Cinema camera or Video camera?

[Jannard]

There are a few ways to make your RED ONE look like it has 7 stops of dynamic range... two primary ways.

1. Over-expose. Just clip the highlights while you record and there won't be any need to screw up anything in grading. This is the easiest way to turn a Digital Cinema Camera into an ordinary looking video camera (although you will still have more resolution).

2. Grade it improperly. Use ISO or exposure to set the mid-grey (or make most of the image look right) in RED Alert!, REDCINE, Scratch or Final Cut... without any regard for the histogram. Let the highlights go because the "middle of the image looks right". Don't use a curve to pull up the middle because that would hold the highlight you wish to blow out. Clipping is an art. If you can get really good at it, you may also learn what not to do "just in case you want the full info for a specialty shot".

As has been pointed out, the RED ONE does not have as much DNR as film. While most people think it is plenty to work with... it certainly does not if you throw away info.

Here's some good news. We are here to help. Build 16 makes it MUCH more difficult to miss exposure. And REDspace does a great job of helping to see your image without tempting you to make grading errors.

There are some of you that are saying "I don't do that. I know exactly what to do and how to get the most DNR with my RED ONE". OK, then. I'm not talking to you.

Here is an example of how to get an image to clip with one grading button.

I know... this grab is not that sharp. That isn't the point here...

Jim

http://www.reduser.net/forum/uploaded/9_forcea.jpg

http://www.reduser.net/forum/uploaded/9_forceb.jpg

[conrad gaunt]

Is there ever any chance of getting the full 12bit sensor data out of redcine in a DPX, I know 10bit log doesn`t lose much useful data, but I agree with you about not throwing it away?

[Graeme Nattress]

That's a great example Jim.

Graeme

[Graeme Nattress]

10bit Log or Gamma encode from a 12bit source does indeed loose you some code value precision. However, it does so from the brightest stops where you have an over-abundance of code values, and hence, if done correctly, does not diminish the image quality in any meaningful way.

Graeme

[10s]

Do you mean I have to actually use the camera properly? that's crazy! :blink:

I've thought for a long time there is a need for a special DVD on how to properly use the camera. It should go out with every unit.

[mrmuccy]

Jim,

Can you tell me about the wavelette compression you're doing & where it is introduced in the image chain? I'm wondering if this is connected to DNR? Although I imagine its after the AD conversion.

[I Bloom]

Jim,

Did you just bust out some build 16 preview frame grabs!

IBloom

[Jannard]

We are constantly amazed at how much footage we see being "roughed up" during grading. More than we see over-exposed. I think it is safe to say that much of the "RED footage looks like video" and "I see so many clipped highlights" are a result of grading. RED RAW footage is not like film, in an inverse way. With film, the tendency is to under expose a negative to keep the shadows from blocking. RED RAW footage is exactly the opposite... which I'm sure has contributed to some difficulties.

Build 16's primary goal is to help our customers make easier decisions that produce better results. It is a major change in how to shoot and grade RED footage.

Jim

[Jannard]

Jim,

Did you just bust out some build 16 preview frame grabs!

IBloom

Nope... old Build 12 or 13 footage.

Jim

[Graeme Nattress]

With RAW recording to R3D files, the process is simple - we take the RAW sensor data, and pipe it into the wavelet compression. It does not effect dynamic range at all, being 12bit as the sensor is 12bit.

Graeme

[Jim Exton]

A little instructional tape or clip put up on Red.com, like you did with the myth buster one, would be really helpful.

Glad to hear build 16 helps, but it sounds like there is still further instruction needed.

Hopefully you all or someone else will get to that pretty quickly once build 16 is out.

[Mark Crabtree]

I'm buying a preset lens. Preset iris. Preset focus. Fool proof!
It's what we've all been hoping and praying for, true point and shoot technology. No need to look at histograms or learn how to use a complicated digital readout light meter. Your choice of Pinhole or Bell & Howell optics. PM me for group buy info.

[I Bloom]

Jim,

Can you tell me about the wavelette compression you're doing & where it is introduced in the image chain? I'm wondering if this is connected to DNR? Although I imagine its after the AD conversion.

If I could attempt to answer.

AFAIK the compression is the last step before storage, after some digital signal processing and prior to debayering (which as we know happens in post on Red.)

Wavelet compression is not connected to DNR in the sense that you aren't loosing any bit depth, nor are you changing the values in the image signal.

However it does affect subjective DNR in the sense that random noise cannot be compressed. So your noise floor "looks different" when it has been compressed, the grain is not as tight. It's "clumpy". If you think this texture is unattractive you might be less apt to let your image fall into the noise floor. So your effective DNR is reduced.

Who knows how build 16 works, but here is a question: If noise can't be compressed, and any compression will necessarily change the character of the noise, than perhaps it makes sense to change the character of the noise first ourselves, prior to compression, so that it compresses in a more attractive way.

We all have our theories.

IBloom

[conrad gaunt]

10bit Log or Gamma encode from a 12bit source does indeed loose you some code value precision. However, it does so from the brightest stops where you have an over-abundance of code values, and hence, if done correctly, does not diminish the image quality in any meaningful way.

Graeme

Understood, but it would be nice to not have to do any kind of `first lite correction` before outputting from RC, I`m simply going to convert the log back into its non-log form so I can manipulate it with other material, so it seems like needless processing which won`t save any disk space in the end. Log may be useful for filmout people etc, but nobody I know.
I`d rather not rely on RC for anything other than decompressing R3Ds, thats not to say that I won`t want to use RC for other things at other times, when its working well. I just want access to the full sensor data that the camera promises. 12bits = max 12stops. 10bits = max 10stops. Like jim says, you can shift/crush/expand colours many ways, so information that seems irrelevant might not end up that way with correct processing. why not have a non-log option, or at least a 12bit log option so we can dig a little deeper.

edit: I should point out that only having 10bit log output won`t be why some people will struggle to grade and lose information as Jim is highlighting.

ps, Can`t wait to see build 16 footage

[Daniel Browning]

Jim, you forgot to share the second half of the lesson: how to crush the shadow detail and sharpen the image into oblivion.

