Lost 2 Rentals This Week!

[Mike Prevette]

If you can't get an hour of red footage Best lighted in 18hours something is VERY wrong with your set up. On my macbook laptop I can bestlight DV dailies in a 2:1 ratio.

 

[David Battistella]

If you can't get an hour of red footage Best lighted in 18hours something is VERY wrong with your set up. On my macbook laptop I can bestlight DV dailies in a 2:1 ratio.

Share?

David

 

[M Most]

Share?

David

There's no big secret. Most relatively new Macs, particularly the 8 core models, can process at standard def with a standard debayer faster than real time using Redcine. The 8 core machines will process about an hour's worth of footage in about 45 minutes.

 

[Lucas Wilson]

I can get an hour's film to the lab, processed and best light in less than 18 hours. I can't do that with my Red files... on the fastest mac.

Now here is a workflow that makes a lot of sense:

1) Okay... here is a shooting medium... it's incredible, but... don't expose it to light or the whole thing is ruined.

2) Now that you have taken great care to not expose it to any light, get it to "the lab." Once at the lab, take this shooting material out of its case, dip it in large vats of toxic, carcinogenic chemicals in order to extract something useful out of it.

3) Once all toxic chemicals are cleaned, etc. - if you want "4K" data files out of it, you must invest in a machine that costs roughly US$ 1 million in order to load it accurately, and create data.

4) Ah... but the data must then go onto massive amounts of drives, driven by very expensive machines. Another several hundred thousand US.

5) Then load it up into a software device, and... presto! Images!

The film workflow is completely insane when you look at it as an outsider. We are all just used to it, so we think it makes sense.

It takes most SCRATCH customers on the Skulltrail hardware about an hour to get an hour processed. If you have a lab that can get an hour of film scanned at 4K, delivered to you, and loaded into your system in 18 hours over the course of an entire feature at a price similar to compact flash cards and a fully loaded Mac Pro and/or SCRATCH system, then you are a business genius.

Yes, I'm being a s***t ass. But I get very tired of people dissing the workflows that are out there without actually knowing much about them.

Lucas
------
ASSIMILATE, inc.
LA, CA, USA

 

[Dylan Reeve]

When you put it that way Lucas it's quite interesting.

No one in their right mind (besides the likes of George Lucas and Peter Jackson) would consider shooting on film and then personally owning all the equipment necessary to get it to the point they can edit with it.

You shoot film you simply know you are going to be sending your cans off to a lab and then possibly another facility for processing and telecine before you can edit.

But when it comes to RED, because it's small and we can stick a CF card in anything then we expect, immediately, to be able to do it all ourselves. Now.

It's odd. If there was a company called RedLab that took your CF cards or RED Drive and shipped you back a Digibeta or SR tape, would that be good enough? It'd still cost less than film, the workflow at edit would be identical. But I suspect the expectation of do-it-myself would remain?

 

[GlennChan]

Always go back to the source.

Here Sony HDCAM SR Pages

Hi Tekcin,

I was referring to the difference between "true" 24 frames per second progressive and 24psf (progressive segmented frame). In psf, the progressive frame is split into two fields and the fields are sent one at a time. This works well for CRTs since they are displaying 48 fields / second, which is less flicker than 24 frames / second. (The CRT could have a frame buffer and flash the progressive image twice / three times... but that would probably be expensive and I think it would degrade resolution since the beam would move twice as fast.) So I think that's the difference between true progressive and psf. For the camera manufacturer, Red has indicated that it takes extra work to implement psf.

re: HDCAM SR
What Jim+other members of Red said seemed to indicate that recording 1080p24 (true progressive) to a HDCAM SR deck won't work, because the deck will only accept 24psf.

Thread here:
http://www.reduser.net/forum/showthread.php?t=14589&page=2&highlight=psf

Scratch presumably can turn Red footage, debayer it at high quality, and spit it out as 24psf for recording to a SR deck. (Which is, I think, a good way of doing things. As opposed to implementing 24psf output, so that you can have low-quality debayered 1080psf24 coming off the camera live for recording to SR.)

 

[Michael Thornton]

Glenn,

We are now OT for this thread.

We had a Sony deck's input programmed for 24p progressive.

I know for a fact 24p was send to the deck from Scratch Cine, and it works.

