The IBC betting game - realtime 2k or more out of RAW files - How many?

[Gunleik Groven]

ouch.

i was going for a humorous bet here...

 

[Mark L. Pederson]

My point was that it's only a disclaimer if the buyer/reader understands what Alpha and Beta actually mean.

what he said (bluntly, with a grin)

 

[Ariana]

As has always been the case with RED, you can wait a while until things are where you'd like them to be, or you can jump in and accept alpha/beta and make things work as the rest of us are.

This attitude of "I want everything now and I want it my way NOW" is so un-RED like. I'm not sure how people completely miss the whole work in progress concept and being years of ahead of everyone else.

I can also understand Laguun's concern but not Bruce's. At least laguun has bought something from Red while Bruce is an armchair critic. Why should Red listen to Bruce? Does he have any intentions of buying red product or just complain about them without shooting red yet?

 

[Bruce Allen]

I can also understand Laguun's concern but not Bruce's. At least laguun has bought something from Red while Bruce is an armchair critic. Why should Red listen to Bruce? Does he have any intentions of buying red product or just complain about them without shooting red yet?

Ariana, I was shooting and editing Red before RedCine was released, and before native FCP editing.

Please note that apart from personal attacks, nobody (except people who sell Scratch or own Scratch) has managed much of a response to my statements that:

1. Red has a workflow page with a pretty arrow from RedCine -> Avid. No disclaimer saying "coming soon" or "beta". This implies it works now, which is deceptive. Me asking for this feature is not whining.

2. That the argument given for encrypting the Red file format ("maintaining image quality") is not valid (as has been proved with DSLR 3rd party RAW decoders squarely beating manufacturers')

3. That it's really not that as hard for developers to keep up with file format changes as Red and Assimilate say it is (anyone who has actually written some C++ graphics code want to disagree?)

4. That it would help Red hugely if the major players could read their file format.

Look, if you are all happy with Red taking the Microsoft Word approach, great.

But to me it's a mistake.

Red has so many, many strengths. A closed file format isn't one of them. Unless they had a truly industrial-strength coding and management team behind it. Seeing as they can't even release a Windows QT wrapper and find it too overwhelming to talk to all of the major post-production players at the same time (incl discreet, Avid, etc), then I say they'd be better off giving out file format specs and sharing the load.

I'm not talking about individual skill. Everyone I've ever met at Red and Assimilate seem to be extremely talented, fantastic people.

I'm talking about insufficient resources to deliver what is already advertised on their workflow page.

Bruce Allen
www.boacinema.com

 

[mikeburton]

If you are a post professional you find ways to work around any scenario that may be a speed bump. If you do not consider yourself a "Post-Professional" and refuse to spend the time it takes to be one (if your dedicated to production work ie DP, Director etc) than what the hell are you doing complaining about post workflow's for anyhow? Let the professionals work out the kinks. The difference in software developments will only make the post process easier as it develops but these things do not happen over night and it makes me laugh when people expect them to.
Before Crimson and without a Scratch we were exporting ProResHQ from the first Beta version of REDCINE and manually conforming the Online. Wasn't fun, but we did what we had to do to get the pretty pictures. In the time most people spend complaining about this workflow not being there or that workflow is insufficient etc they could be finding a solution to their problem and finishing their projects. Quit complaining about what you don't have, man up, and get the job done with what you have. Rest assured RED knows what is still needed to streamline the post processes and will do what they can to assist the majority of the requests in time. Requesting what you would like to assist your needs is not a bad thing but some of these posts are just down right disrespectful and insulting to those of us who make it work no matter what!

 

[Bruce Allen]

Back to personal insults. That's cute. I was doing Red projects before even the first Beta of REDCINE so I am one up on you Mike. And yes I have done plenty of workarounds.

That totally fails to address my point though:

1. Red has a workflow page with a pretty arrow from RedCine -> Avid. No disclaimer saying "coming soon" or "beta". This implies it works now, which is deceptive. Me asking for this feature is not whining.

Bruce Allen
www.boacinema.com

 

[laguun]

If you are a post professional you find ways to work around any scenario that may be a speed bump.

In this case here, as we have hdcam, 35mm mechanical Arri and red cameras on offer, many "post professionals" choose not to shoot on red.

Because they don't want to work around.