[Matthew Verkler]

Jim, you forgot to share the second half of the lesson: how to crush the shadow detail and sharpen the image into oblivion.

Yes, over-sharpening is key to achieving that much sought after "video look". I have successfully turned boring Red shots into much superior "digital video" with the techniques Jim has outlined, coupled with copious amounts of sharpening, or, as I like to call it, "enhancement".

Thanks Jim, for helping everyone figure out how to match the footage we have become used to!

[Jannard]

Jim, you forgot to share the second half of the lesson: how to crush the shadow detail and sharpen the image into oblivion.

That's next week... :-)

Actually, this clipped highlight thing is the single biggest issue facing RED ONE users. If it was all about the camera, that would be one thing. But so much footage is "reduced in quality" during grading that it deserves most of our attention. Again, it is the primary benefit of Build 16.

Jim

[10s]

This RED experience is a learning curve for everyone. On top of this, there are the skeptics, and then there's actual technological challenges making this even more complicated. Sometimes the cutting edge really hurts!

Maybe build 16 will solve many issues but it seems to me that the best way to get the word out and reset everyone's learning curve is a great video via DVD & WEB on how the whole system works. This will cut down on all the chaos, chatter and skeptics who are having a hey-day with poor operator execution, and blaming it on the hardware. Of course this takes time, money and it's another project/headache, but isn't it even a bigger headache to introduce new a photographic system and end up having so many get it wrong and spread rumors.

...I do like Mark Crabtree's idea on just settling this once and for all, & getting pin hole cameras! :wacko:

[Jason Ing]

Edit: I reread Jim's preview post and answered my own question. Never mind. :)

[Brandon Fraley]

With film, the tendency is to under expose a negative to keep the shadows from blocking. RED RAW footage is exactly the opposite... which I'm sure has contributed to some difficulties.

did you mean over expose? I'm not sure i understand.

[Jesse Rosten]

That's next week... :-)

Actually, this clipped highlight thing is the single biggest issue facing RED ONE users. If it was all about the camera, that would be one thing. But so much footage is "reduced in quality" during grading that it deserves most of our attention. Again, it is the primary benefit of Build 16.

Jim

Maybe some more in-depth Redcine tutorials are in order? It'd be great to see examples of what not to do....and how to squeeze the best image out in Redcine. Remember, you guys are changing the rules when it comes to digital cinema. It'd be great if you'd also supply us with a rule book :)

[Mark Thorpe]

did you mean over expose? I'm not sure i understand.This is what I am a bit fuddled with too at the moment. I hear people using the term "Shooting to the Right" I always took that as being a direct reference to the histogram which would then tend for the image to be slightly over exposed. On the other hand I was not sure if they were talking about an RGB histogram or one indicative of exposure, same thing in a way I guess.

Anyone care to clear that up?

Cheers,
Mark.
P.S Even though my system was delivered a long while ago I haven't actually touched it yet. The unwrapping ceremony happens next week in San Diego. Looking forward to that.

[Jannard]

did you mean over expose? I'm not sure i understand.

Sorry... I did mean over-expose. I got twisted up in the double negative RAW thing and a short attention span.

Jim

[Dominic Cochran]

There are a few ways to make your RED ONE look like it has 7 stops of dynamic range... two primary ways.

1. Over-expose. Just clip the highlights while you record and there won't be any need to screw up anything in grading. This is the easiest way to turn a Digital Cinema Camera into an ordinary looking video camera (although you will still have more resolution).

2. Grade it improperly. Use ISO or exposure to set the mid-grey (or make most of the image look right) in RED Alert!, REDCINE, Scratch or Final Cut... without any regard for the histogram. Let the highlights go because the "middle of the image looks right". Don't use a curve to pull up the middle because that would hold the highlight you wish to blow out. Clipping is an art. If you can get really good at it, you may also learn what not to do "just in case you want the full info for a specialty shot".

As has been pointed out, the RED ONE does not have as much DNR as film. While most people think it is plenty to work with... it certainly does not if you throw away info.

Here's some good news. We are here to help. Build 16 makes it MUCH more difficult to miss exposure. And REDspace does a great job of helping to see your image without tempting you to make grading errors.

There are some of you that are saying "I don't do that. I know exactly what to do and how to get the most DNR with my RED ONE". OK, then. I'm not talking to you.

Here is an example of how to get an image to clip with one grading button.

I know... this grab is not that sharp. That isn't the point here...

Jim



I guess I'm playing Devil's Advocate here but don't you think it would be more constructive to make a post saying: "How to make your Red look like it has 11 stops of Dynamic Range," and not telling us all what what NOT to do?

Personally, my company has always done better by ignoring the naysayers rather than acknowledging them. It just seems more clear as well.

[jbeale]

Actually, the way Jim wrote it is more memorable, for me.

[Michele Gavazzeni]

With film, the tendency is to under expose a negative to keep the shadows from blocking. RED RAW footage is exactly the opposite... which I'm sure has contributed to some difficulties.

OK this means:
When you shoot film you usually rate the film stock at a lower ISO rate so you slightly over-expose it with the intent to keep the shadows from blocking.

When shoot Digital RAW you usually rate the chip at a higher ISO rate (RED is rated at 320 but you shoot it at higher ISO; someone says 500) so you under-expose your footage with the intent to keep the highlights from clipping.


