Comparing the Cinema Camera: Part 2 The Impact of 12-bit RAW

The place for questions about shooting with Blackmagic Cameras.
  • Author
  • Message
Offline
User avatar

Marco Solorio

  • Posts: 118
  • Joined: Wed Aug 22, 2012 3:40 am
  • Location: Alamo, California

Comparing the Cinema Camera: Part 2 The Impact of 12-bit RAW

PostSat Oct 27, 2012 8:21 pm

HOT OFF THE PRESS! Here's the new follow-up video on Vimeo to our last Cinema Camera comparison video. Get your popcorn ready! This one goes deeeeeeep!

And a HUGE thanks to all those that watched and made comments on the first video! Always appreciated!



Cheers!
Offline
User avatar

Jason R. Johnston

  • Posts: 1615
  • Joined: Wed Aug 22, 2012 7:05 am
  • Location: Nashville TN USA

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSat Oct 27, 2012 8:26 pm

Thanks, Marco!

As much as I just really want to yell at most of the people on this forum, it's posts like this that keep me here. And the BMD guys, of course.
JASONRJOHNSTON.COM | CINEMATOGRAPHER | DIRECTOR | EDITOR | COLORIST
RED Komodo | DaVinci Resolve Studio 18.5 | 2023 MacBook M2 Pro 14
Offline

Felix Steinhardt

  • Posts: 205
  • Joined: Sat Aug 25, 2012 9:45 pm
  • Location: Karlsruhe / Germany

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSat Oct 27, 2012 8:40 pm

Thanks a lot for the video! :) The day for night grading was amazing!

But unfortunately you made a mistake at the point were you created the 8-bit file.
RAW in ACR isn´t shown in log and it doesn´t have to be. It´s interpreted to bright, too. A non RAW File has a look baked in. Therefore the BMCC shoots log in Prores just like the Sony F3 or Arri Alexa.

You just exported the interpretation of what ACR sees in the DNG which is too bright and very contrasty. If you converted the 12 bit RAW into an 8 bit log or loglike (like cLog) file you would have those highlight infos in the 8 bit file. But of course it would be tough to grade this because of the limited bitrate ->Banding when you strech it back out.
Offline
User avatar

Abobakr M. Alshiblie

  • Posts: 104
  • Joined: Mon Oct 22, 2012 6:35 pm

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSat Oct 27, 2012 9:05 pm

Thank thank thank you... That was so creative and I loved it.. Thanks a much <3
Offline

Costa Louvieris

  • Posts: 343
  • Joined: Wed Aug 29, 2012 8:06 am

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSat Oct 27, 2012 9:50 pm

Abobakr wrote:Thank thank thank you... That was so creative and I loved it.. Thanks a much <3


Thanks Marco. You are quickly becoming my fave reviewer/tester. You are very knowledgable and explain things so well that even my limited brain can understand them! Loved the day to night stuff too. Awesome.
Offline

Bill Rich

  • Posts: 292
  • Joined: Fri Sep 07, 2012 10:19 pm
  • Location: Los Angeles, California

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSat Oct 27, 2012 10:01 pm

Great test Marco! It really shows the power of 12bit raw!!
Bill Rich
PhotoJournalist/Editor/Producer
Los Angeles, California
Offline

John Waldorff

  • Posts: 197
  • Joined: Wed Aug 22, 2012 10:19 pm

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSat Oct 27, 2012 10:02 pm

Like!
Offline

John Waldorff

  • Posts: 197
  • Joined: Wed Aug 22, 2012 10:19 pm

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSat Oct 27, 2012 10:22 pm

>You just exported the interpretation of what ACR sees in the DNG which is too bright and very contrasty. >If you converted the 12 bit RAW into an 8 bit log or loglike (like cLog) file you would have those highlight >infos in the 8 bit file. But of course it would be tough to grade this because of the limited bitrate ->Banding when you strech it back out.

It is worth pointing that out.
But it is not a problem in the test.
The produced 8-bit file corresponds not the prores output because that was not the point.
It corresponds a possible 8-bit camera shot and shows the limitations to bring a badly exposed shot back to life.
With this way of doing the file it can be clearly demonstrated.

Sure, if shot in Prores (or a perfectly exposed shot with a 5dmkIII) there would have been little less overexposure and therefore a little more information in the out-of-the-window shot.
But the point is still clear.

Cheers
Offline

CaptainHook

Blackmagic Design

  • Posts: 2054
  • Joined: Wed Aug 22, 2012 4:50 am
  • Location: Melbourne, Australia
  • Real Name: Hook

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSat Oct 27, 2012 10:48 pm

John Waldorff wrote:It corresponds a possible 8-bit camera shot and shows the limitations to bring a badly exposed shot back to life.

But what about an 8-bit camera shot in log or some kind of flat profile? I wonder if Marco would upload that DNG of the room for people to do their own tests?