It has been a year now. And no Windows, Avid, direct raw and sound are not small workarounds. They are huge obstacles.

 

[mikeburton]

Back to personal insults. That's cute. I was doing Red projects before even the first Beta of REDCINE so I am one up on you Mike. And yes I have done plenty of workarounds.

That totally fails to address my point though:

1. Red has a workflow page with a pretty arrow from RedCine -> Avid. No disclaimer saying "coming soon" or "beta". This implies it works now, which is deceptive. Me asking for this feature is not whining.

Bruce Allen
www.boacinema.com

Bruce, I never quoted you personally and was not attacking you in any way. If for whatever reason you thought I was, well, sorry, but you shouldn't assume things. I see my post ended up under yours but it was not intended to be in response to you, nor was I trying to get into a pissing contest to see who did what first.
I was simply stating that "some" seem to spend more time bitching than working (happens a lot here) which is a total waist of time IMO. If you fall into that catagory then well, in an informal way, maybe I was addressing you. I have read many of your posts in the past and consider you a valuable member of this community. I would like to think most of your requests are reasonable and not spoiled like some.
I hear your point on the Avid thing but you obviously know the work around and I'm sure RED will address this when it comes up on their list of a billion things to do. In the meantime, I'm sure they here this loud and clear and it won't do anyone any good to bug the crap out of them.
BTW - I want de-squeezed anamorphic preview still on the RED LCD!
J/K Jim

 

[mikeburton]

In this case here, as we have hdcam, 35mm mechanical Arri and red cameras on offer, many "post professionals" choose not to shoot on red.

Because they don't want to work around.

It has been a year now. And no Windows, Avid, direct raw and sound are not small workarounds. They are huge obstacles.

Granted, some post facilities are slow to adopt and don't deem it necessary to keep up with technology as it changes by the day. In that case, that may not be the post house for you then if you are shooting RED.
The thing is that there is a working post workflow that many have adopted and are turning out RED projects constantly. The workflow may not be what you are accustomed to but it does exist and by being a RED owner or supporter it is partially your job to guide these hesitant clients in the best possible direction based off their inparticular needs. In some cases you have to be a salesman. I've found, your enthusiasm and confidence when discussing their post options are key which stands true with any non-traditional methods of technology.
If the client trusts you and thinks they are in good hands you just might get the job. If those needs are too cumbersome for what you think RED's limitations may be, guide them elsewhere. IMO, you get out what you put in and not everyone will want to ride the train right now, but its nice to have a front row seat

 

[laguun]

Granted, some post facilities are slow to adopt and don't deem it necessary to keep up with technology as it changes by the day. In that case, that may not be the post house for you then if you are shooting RED.

Mike,

its not the posthouses who dont keep up.
4K with sound and EDL basing conform is pretty basic - since years.

The problem is that the reds camera raw format (still) is incompatible with >90% of the existing top-tech, top-up to date gear & houses. FCP and scratch maybe nice, especially for those who havent be spoiled by the power of a smoke/flame/filmlight and recently bought their first DI - but they are clearly no alternative to the huge majority.

One doesnt usually buy a poststudio for *one* camera-model.

The thing is that there is a working post workflow that many have adopted and are turning out RED projects constantly. The workflow may not be what you are accustomed to but it does exist and by being a RED owner or supporter it is partially your job to guide these hesitant clients in the best possible direction based off their inparticular needs.

Its not "us" - we are pro red. deeply pro red. And we have several working workflows inhouse.

But customers who are used to an online finishing (no offline), realtime (not 2K / quarterrez) colorcorrected piece with *sound* neither find an answer in FCP or scratch or the combination both. It, lets face it, requires several additional steps and people who mainly focus on workflow, not creativity added to any team.

That should change and it makes no sense to deny this.
Thats all we are talking about.

If i.e. discreet smoke finally one day directly supports redraw, or iridas speedgrade is able to perform its GPU-basing realtime debayer instead of the slower CPU-basing right now, it will be pretty easy to understand what many clients want to have.

The red is a first class camera, hindered by second class workflows yet, and red is about to change this. As many have pointed out, its really necessary to gain *broad* support in the postindustry and allow the top-class postpro manufacturers to *directly* access the redcode raw JPEG2000 without speed penalty. Its in the best interest of any red owner/user, no matter if satisfied or unsatisfied with the given tech, that the posthouses want to support, not avoid, redcode.