Right?

[Dominic Cochran]

Actually, the way Jim wrote it is more memorable, for me.

Sarcasm is definitely more memorable, but rarely useful.

[Michele Gavazzeni]

did you mean over expose? I'm not sure i understand.

I hear people using the term "Shooting to the Right" I always took that as being a direct reference to the histogram which would then tend for the image to be slightly over exposed.
No never over-expose a shot
Always try to keep your histogram as much to the right as possible withough clipping the highlights.

[Mark Thorpe]

Thanks Michele, maybe I was just looking out at the world from the Histograms POV !!

Cheers,
Mark.

[Chris Nuzzaco]

One of the first things I noticed about Red Cine was how something like improper use of the ISO settings could permanently clip highlights (assuming you exported the file with those settings, the RAW data is fine).

I've also noticed this type of thing in FCP's Color. If you clip a highlight in the Primary in Room, you can't "resurrect" that highlight in the Primary Out room. I personally find that very annoying, as it makes the grading process more cumbersome. I don't have that issue grading in other programs like After Effects. I don't remember if Avid Adrenaline had this issue...

[Jan Reiff]

there is a lot of clipped 35mm footage out there, in commercials, features...

i am not worried about that with red. my philosophy always was to set a detailed light plan and not to shoot "guerilla" stuff. and to direct the grading in the post to avoid it (there are many operators that put too much contrast in the images IMO) i never had issues with clipped areas in 35mm, and i will never have it with the red footage.

in case of Steven Soderbergh (i don´t like the short "SS", not a short thing we like here you know) it´s maybe caused by a very closely time-table of shooting, 4:30 feature is a long way...

[gdv]

Do you mean I have to actually use the camera properly? that's crazy! :blink:

I've thought for a long time there is a need for a special DVD on how to properly use the camera. It should go out with every unit.

I've thought for a long time there is a need for a special DVD on how to properly use the camera. It should go out with every unit.[/QUOTE]

When I, a few days ago, after unboxing my new Reds, wrote on this forum that I find it strange that there is no manual, reference guide, quick starting guide, not even a piece of paper which say something "... you have just bought the most innovative camera..." etc. there were responses like "... what do we need it for..", "... it keeps changing anyway..." etc.
I find it too out of normal that a company deliver a camera with accessories just like that in boxes and the only written words about it was for the Red power Packs.
I think red is doing something wrong that maybe very painfull and costly in the future. There are ways of doing things.
i am still learning the camera and i already have a list of things that are not properly made if you think about this camera as something to work with for the next few years.
If the concept of Red, for the Red One camera, was to manufacture a tool that will be good for the next 12-24 months then this is something else.
Look at Arri, I don't see any panic there because of the Red versus the D-20 and D-21, because they were making cameras for some hundreds years and they are thinking long term. "So", they say, "now we have a new player in the field, which threatening our hegemony but what counts is what will be in five, ten, and twenty years from now".
Red it is not too late, you can still correct some mistakes which will pay off in the long run.

[Red Istanbul]

hey

I think RED ONE should be delivered with a set of DVDs that gives a detailed information about getting most out of the camera. It would be cool to see some professional DP's explaining how to deal with different situations or an editor expaining the details workflow.

best regards

[Clint Johnson]

Hi gdv... didn't anyone point out that the most up to date manual is available on the Red SUPPORT (http://www.red.com/support) page? It is zipped in with the Camera Firmware download- which you should have on your SD card just in case you need to re-install the build anyway.

[Hrvoje Simic]

Build 16 makes it MUCH more difficult to miss exposure.

"Highlight protector" built in, with three levels:

1. RedTold'ya mode - text and sound message "Burning!" in the camera and another message in RedCine : "Told 'ya!"
2. RedSlap mode- when highlights cross the limit hand comes out from inside the camera and gently warns the camera operator of reduced picture quality, by leaving a subtle red mark and pulsating feeling on the camera operator's face.
3. RedZapper mode - gently sends higher amounts of electricity into the hand which opens the iris too much


excerpt from a press release:

"Lake Forrest, CA - We have tested our new "Highlight protector" feature with great results. Our three levels of protection are extremely versatile and adaptable to various types of negligence. The system is programmed to automatically switch modes measuring the amount of mistakes made in a specified timeframe. That way each user gets the appropriate attention and guidance and learns to avoid future mistakes much faster than with usual solutions. "

[Jan Reiff]

sound with acoustic guitar please and deep voice

"and it burns burns burns..."

[Denis Buhot]

I sure don't want to pose as an expert - I'm not, by far - but doesn't all this seem very usual, familiar stuff to any one who's got used to a digital still pro camera and made the effort to understand what iso rating and histograms are ! Is'nt the Red one an incredibly fast recording still camera, in some sense ? How to expose "digitally" and use a histogram not to clip the highlights is all over the internet, I had thought... And good movies with clipped whites, terrible grain and dead blacks are all over the theaters, with nobody suggesting Spielberg should pay more attention to the cameras he uses, nor change DP for a competent one...

Jim and the Red team are being very patient, at times.

Indeed, the technical side - not to say the fuss - seems a bit "over exposed" to me here... Especially when, lacking a sleep mode, I shall not be given a chance to clip any highlights nor kill any blacks in the very shot I just couldn't shoot because of power shortage :-) *

However, who ever came to think Red one would make basic attention, lighting, filters, etc. dispensable ?... I thought I'd heard of a mattebox ? Whoever happened to shoot under forest cover at noon, not to say in the jungle...- should know the contrast is not an easy one to deal with if you don't want the extra hassle of proper lighting and just need the image to reflect the raw reality of what jungle had to be for the Che near la Higuera... I don't mean to be sarcastic nor go emotional, but blacks had to be very black, and whites awfully clipped, when he was lying wounded in his dark cell, and the door opened to the man who came to shoot him. And the sudden light must have blinded him for a while, and the bullets from that machine gun must have "lacked some detail", indeed... Maybe SS didn't care to much of a clean, soft and perfectly exposed image here, who knows... I wouldn't, for sure.

