We will always tell you the truth about our development.

[Gavin Greenwalt]

If Avid called up RED and asked for the REDCode APIs to integrate into Composer I'm sure RED wouldn't turn them down.

This is stupid. We all knew 4k post production was going to untennable on feature films for a while. It's bleeding edge limited availability. You aren't going to see iMovie support realtime 4k until everybody needs it.

In the meantime there are numerous paths to a 4k deliverable. And many of them are affordable.

The companies you list as having tight integration with RED isn't a result of RED only paying them attention. It's a result of only them paying attention to RED.

 

[laguun]

the frustration for me is simple:

we have an oustanding camera coming. 3 of them to be precise.

and i am getting more and more irritated by the unexisting workflow.

red has many excellent folks on board - and therefore i wonder why my fears are called silly.

once more - i have to manage 2 redbasing fullfeatures vs. 35 within the next 60 days, and i can´t offer them a precise cost & calculation. FCP and scratch are fine, but both are technically inferior to our needs - no rgb, no 4, or no 4k or 2k realtime in scratch, additional investment etc.

the inexpensive 4krgb10bit route is adobe, the others have to be decompressed - all that leaves me in certain uncertainity regarding the workflow, and hey, i am one of the outspoken promoters of red.

i do apologize for my somewhat biting tone, and i am certainly one of the most intrepid and advanced customers of red in germany - and this is why i need HQ postproductions solutions above fcp&scratch and a roadmap of the compatible systems, as i did manoveer 2 fullfeature near too change from 16/35 to red.

i have, so far learned that J Jis a somewhat exceptioal entreprneur, and i might be a paranoid victim of this whole biz after 19 years in it, >10 spendt on the *dark side of the force*, at discreet.

red did announce that they will support 3hrd parties, and i think it is necessary to give a ETA - not in weeks, not in monhths, but something as Q1/2008 for the 4k DI manufacturers would be highly reassuring.

Important people don´t talk much - and i respect and understand that.

but i should be able to explain my customers a somewhat easy post-route for their 4k.

regarding being "silly" or whatever, i suppose i better don´t comment on that.

p.s.
@iridas: achim is at full speed implementing iridas DI here, btw. We have a solid 15-20p playback, on 4k, uncompressed as of now.
@everyone - don´t get this as a rant. i am frustrated, yes, red has homework to do, yes, and there are unasked important questions. once more, i will gladly invest 50.000 for another 3 red cameras, but if i am -forced- to decide on post or aquisition investment, i have to speak out.

 

[Jeff Kilgroe]

laguun,

What is your workflow when shooting 35mm? What is your DI process or workflow and file formats used? Chances are, RED will drop right into that workflow. If you're using DPX sequences, not a problem.

I guess what I'm saying is that if you look at RED in the same way you do a 35mm film camera, it might relieve some of the headache. If you shoot film, you develop and scan into your DI workflow. If you shoot RED, you do a base CC in REDCINE and then export the files you need into your DI workflow. For now there's RED Alert! on the Mac, but REDCINE should be here soon... It might even be here with cameras 51-100, if not already to some users, since there is no RED Alert! on Windows systems. You don't have to use FCP or SCRATCH or rely on any post software that directly supports .R3D files or QT ref files to REDCODE.

Tell us, what is your current workflow and / or what workflow do you want to use? There are ways to make this work.

As for RED giving an ETA on third-party support, I doubt that is something they can or will do. It's going to primarily be up to third party developers to provide RED support themselves. Obviously, they will have to work with RED on the technical details, acquiring APIs, etc..

Like Gavin said above, it's a case of these companies not paying attention to RED. And I agree with everything else in his post. Now that cameras are shipping, I'm sure plenty of third parties are taking notice, but I'm betting that most are still waiting to see how this whole RED ordeal shakes out or they are only now getting the ball rolling on attempting to support RED directly.

RED wants to support as many workflows as possible and help others support RED. They are working on it, this is what Jim posted to this thread earlier.

We intend to support as many platforms as possible in due time. We do not have the resources in place to accomplish everything at once. We also don't have control of the other companies policies.

RED Alert! is OSX only. REDCINE is OSX and Windows. There is growing Apple support to native. DPX files can be generated for Quantel, Baselight, etc. Discussions are ongoing with several other companies.

Jim

 

[Roberto B]

laguun,

What is your workflow when shooting 35mm? What is your DI process or workflow and file formats used? Chances are, RED will drop right into that workflow. If you're using DPX sequences, not a problem.