You just need to programmed it.

Internal conversion maybe programs also.

* HDCAM SR and HDCAM recording
* MPEG 4 Studio Profile Compression
* 4:2:2 and 4:4:4 (optional) record capability
* 10 bit per sample video record capability - log or linear
* Records 1080 line (Interlace and/or Progressive) and 720 line Progressive images
* Records 12 channels of audio at 24 bits per sample
* Optional Digital Betacam playback
* Pre-Read Editing, Edit/Record Confidence Playback
* Records non audio AES/EBU data in audio tracks
* Internally routable audio outputs


Features
Feature Benefit
* HDCAM and HDCAM SR recording
Maximum flexibility in one machine to meet current and future HD demand.
* 4:4:4 RGB record capability in HDCAM SR format (optional HKSR-5003)
4:4:4 recording is competitive with 2K data for high end archiving and digital intermediate workflows.
* 12 channels of audio
Provides maximum benefit and flexibility to broadcasters and archivist that require more than 8 channels of audio for broadcast and DVD mastering.

Why would you want interlace any way?

I always try to stay as closed to the final 24p as I can for film out.

Wouldn't you?

Tek

 

[Harry Clark]

Tek,
Glenn Chan is right. It's not that anyone "wants" interlace, it's the way HDCAM SR is recorded. But it's not an interlace-captured image. It's the original, true, 24fps (23.98 fps) SPLIT and recorded in segmented frames, similar but NOT THE SAME as interlace. It's done for recording ONLY. Panasonic's HDX-900 camera does the same thing when in 1080 mode. It's simply a way of halving the data in each frame to write it to tape. Nothing is lost, there are no interlace artifacts, there is no pulldown. Just a clever way of getting it all on a small tape.
Now, having said that, the deck (I think) can take a 23.98p 1080p signal and make it 23.98 pSF 1080p. But I'm not sure the field recorder will do that. You might need the studio deck.
Cheers,
Harry

 

[Michael Thornton]

SPLIT and recorded in segmented frames

It shouldn't mater how it is stored to me.

Just like the one and zero, as long as I get the image progressive in the end.

24p going in >&^%^& splits up to many pieces $&^%&^%&> and 24p going out without loosing bits.

That's all I care.

The fact is, there will be 1080p play back from camera running build 16 that can be captured with very little degrading from 1080p output.

K.I.S.S.

What ever the client want the client gets, 1080p, Digibeta, BetaCam, or DVCPRO### what ever.

It's their choice.

Tek

Panasonic's HDX-900 camera does the same thing when in 1080 mode.

Before RED cam out we were very happy with Panasonic DVCPRO100 even on tapes.

Red made my dream come true as far as DOF and Cine lenses without adapter.

some time we still shoot HVX200.

 

[Harry Clark]

Luki,
In the same spirit of tongue-in-cheek...
Imagine a world where there has always been video, electronic imaging, etc. Film never existed.
Some guy named Edison, who used to be in the light bulb business, announces he's going to make a new camera system from scratch. The goal is to develop a system that can maintain consistent high-quality, 500-year archival imaging. Everyone laughs and asks why this outsider from the light bulb business thinks he can achieve what camera designers like like Mannard and Briens have not!
He partners with a guy named Eastman and they create a system that is format agnostic, has a wide dynamic range, and is free of the artifacts of the standard imaging systems. No more fixed pattern noise, rolling shutter artifacts, dead pixels, codec errors, backfocus issues from thermal expansion of the imager, no fans, a clear viewfinder that always works, and above all, an archival medium that will port to any known system, and likely any system that will be developed. You can even get through the day with just one battery, as the cameras do not use power when they are not actively shooting! High and low speeds are easily achieved by speeding up the cameras or slowing them down, and to everyone's surprise, high speed does NOT imply a loss of resolution!!! Best of all, Eastman promises to update the "film" that goes in each camera periodically, taking advantage of improvements in this fast-moving area. Owner-Operators' collective jaws drop as they realize that they do NOT have to buy a new camera every 9 months!
DPs and directors trip over themselves to use this exciting new medium. Some guy named Blairmont buys 76 of the new cameras, and a guy named Petersdad starts a company called OnHollywood with the mission of creating the first "lab" to "process" and "transfer" this new medium. He becomes a bazillinaire, as does Blairmont, Eastman, and Edison. A fast-moving director named Doderberg uses a rickety prototype to shoot the first two features on the new medium, while another director named Steelberg declares that he will NEVER use this new medium and will stick with video "until the last VTR shuts down" But it's all over in 18 months and the world never looks back.
Just a joke, no flames please...
Cheers,
Harry