To me, what counts is that the red one is able to record a superb image in almost any conditions, at an affordable price, provided you don't expect to be just pressing the record button. What may seem boring to "spoilt" people in the 35mm film industry do look an incredible opportunity to people who ever dealt with any middle or even supposedly high end video camera.

Well, people like me still badly need a sleep-mode, but that's another story, and I don't want to put more strain on Jim's nerve...

[Graeme Nattress]

I just want access to the full sensor data that the camera promises. 12bits = max 12stops. 10bits = max 10stops.

And you get it! Your statement above is false in this context. As soon as the data is put in a non-linear form, you can encode > 12stops in a 10bit container. Just like scanning a film neg to 10bit dpx doesn't suddenly give you only 10 stops to work with.

If you don't want to mess with curves, use 16bit tiff output and linear light. Nobody is stopping you. However, unless you really know what you're doing, at least do white balance with our tools.

Graeme

[Brian Langeman]

I understand what Graeme is saying, and it really makes a lot of sense. I think you can output 16bit tiffs and you'll get what you're looking for with the 12bit sensor data coded linearly, because I don't think there are any 12bit standard formats.

But IF you have the ability to work with log files until your final output, even if you're not doing a film out, then that is the most efficient way to work. You save a lot of disk space, and also consequently rendering time, when working with 10bit files as opposed to 12bit or especially 16bit. You still have all of the dynamic range from the sensor, it's just coded differently, because basically too much data from the linear file is given to the highlights which is wasted because you will NEVER see a change from code values 4024 to 4025 even while grading. Converting it into a log file takes some of the info away from the highlights (which you will NEVER see anyways) and gives some more to the shadows, which is needed).

I recommend the book "Digital Compositing for Film and Video". Although a lot of you probably aren't compositors, the last few chapters describes this subject the best that I've seen so far.

Basically, you can work with 16bit tiffs if you want, but that's wasting a lot of disk space.

[conrad gaunt]

And you get it! Your statement above is false in this context. As soon as the data is put in a non-linear form, you can encode > 12stops in a 10bit container. Just like scanning a film neg to 10bit dpx doesn't suddenly give you only 10 stops to work with.

If you don't want to mess with curves, use 16bit tiff output and linear light. Nobody is stopping you. However, unless you really know what you're doing, at least do white balance with our tools.

Graeme

My concern is that, yes, values in the log code can cover the range that 12bit does (because its non linear, same DR roughly), but I`d still rather have 4096 code values going out of RC as apposed to 1024. Also, I don`t intend to scan any film, thats why I`m buying a 12bit digital cine camera, thanks to you Red guys, and your hard work. I won`t thank you too much because I know your having fun too :)
If the tiff output does get you the full sensor range, thats great, I didn`t know, and I will go down the tiff route. I will white balance with your tools when RC starts working for me, but I am developing freeware that has histograms and "stuff", just like the camera does, so not always. I only get to play with other peoples dpx posts at present. As for curves, I`ll be using them a lot, just not on the log data.

[conrad gaunt]

I understand what Graeme is saying, and it really makes a lot of sense. I think you can output 16bit tiffs and you'll get what you're looking for with the 12bit sensor data coded linearly, because I don't think there are any 12bit standard formats.

But IF you have the ability to work with log files until your final output, even if you're not doing a film out, then that is the most efficient way to work. You save a lot of disk space, and also consequently rendering time, when working with 10bit files as opposed to 12bit or especially 16bit. You still have all of the dynamic range from the sensor, it's just coded differently, because basically too much data from the linear file is given to the highlights which is wasted because you will NEVER see a change from code values 4024 to 4025 even while grading. Converting it into a log file takes some of the info away from the highlights (which you will NEVER see anyways) and gives some more to the shadows, which is needed).

I recommend the book "Digital Compositing for Film and Video". Although a lot of you probably aren't compositors, the last few chapters describes this subject the best that I've seen so far.

Basically, you can work with 16bit tiffs if you want, but that's wasting a lot of disk space.

I do too. My own software will do compositiing and many things. 12bit values save little space compared to 10bit and are no quicker to process. I can load/filter and save a dpx in about 1.7 secs on a single drive laptop (over two years old). If you use windows postly and ntsc drives, drive compression is your friend often, and the compressing/decompressing speeds things up (since disk reads are many times slower than decompressing in memory, and this has been true of disk compression since pentiums hit 120mhz!). If you keep the original sensor data (whatever the container 12bit or 16bit), they will compress roughly the same.
Converting 12bit log -> 10bit is slightly lossy, as Graeme admits, and is pointless if you don`t need log. As you`ll know, mixing log and linear isn`t great fun when compositing.

[Brian Langeman]

Also, I don`t intend to scan any film, thats why I`m buying a 12bit digital cine camera, thanks to you Red guys, and your hard work.

I'm not sure how much you know about the film scanning process Ronx. This isn't an attack or anything, I'm learning about it now too. But a lot of these issues with clipping and converting between lin and log would be solved if people did a little research on how people have been working with scanned film for the past while already.

Although the RED camera is different in how you would expose for film in some ways, how you treat the R3D file is VERY similar to how you would treat a film negative. So even if you don't plan on scanning film, it would still benefit to learn from the film's DI workflow that has already been stable for some time.