 

[M Most]

The problem is that the reds camera raw format (still) is incompatible with >90% of the existing top-tech, top-up to date gear & houses. FCP and scratch maybe nice, especially for those who havent be spoiled by the power of a smoke/flame/filmlight and recently bought their first DI - but they are clearly no alternative to the huge majority.

The last time I looked, there were exactly 4 cameras that fit into the "raw" category: Red, Dalsa, SI, and Arri. Also the last time I looked, ALL of them fit your description of "incompatibility."

Fact is, RAW is a capture method. The fact that Red is making strides towards allowing it to be more than that is pretty unique, original, and forward thinking. But in almost all cases to this point, anyone doing a 4K - or even a 2K - DI using material from a digital camera in any high end DI system has done it by converting those original captured RAW files to DPX or some other RGB format. And that works quite well, because to create a deliverable product, that conversion must ultimately be done anyway. The fact that the game changes somewhat with Scratch and Red should be looked at as the beginning of an era in which there are alternatives to this methodology, brought about by a forward thinking camera company and made possible by faster, more powerful hardware. But those alternatives are not necessarily the only, or sometimes even the best way of accomplishing the task. I would contend that anyone claiming that Red - or Dalsa, or Arri, or SI - material cannot be handled by high end DI systems is not trying very hard to make it work. Because those that have seem to be completing projects quite well. The appearance of helper programs like Ian Bloom's software and some others has made this rather straightforward even for the less adventurous.

None of this is rocket science. If one is to be a post professional, that implies having the skill to put together ways of doing things successfully, regardless of whether there are complete, turnkey, off the shelf solutions or not. Quite frankly, doing film DI work is, to me, far more complex (if you're doing it right) than finishing a Red project, regardless of whether you're using Scratch or anything else.

 

[laguun]

The last time I looked, there were exactly 4 cameras that fit into the "raw" category: Red, Dalsa, SI, and Arri.

No, there are 6 camera series who fit in the raw category.
The Vision Research Phantom Series and the Weisscam HS-1 and 2.

Also the last time I looked, ALL of them fit your description of "incompatibility."

No.

As example, the SI 2K and SI 2K mini write directly into cineform raw, which can be played back, rendered, edited crossplatform in full master quality as well in Adobe Premiere as FCP, as well on OSX and Win, as well as used for VFX in combustion or After Effects etc. It would be excellent if redcode would offer that level of flexibility.

Also, all the raw cameras as well as cineform are natively supported by *the* raw DI pioneer - iridas. Iridas would benefit a lot if red allows direct raw data access, their GPU-basing debayer is next generation technology compared to the existing CPU-basing approaches.

red is coming there - and due to its well merited huge marketshare (should be >80%), many postproduction system and software manufacturers want to support them. But without access to the raw data the support has to be and will remain second class.

Fact is, RAW is a capture method.
The fact that Red is making strides towards allowing it to be more than that is pretty unique, original, and forward thinking. But in almost all cases to this point, anyone doing a 4K - or even a 2K - DI using material from a digital camera in any high end DI system has done it by converting those original captured RAW files to DPX or some other RGB format.

No. For Arri, SI, Vision Research, cineform, Dalsa and Weisscam there is no need to to convert, as there isnt for red. For many scenarios it makes sense to debayer before the DI, for many others its vice versa.
Iridas Speedgrade XR supports all of the cameras btw:
http://doc.iridas.com/index.php/RAW_Support

to create a deliverable product, that conversion must ultimately be done anyway. The fact that the game changes somewhat with Scratch and Red should be looked at as the beginning of an era in which there are alternatives to this methodology, brought about by a forward thinking camera company and made possible by faster, more powerful hardware.

Si and cineform started the era. Scratch followed more than one year later.
april 2006:
http://cineform.com/press/rel-SI1920HDVR.htm
november 2006:
http://cineform.com/press/rel-SI2K.htm

Also its important to understand that the "next generation" raw technology is already on the market - and quite impressive.

Iridas, as example, is delivering Speedgrade XR which is using GPU-basing debayer, which is much faster and allows better quality debayer in less time than the CPU-basing approach.