I guess what I'm saying is that if you look at RED in the same way you do a 35mm film camera, it might relieve some of the headache.

a big NO NO here!.. then, where's the great advantage to go with Red?

 

[MichaelP]

If Avid called up RED and asked for the REDCode APIs to integrate into Composer I'm sure RED wouldn't turn them down.

....

The companies you list as having tight integration with RED isn't a result of RED only paying them attention. It's a result of only them paying attention to RED.

All manufacturers are paying attention to RED as they would any other acquisition format. To suggest otherwise is rather naive. Avid has asked RED for the codec and ability to read native files in order to provide a an efficient workflow regardless of post production solution desired (and I can only guess others have also). At this time, the tighter integration is with Apple and Assimilate and not from lack of asking.

Michael

 

[Gunleik Groven]

a big NO NO here!.. then, where's the great advantage to go with Red?

Well... this is where we happen to disagree... -

The advantage of shooting Red in a 35mm scenario is

1. You don't have to shoot on 35mm
2. You don't have to scan.

Do some calculations on thos two factors, and ... well... I'm fine -

When you IN ADDITION to that can go the 1080 route, either through HD-SDI or through downscale in Redcine for TV/commersial stuff...

The advantage is quite .... stunning -

You could choose an F23 workflow (given that the HD-SDIs works as announced) and record directly to 4:4:4 1080/2k OR you could do both - through the Redcine option (futureproofing low-budget productions)

So

The advantage is still there. By far -

Gunleik

 

[Mr. Paul White]

a big NO NO here!.. then, where's the great advantage to go with Red?

I understand you.

 

[Jeff Kilgroe]

a big NO NO here!.. then, where's the great advantage to go with Red?

I think you misunderstood my point. From a workflow point of view, if you drop RED right into an existing 35mm film workflow it's a huge advantage. You essentially go right from shooting, direct to DIs without the time and expense of developing film and scanning your negs. ...That also doesn't take into consideration the other advantages we're seeing with the camera itself. IMO, film is rapidly becoming a medium for specific artistic or sentimental purposes. The days for film in mainstream production are definitely numbered.

 

[Gavin Greenwalt]

All manufacturers are paying attention to RED as they would any other acquisition format. To suggest otherwise is rather naive. Avid has asked RED for the codec and ability to read native files in order to provide a an efficient workflow regardless of post production solution desired (and I can only guess others have also). At this time, the tighter integration is with Apple and Assimilate and not from lack of asking.

Michael

I have no inside knowledge of Avid (although I love it) however what I meant was I doubt Red is the hold up. It's a question of priorities. And quite honestly I don't see why Red should be their priority. The number of HVXs (DVCPro 100) in the wild is enormous comparatively. If I was Avid I would be paying attention too but I wouldn't let it get in the way of other priorities.

I would love to see 4k integration with Avid but not if it ruins it. I don't want to see REDCode integration until they can figure out a way to work it into avid in such a way that it's as fast as DV. Aka the wonderful avid timeline feedback. No rendering etc etc.

 

[Kyle Mallory]

After the big crash... I think I'll bump this thread again.

I was talking with a post/finishing software company the other day about their need to support the RED codec natively. The word, directly from them, was that they inquired, but that Apple and Assimilate have exclusive contracts with RED for editing and coloring, respectively. I'm not going to name names, its a small industry, and I don't want to jeopardize relationships. But, given RED's silence on the matter, and the increasing number of statements outside of RED reinforcing this position, I can come to no other conclusion.

 

[jimhare]

Is the lack of native support that big a deterrent? I was at a RED shoot day last and had a discussion with the local post house and we agreed on a clean work flow right away. I could edit in whatever res was convenient, output an EDL and DPX files, and they could take it all the way through grade and to a 35mm print if that's what I was after.

Other than the need to create DPX and have some large drives, what is the inherent issue? I'd still take that over stock costs, labs, transfers and the rest any day. Buy a dedicated machine that does nothing but transfer DPX all day and it would still cost less than a single telecine session.

If they do end up making it native, bonus. If not, still a great work flow and spectacular results.

I saw this work flow in action from on the set, through to final grade.

Looked pretty good to me.

Jim

 

[Patrick Tresch]

I had a dream...

I had a dream...

Is the lack of native support that big a deterrent?

It's about REDvolutionning the whole post-production workflow!!!

Why should the post-production be left appart? 4k/2k DPX workflow for DI is "old school" and is (was) the only available until now.