Maybe you're right about 10bit files not saving much space over 12bit ones. (I think that's what you were saying.) You probably know more about it than me, considering you're developing your own freeware. I would just assume it would save a lot of space over thousands and thousands of frames.

[Graeme Nattress]

And if there was a standard 12bit file format, we might have used it... But, the image processing we do in RedAlert, say, can look better if you keep it 16bit than even going back down to 12bit if you stay linear. It's very simple to apply a LUT to the 10bit data to get back to whatever the native bit depth of your compositing is though.

Graeme

[conrad gaunt]

I'm not sure how much you know about the film scanning process Ronx. This isn't an attack or anything, I'm learning about it now too. But a lot of these issues with clipping and converting between lin and log would be solved if people did a little research on how people have been working with scanned film for the past while already.

Although the RED camera is different in how you would expose for film in some ways, how you treat the R3D file is VERY similar to how you would treat a film negative. So even if you don't plan on scanning film, it would still benefit to learn from the film's DI workflow that has already been stable for some time.

Maybe you're right about 10bit files not saving much space over 12bit ones. (I think that's what you were saying.) You probably know more about it than me, considering you're developing your own freeware. I would just assume it would save a lot of space over thousands and thousands of frames.


sorry, I didn`t take it as an attack, I just have no use for established film processing techniques as I`m not coming from film or going to film, I`m processing DATA (image), and this is how I look at it. If its usefull I use it, if not, I don`t. If you preserve all the information in the data, throughout your processing, you`ll get good results. A final grade can be made later (in log if need be), but I`m truly digital in nature so traditional film processing routes don`t all apply here. Of course, Red have to supply these traditional log/film pathways for all the people who do need them, which may be most customers?, who are in the film business and who are buying R1s. CGI is never log, and I think your book will suggest converting the log footage to lin before mixing them up. Anyway the important thing is the camera produces great images :)

[Brian Langeman]

Didn't think you took it that way, just being cautious.

I feel I might be repeating something here, but I think it's pretty important the way people think about processing the DATA from their R3D files. Of course lots of things from the film workflow don't apply to the RED's, and some are even backwards. But some things are exactly the same. One conversation going on is about if using light meters is valid with the RED or with Video due to zebras and things like that. I would argue that for consistency and for good results all the time a light meter is just as useful while working with the RED as when working with film.

Now back on topic. If people were to think just a little bit harder with what they're doing in RC, these clipping problems that Jim is addressing would be gone. Do people scan film as 8bit linear files and clip the superwhites? No, because they want to keep the dynamic range. Therefore, while working in RC if you think about your job as the same thing as the film scanners or telecine's job, then you will end up with better results.

[conrad gaunt]

agreed, I use a light meter and will use the histograms in camera when it arrives as well as histograms in post too. Ill be investing in some books on colouremitry(!).. and a dictionary too! :)

[Paul Hazlett]

"Highlight protector" built in, with three levels:

1. RedTold'ya mode - text and sound message "Burning!" in the camera and another message in RedCine : "Told 'ya!"
2. RedSlap mode- when highlights cross the limit hand comes out from inside the camera and gently warns the camera operator of reduced picture quality, by leaving a subtle red mark and pulsating feeling on the camera operator's face.
3. RedZapper mode - gently sends higher amounts of electricity into the hand which opens the iris too much


excerpt from a press release:

"Lake Forrest, CA - We have tested our new "Highlight protector" feature with great results. Our three levels of protection are extremely versatile and adaptable to various types of negligence. The system is programmed to automatically switch modes measuring the amount of mistakes made in a specified timeframe. That way each user gets the appropriate attention and guidance and learns to avoid future mistakes much faster than with usual solutions. "

very funny, sign me up!!

[Sam Roberts]

Yeah those of us coming up from video where the only grading we did was a few tweaks to fix a color balance problem now and then could really use some kind grading tutorial using RedAlert or Red Cine or better, both.

[Jason Diamond]

yeah i believe the answer is very simple:

HIRE IAN BLOOM!

hahahaa

[Dj Joofa]

but here is a question: If noise can't be compressed, and any compression will necessarily change the character of the noise, than perhaps it makes sense to change the character of the noise first ourselves, prior to compression, so that it compresses in a more attractive way.

We all have our theories.

IBloom

Ian, there are two fundamental approaches in signal processing. Actually one. The other one is borrowed from control systems. The original signal processing just looks at the data and tries to do meaningful things with it, say take a Fourier transform or a Wavlet transform. The other, control systems approach, tries to build a model of the why that particular data was generated using but not limited to state space approaches.

A typical compression works on the former approach. However, noise modeling is extensively done using the later approach.

Let me give you an example. MP3 type audio compression uses perceptual coding that is a former approach as it looks at frequency data and tries to remove certain frequency band that are not audible. It does not model the source that how that sound data was generated by your vocal cords. On the other hand there are approaches that do model the source (vocal cords) and have certain advantages when information from the recorded data is missing.

[laguun]

Reds raw workflow is a real paradigm shift - and it is, lets make no mistake about that, here to stay.

it *can* be confusing in the beginning for people who so far only used negative film, especially taking into account that one might rate the "sweet" asa spot different depending on light temperature.

The DNR of red is good to get *outstanding* result, shown, proven and experienced by ourself countless times. I, no matter if using hdcam, film or red, always try to light for slightly less than the maximum of the camera. However, DNR is addictive. If Epic would offer a HDR DNR (by using 2 sensors etc) i would be much more tempted to upgrade our reds to epic - the resolution already is much better than 35mm can transport it to audiences anyhow.

[Ed Watkins]

Just a quick question:
What is REDspace? Is it a new piece of software, like REDcine? If so, will we have to purchase it? Or, is it the name given to a new user interface in build 16 firmware?


Thanks.