To allow all red owners/users to benefit from these technologies requires, once more, nothing else than direct access to the red ones raw data - and red stated often that they want to offer this.

And things like the adobe raw support by red/adobe look very very promising.

But as of now, the HQ-workflows are requiring complete debayer first and multiple transcoding steps in order to get into an avid and/or autodesk system, or a well planned multi-platform, multi-product, multi-render in order to use FCP offline and then online from the raw which isnt realtime in 4K and cant be displayed in 4K. For you and us thats daily business - but for most potential customers its highly confusing.

.

 

[M Most]

No, there are 6 camera series who fit in the raw category.
The Vision Research Phantom Series and the Weisscam HS-1 and 2.

That is true. I would point out, however, that it is very rare that either the Phantom or the Weisscam is used as an "A" camera. They are both specialty devices, although, as you accurately point out, they do record RAW.

No.

As example, the SI 2K and SI 2K mini write directly into cineform raw, which can be played back, rendered, edited crossplatform in full master quality as well in Adobe Premiere as FCP, as well on OSX and Win, as well as used for VFX in combustion or After Effects etc. It would be excellent if redcode would offer that level of flexibility.

Also, all the raw cameras as well as cineform are natively supported by *the* raw DI pioneer - iridas. Iridas would benefit a lot if red allows direct raw data access, their GPU-basing debayer is next generation technology compared to the existing CPU-basing approaches.

Well, a few quotes ago, your point was that the big iron - Quantel, Baselight, Lustre, Resolve, Nucoda - didn't support Red. They don't support Cineform, either. Hence my point about all formats (and I included SI on purpose here) needing to be converted in order to be used in high end systems. In my world, Adobe Premiere and Final Cut are not high end DI systems. I'm not talking about offline editing, I'm talking about finishing.

No. For Arri, SI, Vision Research, cineform, Dalsa and Weisscam there is no need to to convert, as there isnt for red. For many scenarios it makes sense to debayer before the DI, for many others its vice versa.

Yes, there is, if you actually have to create deliverables. That includes Red as well. WHEN you do the conversion is the only question. Once again, I'm not talking about offline editing here.

Iridas, as example, is delivering Speedgrade XR which is using GPU-basing debayer, which is much faster and allows better quality debayer in less time than the CPU-basing approach.

I'm very aware of Iridas' products, I've been using and testing them for years. But that's not what's being discussed.

But as of now, the HQ-workflows are requiring complete debayer first and multiple transcoding steps in order to get into an avid and/or autodesk system, or a well planned multi-platform, multi-product, multi-render in order to use FCP offline and then online from the raw which isnt realtime in 4K and cant be displayed in 4K. For you and us thats daily business - but for most potential customers its highly confusing.

That I sort of can agree with. What I would point out is that you have numerous quality vs. speed trade offs available in the initial conversion for offline editorial, so it's not necessarily the big deal it's made out to be in this thread.

 

[Lucas Wilson]

Iridas, as example, is delivering Speedgrade XR which is using GPU-basing debayer, which is much faster and allows better quality debayer in less time than the CPU-basing approach..

How is it better quality?

Lucas

 

[David Newman]

It is the nature of demoasic algorithms, there is no true best quality, as all about guessing the missing pixels to turn CFA Bayer into 4:4:4 RGB. This means no company can have a mathematically ideal demosaic, including Red, Iridas or CineForm, yet one can try for subjectively better results. Better can be in many factors, more resolved detail, more suppression of noise, more suitable for heavy color manipulation, typically not all can be achieved at once. It is completely reasonable that Iridas has a demosaic that is subjectively better received by their customers. The good thing that the R3D SDK has the RED demosaic within it, once SDK RAW is enabled, it allows the end user the choose, that what we do. The demosaic is just another creative choice.

 

[Lucas Wilson]

Hey David,

I get the subjective nature of demosaic algorithms, and what can be perceived as "better."

I think I may have just misread Laguun's post. I initially thought he was saying that GPU-based debayer was universally better. Rereading it, I don't think that's what he meant. : )

Tx,

Lucas

 

[Steven Caesare]

I can't help but wonder if all those who express such deep dis-satisfaction with the state of RED's development* as it stands today would REALLY be happy with the alternative: A completely finished camera/workflow package... available in 2010.