 

[Jeff Coatney]

I'm fascinated by the constant "Workflow" debate. Since Red is neither "Film" nor "Video" why does Red have to create a workflow that is tethered to those now arcane and inefficient "workflow models"? The Red workflow is SIMILAR to Film and Video, but it is much more efficient, much higher quality and much easier to learn. Red's workflow is as unique as the camera, and once you bother to learn it, you discover what an improvement it is over Film & Tape workflows.

What is it about Post that makes people think it is supposed to be completely free of improvement or change? Your software is constantly changing, formats are constantly changing, the internet is constantly changing. Is Post supposed to remain locked in a state of suspended animation? Are Post people that inflexible that they can't learn a few new ways of producing their work? We're not talking about a huge leap here, it's not like we're asking cavemen to suddenly start walking upright on two legs overnight. I wish some of you workflow critics could hear yourselves. You sound like children. Honestly, do you really need to be spoon-fed?

 

[Mark L. Pederson]

Luki,
In the same spirit of tongue-in-cheek...
Imagine a world where there has always been video, electronic imaging, etc. Film never existed.
Some guy named Edison, who used to be in the light bulb business, announces he's going to make a new camera system from scratch. The goal is to develop a system that can maintain consistent high-quality, 500-year archival imaging. Everyone laughs and asks why this outsider from the light bulb business thinks he can achieve what camera designers like like Mannard and Briens have not!
He partners with a guy named Eastman and they create a system that is format agnostic, has a wide dynamic range, and is free of the artifacts of the standard imaging systems. No more fixed pattern noise, rolling shutter artifacts, dead pixels, codec errors, backfocus issues from thermal expansion of the imager, no fans, a clear viewfinder that always works, and above all, an archival medium that will port to any known system, and likely any system that will be developed. You can even get through the day with just one battery, as the cameras do not use power when they are not actively shooting! High and low speeds are easily achieved by speeding up the cameras or slowing them down, and to everyone's surprise, high speed does NOT imply a loss of resolution!!! Best of all, Eastman promises to update the "film" that goes in each camera periodically, taking advantage of improvements in this fast-moving area. Owner-Operators' collective jaws drop as they realize that they do NOT have to buy a new camera every 9 months!
DPs and directors trip over themselves to use this exciting new medium. Some guy named Blairmont buys 76 of the new cameras, and a guy named Petersdad starts a company called OnHollywood with the mission of creating the first "lab" to "process" and "transfer" this new medium. He becomes a bazillinaire, as does Blairmont, Eastman, and Edison. A fast-moving director named Doderberg uses a rickety prototype to shoot the first two features on the new medium, while another director named Steelberg declares that he will NEVER use this new medium and will stick with video "until the last VTR shuts down" But it's all over in 18 months and the world never looks back.
Just a joke, no flames please...
Cheers,
Harry

LOL!

You just missed the part when Edison's firechief launches a website called FILMUSER.NET - everybody freeks out and bangs their spoons on high-chairs and demand that Edison make the camera shoot 29.97 fps, insist the camera is useless until Edison makes a low cost telecine device - and they refuse to pay their outstanding balance on the cameras they reserved until the camera can record in-camera audio.

Meanwhile - on another webforum - devoted to hard core video shooters - a few posters insist that Edison is a "one trick pony" that "just got lucky with the light bulb" - and they insist that color grading with printer lights will never work.

 

[Steve Sherrick]

I'm fascinated by the constant "Workflow" debate. Since Red is neither "Film" nor "Video" why does Red have to create a workflow that is tethered to those now arcane and inefficient "workflow models"? The Red workflow is SIMILAR to Film and Video, but it is much more efficient, much higher quality and much easier to learn. Red's workflow is as unique as the camera, and once you bother to learn it, you discover what an improvement it is over Film & Tape workflows.