[Jens Jakob Thorsen]

JJ wrote:"With film, the tendency is to under expose a negative to keep the shadows from blocking".
Well no. Not in any way shape or form. Quite the opposite in fact. In my 20yrs as a DP I have allways been taught to be aware of one thing: the too thin negative. Epecially for telescine too much underexposure is deadly. Overexposure on the other hand is easy to deal with and no one will notice as happened to me when my AC forgot to tell me he pulled the Pola on a beach shoot.
The other way round would have been bad, using the 5245.
This is so much the case that tru all the 90ties I allways overexposed 1/2 stop for the telescine transfer.
Everything is more refined now but its still the truth. I like underexposing as much a the next guy now, but if u want to play it a little safer you overexpose your neg.
Shooting with RED I found that the "old" JJ saying: "Protect your highlights"
meaning expose as much as u can while maintaining the highlights of your shot, still is the mantra and the way to go with RED.
If you do it right in camera and grade in 2K DPX, you can get beautiful results.
I know I have.
As we all know(or should know) its not 35mm film and we dont have as much to play with in the grading stage. So the new school of cinematography as per Cersar Charlone(City of god) among others, of just shooting and fixing it later is NOT the way to shoot RED or any other digital media.
Best JJ
EDIT:And he did indeed correct this error of over/under. My mistake for have an equally short attention span and not reading the entire thread before posting. never do that.
BTW Im JJ, Jim is at most J.J(my mistake again; not to be repeated by ANYONE)
Best
Jens Jakob Thorsen DP

[Michele Gavazzeni]

JJ wrote:"With film, the tendency is to under expose a negative to keep the shadows from blocking".
Well no. Not in any way shape or form.

I was trying to discuss this before but the post has been hijacked to a over discussed 12 or 10 bit log or lin ...etc

Trying to interpret Jim post...

When you shoot film you usually rate the film stock at a lower ISO rate so you slightly over-expose it with the intent to keep the shadows from blocking.

When shoot Digital RAW you usually rate the chip at a higher ISO rate (RED is rated at 320 but you shoot it at higher ISO; someone says 500) so you under-expose your footage with the intent to keep the highlights from clipping.

What do you think?

[George A.]

It was just a slip from Jim - he meant exactly the opposite from what he said - the late hour etc. It happens - but we all got what he meant:

Treat RED exactly opposite from film.

[Derek Wiesehahn]

It's never a good idea to intentionally underexpose with a digital camera, especially when a camera is especially susceptible to shadow noise as the Red One is (with build 15). In my opinion, the best way to expose the Red is to choose the ASA you need (the lower the better), and then set your iris to a stop where the whites are just below the clipping point, using the zebra or histogram. Then add shadow fill with lighting if possible.
As Jim mentioned, one can then use curves or secondary correction to adjust midtone and shadow levels in post, without bringing the highlights to the clip point.

So what is Redspace?

[Michele Gavazzeni]

ASA you need (the lower the better)

Red experienced owner suggest exactly the opposite.

[Sean]

Just saw Flight of the Red Balloon (dir: Hou Hsaio-Hsien). Never seen so much clipped and over-exposed 35mm film in my life. I didn't think it was pretty, because now I'm attuned to such things. Nevertheless...

"A work of art on the order of a poem by Yeats or a painting by Rothko." John Anderson, Washington Post

"Hou's latest film feels to me like a masterpiece." Michael Phillips, Chicago Tribune

"Flight of the Red Balloon succeeds magnificently." Philip Marchand, Globe & Mail

Seriously, never seen so many blown out images ever. Hm...

[Joseph Mastantuono]

"There's no accounting for taste"

Personally, I've encountered a bit of the opposite from DP's.

There's such a absolute fear of clipping that there's a inclination to shut down a bit too far.

[Jason Sinclair]

Yes well then that certainly makes everything clear.... Thanks for the response. Your camera is still cool.

[Jason Sinclair]

I press f in redcine and i see a cool dynamic range with all color processing off. Is there anyway we can just expose, view and export just to this data the sensor sees? Do we have that choice?

[jpp]

Just saw Flight of the Red Balloon (dir: Hou Hsaio-Hsien). Never seen so much clipped and over-exposed 35mm film in my life. I didn't think it was pretty, because now I'm attuned to such things. Nevertheless...

"A work of art on the order of a poem by Yeats or a painting by Rothko." John Anderson, Washington Post

"Hou's latest film feels to me like a masterpiece." Michael Phillips, Chicago Tribune

"Flight of the Red Balloon succeeds magnificently." Philip Marchand, Globe & Mail

Seriously, never seen so many blown out images ever. Hm...

For at least the first 1/2 hour, I thought the damn thing must have been shot on HDCAM for precisely that reason.... A lot of Hou's 35mm work does show clipping, probably thanks to budgetary limits and his interest in street shooting in broad daylight with small crews.

His DP is among the best working today, but you do really have to wonder what those critics were drinking when they saw this film. The shot of the balloon over Paris is lovely, but that's about 1% of the movie. And "Balloon" is probably Hou's weakest film ever. Down with critics.

[Jason Murphy]

OK, just a quick point here.

Blowing out 35mm film IS NOT the same as clipping on a digital sensor. The problem with clipping isn't just that you irrevocably lose information in the highlights, it's also that there is a distinct clip line around the clipped area, and frequently you have some channels clipping before others, leading to certain colors in the clipped area. With film, there's a roll off into the blown out area that makes it much more visually pleasing. Furthermore even on blown out film, there's information in the blown out area that allows you to recover parts of that area. Not so on the digital side.