Let's dream of 27Mb/s 4k based on REDCODE RAW solutions. No time/space/$$$ consuming DPX itermediate.

I can still wait 1 or 2 months until there is such a sultion.:tongue:

Some call me a dreamer!!!:biggrin:

Pat.

 

[Kyle Mallory]

It's about REDvolutionning the whole post-production workflow!!!
...
Let's dream of 27Mb/s 4k based on REDCODE RAW solutions. No time/space/$$$ consuming DPX itermediate.
Pat.

Exactly! Sure, I can export DPX sequences, etc. But REDCODE provides soo much more flexibility because of its nature. For starters, as soon as I export DPX files, I am losing all of the ability that REDCODE offers. Suddenly, things like gain, gamma, saturation, etc, that were stored as metadata become baked into the DPX files, limiting my ability to change them in the future. Sure, I can go back and re-export the footage from REDCINE, but... that could set me back hours, even days depending on where the original footage is, and whether I have a copy of REDCINE available to me. Another benefit of REDCODE is that it scales easily, immediately, for proxy files. As I understand others (don't make me go searching past posts), I could, in theory, tell my NLE to render a 1k version as a proxy, direct from the footage, in realtime, for editing, compositing, correction, etc, without having to actually EXPORT a 1k proxy. Then bump up the quality to "Best" just before I render. This is HUGE.

Right now, for me personally, my biggest issue is that I have to keep a copy of OSX or Windows around to run REDCINE, Scratch, or FCP on to get at my footage (By the time I get my RED, I fully intend to be 100% Linux-- RED is the only thing that will stop me). In the case of OSX, its even worse because I have to run on Apple hardware. Part of being an OPEN environment, an OPEN platform, is not locking me into a particular platform, or tool. Unless I misunderstood, when RED first showed up on the scene, they were all about "OPEN". Sure, they can have a preference for the hardware/platform/tools that they recommend and promote, but I was under the impression they weren't going to lock me in to their school of thought. If they've made exclusive contracts with a couple of select companies, that's fine too, but they should clarify that with their customers who expected differently.

Ultimately, my biggest issue with this topic, is simply that Jim/RED won't comment. And don't tell me they're too busy putting camera's together. If Jim's got time to post asking where Acehole is, I would hope he's got time to write 2 or 3 lines that are relevant to an important topic. (Jim, don't stop asking where Acehole is, its fun, and adds an easyness to the board... just take a moment to address some of our *other* concerns, too.. Please!)

Either way, exclusive contracts that lock me in, or open architecture that truly promotes 3rd party vendors and tools is fine. I just need to know so I can make plans to have the money available for the right tools (whether Linux, Windows, or Apple), and prepare to explain the workflow to my potential customers. I don't think this is too much to ask.

 

[Joel Kaye]

my biggest issue is that I have to keep a copy of OSX or Windows around to run REDCINE, Scratch, or FCP on to get at my footage

If Color reads RedCode Raw and Final Cut edits it - then we've got something. Is that the plan?

Or does Color compete with Assimilate for REDCODE access in a color editing app? That would seem like a pretty stupid situation to get RED into so I have a hard time believing it... but hey, maybe the deal was made before Color was announced and now there's a complete clusterf*&k.

I might just RedAlert to ProRes 1080P and do everything from there if that's the case. I wouldn't be surprised to fine RedCine is bootcamp only for a while considering Assimilate usually develops for XP. Perhaps the holdup is getting Apple to release ProRes on XP so RedCine can get to it.

:detective2:

 

[Joe Carney]

I will be surprised if Apple delivered anything before NAB. For Mac users, I hope they do. RedCine on Windows and Mac will help quite a bit.

 

[Mike McCarthy]

As I see it, there are 3 major advantages of finding a way to natively edit Redcode files over DPX exports of them.
1. Time and storage-Redcode files are produced by the camera live and don't take time to generate or export. They also require significantly less storage space.
2. RAW attributes-As noted above, file characteristics stored as metadata can be adjusted with no quality loss until they are baked into an RGB or YUV export, plus the RAW factor is contributing to the low file size.
3. Wavelet Compression-This is the attribute of Redcode that allows performance gain by playing files back at lower resolution without making proxies. In AfterEffects, if you set your 4K DPX to 50%, it READS the whole large 4K file, and RENDERS from that a 2K result. With Wavelet compression, the data required to generate a 2K image resides in the first 25% of the data for a 4K image. The system only reads the data required for the 2K image, lowering the storage performance requirements.

