at scaling.
Purifier witch uses the same scaler as BitVice is indistinguishable from FCPs scaling (Motion Estimation set to ‘best’) even at pixel level. As if both were using 100% the same algorithm …

If you need to scale for web, I recommend Matrox MAX (as ComporessHD board or in a MXO2).
It uses Matrox’ hardware scaler file based!!!
Unfortunately it only outputs progressive H.264 (8bit) at the moment (but with HW-deinterlacing!).
This is perfect for web or BluRay, but not usable for further workflow processing.

It’s really too bad we can’t use the hardware scalers of our I/O devices file-based yet!
It’s a shame – the perfect solution is so near …

[Bernhard G.]

Hello,

I’m 100% with argument 4.)

As seen now, Apple seems to dominate the market with ProRes.
Every major software and hardware is advertised being compatible with ProRes.

In my personal opinion Adobe should had overtaken the company ‘Cineform’ AT ANY PRICE
when PP was released for OSX years ago and placing it in the market very very aggressively.

Maybe the strategy using native codecs was right on a long term. Maybe not.
I notice the tendency that camera manufacturers will loose their monopoly
of defining recording codecs to software companies and 3rd party hardware vendors –
see AJA KiPRO, Arri Alexa, Atomos Ninja, etc.

And who dominates the codecs that are used in the field,
determines with software is used afterwards – see Apple.

Only my personal opinion.

Bernhard

[Bernhard G.]

Forgot to mention Purifier uses the same scaling as BitVice 😉