As an aside, and I know it boils down to a matter of taste, but I think that 'Flight of the Red Balloon' was both lovely visually, and a great film. Didn't at all feel like it was shot on HDCAM, at least to my eye. Mark Li Ping-bin really likes to work with blown out areas at times, though (and not just with Hou; Springtime in a Small Town and Vertical Ray of the Sun both have significant uses of blowout in the image).

(As far as it being a weaker film, I have to disagree as well, but with Hou, I'd also have to go back to 'Daughter of the Nile' to find a film of his that I thought was relatively weak, and even that movie is interesting).

[Graeme Nattress]

Even with film, if it's blown, it's blown. Yes, there's a roll-off, but that can also be the same with digital. Under-expose by one stop, and you can insert a curve that rolls over that stop. The more you under-expose, the more smoothly you can roll off.

Graeme

[Karl H]

not playing devils advocate here, but has build 16 actually increased dynamic range at all, or are you saying that it just helps us to get the most out of the same DR we already had?

[Graeme Nattress]

Take a look at the images Jim posted and tell me what you think...

Graeme

[Antoine Fabi]

Even with film, if it's blown, it's blown. Yes, there's a roll-off, but that can also be the same with digital. Under-expose by one stop, and you can insert a curve that rolls over that stop. The more you under-expose, the more smoothly you can roll off.

Graeme

Yoooo! voilà!

...and using the "exposure" control in RedCine to lessen the very high luma values on the histogram works great before you add that nice S curve.

[Karl H]

Take a look at the images Jim posted and tell me what you think...

Graeme


Well, it's hard to say because I havent seen how much the noise has improved. If it's marginal in normal daylight conditions then I guess the DR won't have imporoved. But maybe it has in tugsten situations.

I guess you can't alter the point at which the image clips but you can drag more image out of the shadows. So you tell me :)

[Jannard]

not playing devils advocate here, but has build 16 actually increased dynamic range at all, or are you saying that it just helps us to get the most out of the same DR we already had?

DNR is dependent on the noise floor. If the noise is reduced, DNR is increased. How much depends on the circumstances.

Jim

[Fredrik Callinggard]

Take a look at the images Jim posted and tell me what you think...

Graeme

A bit soft :biggrin:


Sorry couldn't help it :)

[Michele Gavazzeni]

Take a look at the images Jim posted and tell me what you think...
Graeme

I think Jim haven't notice while shooting that both John Force's cap and his team assistant where clipping.

when back in Redcine:

1st image the image is processed the same way it was shoot
2nd image he has processed the image at a slightly lower exposure and applied a small curve to pull back the middle tones
Now the white cap is in the more pleasant 233 233 233 area but because it was clipped when shot the cap has no details.

[Jörgen Persson]

I think Jim haven't notice while shooting that both John Force's cap and his team assistant where clipping.

when back in Redcine:

1st image the image is processed the same way it was shoot
2nd image he has processed the image at a slightly lower exposure and applied a small curve to pull back the middle tones
Now the white cap is in the more pleasant 233 233 233 area but because it was clipped when shot the cap has no details.

That's not true. The cap has more detail. And the brightest values are in the specular highlights of the assistant's headphones.

See for yourself: Use levels in Photoshop and drag the white point triangle while holding down the "alt" key.

[Michele Gavazzeni]

The cap has more detail.
yes you're right but its barely noticeable.

if you open the 1st image in PS you apply a curve and you adjust the white point down a bit, the midtones up and the black point moving it at the right,the difference isn't that much

for sure you'll be affecting also the specular light on the assistant earphone
but that joust because are mapped at the same level in the jpg file.

http://homepage.mac.com/michelegavazzeni/Sites/imm/9_forcea_edit.jpg

[Daniel Browning]

yes you're right but its barely noticeable.

if you open the 1st image in PS you apply a curve and you adjust the white point down a bit, the midtones up and the black point moving it at the right,the difference isn't that much


If you can't see the difference then your display is not calibrated correctly.

[Michele Gavazzeni]

A simple what is what.... joudge only with your eyes.
My monitor is ok maybe my eyes are a bit off.


http://homepage.mac.com/michelegavazzeni/Sites/imm/9_forcea copy.jpg

[Jannard]

You can get the blown area to look less white, but you cannot recapture the detail that is lost. If you get it right in grading, you don't have to "try to make it look almost as good". It will never be.

Jim

[Brook Willard]

Here's a more extreme example to add to the mix. In this shot, about 75% of the circle you see around the sun was clipped. The first shot is the RAW data as it came across in the metadata. The second shot was graded by a friend of mine who is quite familiar with SCRATCH, but is not familiar enough with RED footage to help him. The file was baked out RAW and recovered as much as possible. The third image was my own personal take on the RAW file with highlight recovery my priority [the shadows are a little darker, but I could've brought those up without changing the highlights... I just didn't notice that they were dark until I made this post].

http://homepage.mac.com/brookwillard/raw.jpg

While the whites in the JPEG file may be clipped [or near clipped], the only points that are actually clipped with my grade are the points that were clipped in camera. The sky may look more "blown out" in the third image, but the transition is natural and smooth as opposed to dramatic and harsh. The shirt may look dangerous, but it hasn't actually clipped. In a proper projection, all shades would be present.

This is another point to keep an eye out for. There are times when you will need to kiss parts of your image goodbye. It's the in-depth knowledge of the way the RAW file will handle light that allows me to expose my images properly. This image isn't a shining example, but the point is this: You learn to translate what you see on the camera's monitor to what you see in post and what your desired grade is. Then expose accordingly.

If you exposed this image and hoped to keep every last pixel of sky blue until the clip point, you will process an ugly image very similar to the 2nd in this series. It's possible - all of the detail is there right up until the clip point - but it's a bad idea. You must expose images with the knowledge and foresight to know what you're going to do with it in post.