Being a fan of Cineform, they are the best example I can give you. Wavelet compression is what allowed them to edit full resolution 1080 files on regular desktops years ago, by only reading and displaying 960x540 until ready to export. Other solutions required the generating of proxy files because most other forms of compression (MPEG, DCT, etc.) require reading the entire frame before a low resolution image ca nbe formed, eliminating any disk speed advantages of low resolution editing unless proxies were created. I have been using Cineform software for 4 years, and it is not perfect, but their Wavelet compression is the best imaginable solution for high resolution post-production on a reasonable budget. Redcode will offer many of the same advantages if it can be editted natively in an NLE.

 

[GlennChan]

DCT is capable of lower-resolution proxies like wavelet... but no one is doing that. Though it's a moot point. It's less computationally expensive so it's easier to have multiple streams of Prores decode than Cineform (AFAIK). I think that is why Apple wanted to go with Prores instead of licensing Cineform... you have the lower quality, but they get to advertise multiple streams of RT; and perhaps there is some advantage as to having the Aja IO/Prores.

2- Ok back on topic... as far as native editing versus uncompressed editing:

Yes, native editing can have some advantages. For finishing however, the benefits of uncompressed editing would be:
A- Industry support for 'standard' formats like SDI, DPX, etc. These formats are supported and *they work*. Unlike MXF, HDV, etc. which doesn't always work.
No showstopping incompatibilities.
B- No overhead to decompress the footage. Redcode RAW is a little different in that high-quality debayering would be slow... so you probably want to transcode anyways (e.g. to uncompressed, or to Cineform, or to prores, or some other intermediary format).

A is a matter of development effort + resources. B would be ok as long as the codec has low overhead.

3- Ack, I think I got three different methods of working confused there. native editing (e.g. working off Redcode RAW), compressed editing (cineform, prores), and uncompressed editing.

 

[Mike McCarthy]

Agreed that DPX is much more compatible, that is the primary thing in its favor. Also, at some point you will have to render out to an RGB format, and if printing to film, you will need DPXs. The later in the pipeline that you do this, the longer you get to take advantage of the benefits I listed before.

Maintaining RAW until you get to CCR has distinct advantages. If the final graded output is the only thing you render to DPX, you save tons of storage space.

Until Redcode is supported natively in certain apps, DPX will be a popular way to go. I am currently onlining a feature with Cineform, but we have to render out to DPX for compatibilty with certain software we use. And we have to render DPX at the end as well for film out of course. But the amount of space we save along the way is enormous, high rez recapture, conformed exports, colored textless, colored texted, would be 1.5 TB each, instead of the 120GB each that I am working with. That is the difference between a firewire drive and a whole array.

 

[M Most]

There is a very major disadvantage to using Raw files for finishing that nobody here seems to be paying attention to (probably because very few people here have ever actually done longform finishing). If you use the Raw files, you must by definition have every clip shot for the production on line simultaneously when you conform and color correct. This means many terabytes of storage, and it also means that the conforming and color correction software must build a database that is much larger than usual in order to track it all. In any case, it is much more unwieldy than extracting the frames that are actually used and conforming from that - which is the alleged workflow with Redcine and a pull list. If the extractions are done correctly, there is no "loss" that is of any real consequence when doing the finishing steps, at least none if it was shot competently.

 

[Mark L. Pederson]

There is a very major disadvantage to using Raw files for finishing that nobody here seems to be paying attention to (probably because very few people here have ever actually done longform finishing). If you use the Raw files, you must by definition have every clip shot for the production on line simultaneously when you conform and color correct. This means many terabytes of storage, and it also means that the conforming and color correction software must build a database that is much larger than usual in order to track it all. In any case, it is much more unwieldy than extracting the frames that are actually used and conforming from that - which is the alleged workflow with Redcine and a pull list. If the extractions are done correctly, there is no "loss" that is of any real consequence when doing the finishing steps, at least none if it was shot competently.

Mike -

We do long form finishing. Using RAW files from RED (.r3d) - keeping all the files "on line" for entire feature (in 4K!!) is under 5T - that is well within the management scope of even a small shop like ours - and yeah, bigger budgets have bigger shooting ratios - but even 8 to 10 T is no big deal these days. We are keeping ALL camera master .r3d on-line on a san connected to our Scratch system. This way at any time during the edit - we can drop in a edl and screen or export or test grades -