Just because the process has been open and you've been invited to ride along doesn't mean a development effort with resource requirements doesn't exist.

Perhaps.... just perhaps... the glass-half-full perspective is that the workflow you seek is coming as soon as it can, and all the other pieces were finished early and you've benefited from being able to get your feet wet far sooner than Sony would have ever let you.

Nah...

-Steve

* (I said "development"... hehehe)

 

[Mike Harrington]

good point steve....


I have whined a bit myself...but the alternative would be waiting...

but only waiting for red...
I think the thing some of us whine about is that there were workflows 6 months ago...4k raw...just not officially supported

either way I have a great working camera...with some cool free tools, and a workflow i can handle....with improvements coming to both the camera and the workflow

 

[Gunleik Groven]

Right, Steve...

I see the argument, and this thread has totally developed to something I didn't intend, BUT

1. @ least Laguun and I, and afaik the others involved are using RED every day and advocating/deploying it into new areas more or less as we speak. Of course we are happy to have the cam... I cannot remember anyone saying otherwise.

2. The issues are real, and even though they will be solved one day, they are a headache for the quick workflow that RED really opens for - given that Scratch isn't a choice, whith which it is almost there, and totally in given scenarios. But @ least in Norway Scratch on every corner is more of a nightmare scenario than a good business decission...

3. Transcoding before delivery isn't really that much of a hassle, if it wasn't for the possibility NOT to transcode at all (in an ideal TV oriented domain with final playout over SDI to tape) or just one transcode (in an ideal file-based workflow).

This is mostly offered by Scratch, and as Lucas hints, there isn't a quality hit through CPU debayer. There definitely is a a convencience hit and a time hit, though. What Scratch offers is soooo sexy (and also proves that it CAN be done today, NOT in 2010.. That is for 2k.) that it unevitably becomes an issue that only Assimilate are allowed to do this. And I cannot imagine that they haven't been prepared for this scenario...

That said: They have done and do something to achieve the current position. Sorta jumping on the ship while it still was called a scam, for one.
Making good solutions of course as a good first/second.
And all of a sudden a lot of people who hadn't previously heard about assimilate, relate to the company. In good (Scratch) and bad (RedCine) ways.

If I was Jim - I would have remembered that, and I guess he does...

That STILL doesn't deflect the fact that there are people out there who prefer NOT to invest in Scratch, dislikes FinalCut and have no urge to start in the "digital onelight" business that the labs used to handle for them with film, and was totally unneccesary with tape. OR they try, and get extremely bad results and concludes that RED is a noisy cam with low latitude, because they do really intuitive stuff in the onelight, that simply isn't right with RAW.
And for that simple reason they do the actually ill-informed choice of not using RED.

And I can very much understand that, even though I have come to that restinance against Scratch most likely is futile.

AND: If I DO buy Scratch, I will have great interests in keeping things status quo... And some have done that.

So... It is a complicated matter....

I do hovever think that GPU debayer and access to the RAW file format should be opened up, as soon as possible. ANd then it would be each vendors problem to keep up with the format as it develops. RED has taken so significant chumps of the markets by now, that I cannot really see anyone serious NOT wanting to follow that jumpy ride...

My 6C by now...

 

[laguun]

I can't help but wonder if all those who express such deep dis-satisfaction with the state of RED's development* as it stands today would REALLY be happy with the alternative: A completely finished camera/workflow package... available in 2010.

Absolutly true.
And its excellent what red achieved so far - really outstanding.
The camera is great - quality, price, features...

However, some peices of the puzzle still have to come together: many people (that includes us over here) are *not* happy with the workflow yet, and it makes no sense denying this.

red improved a lot on feedback, red convinced a lot of skeptics with its quality, and in order to convince the nay-sayers in the post world the best way is simply to solve the problems (sound, windows, edl/tc) and to have the camera supported by the best of the best.

red already succeeded in bringing a real small revolution to the camera market, taking for good reason the lions share of the 35mm market. But in post, most systems are not compatible. As like the camera wouldn´t have an Arri PL mount, but rather an eclair mount, not a V-Mount, but a notebook-battery etc. Sure, us here at reduser would find a way around it - but the common cameraman would be somewhat irritated - and thats still (hopefully soon to end) situation in red postpro.