What is it about Post that makes people think it is supposed to be completely free of improvement or change? Your software is constantly changing, formats are constantly changing, the internet is constantly changing. Is Post supposed to remain locked in a state of suspended animation? Are Post people that inflexible that they can't learn a few new ways of producing their work? We're not talking about a huge leap here, it's not like we're asking cavemen to suddenly start walking upright on two legs overnight. I wish some of you workflow critics could hear yourselves. You sound like children. Honestly, do you really need to be spoon-fed?

I can assure you that post facilities do take time to come around on new technologies. Especially the ones you would least expect, i.e. the big studios. Go into their machine rooms and you will find what looks like antiquated technology, but in fact it is still in use. The reason is that they have a lot of investment, not only in the infrastructure but also the training. It takes time to ramp up to new technology. It takes time to implement it into an existing pipeline.

Having said that, it's often the smaller post facilities (the boutiques if you will) that implement cutting edge technology. Either because they are just starting out and it makes sense to think towards the future or they want to take bigger risks with potentially big reward down the line.

I think in due time Red will permeate into many post workflows. We're only 9 months in right now. By this time next year, Red will be considered to be a common format, and there will probably be far less speculation, misinformation, and general confusion surrounding it. But I think we have to give post facilities the benefit of the doubt right now until they catch up to speed. There are so many formats, codecs, etc out there that the post guys do have a lot on their plate.

 

[Mitch Gross]

Tek-

The playback output on the RED can be set to 1080p. An HDCAM SR deck cannot read this signal. It can read 24psf. That is an output available from Scratch. But this does not answer the original question, which is the desire to play directly out of the camera to a deck without getting a computer between them.

 

[Brandon Kraemer]

Don't blame the editors, blame red...

Don't blame the editors, blame red...

OK,
Just venting I suppose. Take it with a grain of salt... Editors are the SLOWEST in the world to pick up new skills, or update their systems... so help us on the imaging end by letting us coddle them...
I know it's a Digital Cinema Camera blah blah blah... RAW blah blah blah... but...
Please? Help?
Cheers,
Harry

Harry, quite the contrary, editors are the FASTEST at picking up new skills and updating their software, etc. The problem with the RED workflow is with the software development. All the software available to create a RED workflow is only in Beta release. How long has this camera been out? Years now. I'm taking on a RED project this week and I had to write a 4 page manual on how to approach the job because there is no centralized place to get the full story on how to work with this footage.

Furthermore, from the editors perspective, none of this software is properly supported. There is virtually no tech support for Crimson Workflow, because it's just one guy behind that software and I think he has a real job too. Redcine is very prone to crashing even on the newest systems and the docs and tutorials on this software are basic at best.

RED needs to get it in gear to support the post side of the business if this format is going to catch on properly. Releasing the SDK was LONG overdue. Someone please send the editors a RED life raft.

 

[Keith Neff]

Tek-

The playback output on the RED can be set to 1080p. An HDCAM SR deck cannot read this signal. It can read 24psf. That is an output available from Scratch. But this does not answer the original question, which is the desire to play directly out of the camera to a deck without getting a computer between them.

Hey Mitch, we just did a record test with HD-D5 of the Red 23.98 P signal and it was happy. It's not dual link but it does work.

 

[Mitch Gross]

Yes, the Panasonic D5 decks will indeed accept the progressive signal. Don't hurt yourself carrying that deck to set! ;-)

 

[Brandon Kraemer]

So how are they going to "digitize" on set? They would need to shoot at the post house. LOL

Not to mention you don't "Digitize" Red footage.

 

[Michael Hastings]

Yes, the Panasonic D5 decks will indeed accept the progressive signal. Don't hurt yourself carrying that deck to set! ;-)

:biggrin: You cine guys are all wimps! :biggrin: Try hauling around an Ampex DVTR - that's when men were men.

 

[Keith Putnam]

If they're really set on on-set post, it seems to me the best thing to do would be for the on-set RED tech to back up the footage from the REDDrive or CF cards to an extra hard drive (in addition to whatever other backups are being made) for the editor's use. The editor can work with the QT proxies in Final Cut Pro as long as the original .R3D files and QT proxies are present in the same directory, and when editing is complete they can render out the sequence in any flavor necessary.