If I wanted blue sky up until the point where the sky clips [so I wanted it blue until the hard circle starts, then I wanted it to fade to white in the middle], I would've stopped down a stop or two. That would've made the "clipped" circle much smaller, but it would've allowed me to roll off the highlights in a smaller circle.

This particular mentality only applies to the sky. Every other kind of highlight has its own mentality... speculars, skin, clothing, cars, foliage, whatever. Everything must be interpreted just a little differently on the day. If you learn the way that the RAW file handles light and the way that you will handle the RAW file in post, you can learn to expose the camera in a way that will eventually get you to the result that you have in your head.

[Evin Grant]

BTW if you're wondering why the sunny highlight area is so large in those images it's because in Dubai there is often "Dust" as they call it but it's really a mixture of dust and fog that adds a high layer of diffusion to the sun when looking straight up.

[Jannard]

Brook... great post.

Jim

[Jeff Kilgroe]

OK, I hate to be the one that shoots off on a tangent, but... That's a nice fisheye, looks like a 6mm or thereabouts. Is it the Nikon (or Century conversion) or something else? What did you use for a reducer to get the entire FOV on frame?

[Fredrik Callinggard]

I think it's great that you guys show this and it's very enlightening to see and I hate to be the devils advocate, but it's just that I don't understand what this has to do with DR? In the post Jim says that there's plenty of ways in grading to make it look like it has 7 stops of DR.

I understand that smoothing white clippings out makes it not look like video, which is very important and is something that needs to be understood more, but what has that to do with actual DR?


Fredrik Callinggard

[Steve White]

If there's detail in the highlights, the camera has the dynamic range. It isn't a look - it's actual information that's really there.

The problem is that people are throwing that information out when they grade. The RAW data the RED captures is not clipped - but because of where the white-point is set in the grading process, people think the information is gone, without ever putting forth the effort to see if it's there.

Trouble is, if RED sets the white-point higher, people will complain of "flat" footage. If they set it low, people will say it's "clippy". RED could define a set of gamma curves to produce an "optimal" image, but this would defy the principle benefit of the RAW workflow. I suspect Build 16 will include some tools to help facilitate this.

The bottom line is that RED footage needs to be graded by someone who understands what look they want, and what data is available.

[Fredrik Callinggard]

Yes but isn't this example more about grading it properly so it doesn't clip in an ugly way and therefor end up looking like video?

To me there's two problems that can occur, ugly white clippings and this is a great thread showing that this can (a lot of the times) be prevented. The other issue is blue channel noise in shadows, which can be worked around in post with taking the blue channel and compare it with the green and.... etc

To me though these examples here are just showing how to prevent from getting that horrible "line" to white clipping, by giving it a smooth transition and therefor look nicer. It does not really show me DR.

But maybe I'm wrong......

[Brook Willard]

OK, I hate to be the one that shoots off on a tangent, but... That's a nice fisheye, looks like a 6mm or thereabouts. Is it the Nikon (or Century conversion) or something else? What did you use for a reducer to get the entire FOV on frame?

http://www.reduser.net/forum/uploaded/5_1211761866.jpg

It's an Evin Grant special. They don't call him a Ninja for nothing.

[Jannard]

Yes but isn't this example more about grading it properly so it doesn't clip in an ugly way and therefor end up looking like video?

To me there's two problems that can occur, ugly white clippings and this is a great thread showing that this can (a lot of the times) be prevented. The other issue is blue channel noise in shadows, which can be worked around in post with taking the blue channel and compare it with the green and.... etc

To me though these examples here are just showing how to prevent from getting that horrible "line" to white clipping, by giving it a smooth transition and therefor look nicer. It does not really show me DR.

But maybe I'm wrong......
No... it is about both... throwing away info you have and then making it roll to clip (when it actually does) more nicely. Two separate issues. The 1st is DNR, the second is just nicer.

Jim

[jbeale]

yes you're right but its barely noticeable.

You can certainly see some non-clipped detail if you look for it... there's little fine detail, because (I believe) the shot is not crisply focused.

http://www.bealecorner.com/D30/misc/9_forceb-not-clipped-crop.jpg

[Jarred Land]

http://www.reduser.net/forum/uploaded/5_1211761866.jpg

It's an Evin Grant special. They don't call him a Ninja for nothing.

wow thats pretty cool.. I want one :)

[Brook Willard]

Pictures just don't do it justice. It looked like the monster from Cloverfield.

http://homepage.mac.com/brookwillard/6mm.jpg

Alright, I'm done leading this thread astray. Sorry guys.

[Jeff Kilgroe]

bad ass. It can walk, right? :)

[DNomer]

Brook!

WTF IS that thing? For the thread, I know how to use histograms, and I am not going to blow out my images. But that gizmo!? I get that you have some structural items, an EVF and a Red One, but what is that flying saucer thing on top and what is the lens piece right below it?

dn

[Mike Prevette]

One of the only lenses they actually had to "grow" from crystal.

for those interested in how to reduce the image size for the Red: http://www.nearfield.com/~dan/Photo/wide/fish/

[Harrison Diamond]

Another solution ( that might be less cumbersome) would involve the 85mm f/1.0 Repro-Nikkor in a relay system, though there you're dealing with another lens not much more common than the 6/2.8 and on the expensive side.

[Christoffer Glans]

Pictures just don't do it justice. It looked like the monster from Cloverfield.

http://homepage.mac.com/brookwillard/6mm.jpg

Alright, I'm done leading this thread astray. Sorry guys.


That thing gives me the creeps, it feels like it's watching, like some kind of mobile HAL9000 or something. I feel a bit weird when optics start to feel like a real eye watching. :blink:

[Lucerna Studios]

i want to buy this camera to do sweet 16 and weddings!!