And yes, thanks Marco! :)
Last edited by CaptainHook on Sat Oct 27, 2012 11:49 pm, edited 1 time in total.
**Any post by me prior to Aug 2014 was before i started working for Blackmagic**
Offline

Fabián Matas

  • Posts: 130
  • Joined: Wed Aug 22, 2012 9:35 am

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSat Oct 27, 2012 11:42 pm

Nice work Marco, as allways :)
Offline

John Waldorff

  • Posts: 197
  • Joined: Wed Aug 22, 2012 10:19 pm

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSun Oct 28, 2012 12:20 am

CaptainHook wrote:But what about an 8-bit camera shot in log or some kind of flat profile? I wonder if Marco would upload that DNG of the room for people to do their own tests?


A flat profile will help to avoid clipping of color information when grading. But it does in no way convey additional detail.
It is not a compressed histogram horizontally to put more dynamic range into the 8bit range.
Instead it makes the levels of the colors just a bit are lower and therefore have a little more headroom when pushing them into any direction.
So it would not alter the result or change the outcome of the test.

Surely dng is always nice, I want it too + the approach you describe is something that is really essential for shooting with DSLR and when shooting prores as well.
I merely wanted to point out there is no error in the actual test.

Cheers
Offline

bhook

  • Posts: 1024
  • Joined: Wed Aug 22, 2012 8:19 pm

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSun Oct 28, 2012 12:57 am

Very nice Marco! Thank you very much.
Offline

CaptainHook

Blackmagic Design

  • Posts: 2054
  • Joined: Wed Aug 22, 2012 4:50 am
  • Location: Melbourne, Australia
  • Real Name: Hook

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSun Oct 28, 2012 1:29 am

John Waldorff wrote:A flat profile will help to avoid clipping of color information when grading. But it does in no way convey additional detail.
It is not a compressed histogram horizontally to put more dynamic range into the 8bit range.
Instead it makes the levels of the colors just a bit are lower and therefore have a little more headroom when pushing them into any direction.

I think there's some confusion here... Mostly i was suggesting a log transfer curve (or similar) just like Felix but also, a flat profile can help give more 'usable' dynamic range but at the expense of colour/tonal precision in a highly compressed 8bit codec (with more possibility of banding with smooth gradients etc), i.e the other way around to what you've said.

From Technicolor's FAQ on Cinestyle:
"The Technicolor CineStyle provides better dynamic range of the captured content."
**Any post by me prior to Aug 2014 was before i started working for Blackmagic**
Offline
User avatar

Peter J. DeCrescenzo

  • Posts: 2407
  • Joined: Wed Aug 22, 2012 6:53 am
  • Location: Portland, Oregon USA

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSun Oct 28, 2012 1:58 am

Hi Marco: Thanks so much for creating this video and sharing it with us. Great information, well-presented, and an opportunity for all to discuss and learn.

I wonder what the "cool girls" will have to say about it? See:
http://www.herefortheweather.wordpress. ... -becoming/

Cheers!

-
Offline

Andrew Hunter

  • Posts: 203
  • Joined: Wed Aug 22, 2012 2:30 pm
  • Location: Toronto, Canada

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSun Oct 28, 2012 4:07 am

CaptainHook wrote:
John Waldorff wrote:It corresponds a possible 8-bit camera shot and shows the limitations to bring a badly exposed shot back to life.

But what about an 8-bit camera shot in log or some kind of flat profile? I wonder if Marco would upload that DNG of the room for people to do their own tests?

And yes, thanks Marco! :)


A log curve is used to compress higher bit depth values into a lower bit depth container.

In an interview, Grant Petty referred to the log curve applied to the raw sensor data. While you may go "but it's raw!", the sensors ADC units probably operate at 14 bits and use a log curve to compress those values into a 12 bit container.

I also want to point out that there is a fallacy inherent in saying that higher bit depths mean greater latitude. In practice, this is often the result but it is not tightly coupled.

For example, one could have an 8 bit, 12 bit or even 1 bit integer code value representation of the exact same scene luminance values. Each has the same dynamic range relative to scene min and max values. What is extremely different is the ability of an observer to perceive the changes between code values and banding artefacts are the result.

In terms of it's practical applications, the average human observer's tolerance to just noticeable differences (JND) dictates the trade off between scene refered dynamic range and bit depth. In other words, higher bit depths allow for a greater dynamic range without introducing visual artefacts.

Once floating point numbers are introduced, the range of scene luminance values one is able to encode in a given bitdepth increases greatly.
Offline

rawCAM35

  • Posts: 30
  • Joined: Wed Aug 22, 2012 4:31 pm

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSun Oct 28, 2012 7:09 am

Thanks Marco for your generosity of doing this and sharing it with everyone, short and to the point.
Offline
User avatar

Michael Sandiford

  • Posts: 308
  • Joined: Wed Aug 22, 2012 8:48 am
  • Location: Hereford

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSun Oct 28, 2012 7:25 am

Brilliant Marco.

Now this is how it's done. Take note Mytherapy model shootout bods.

This was entertaining to watch, informative, and Marco got to market himself and his company and I don't mind one bit. Brilliant Marco, Brilliant.

Thank you.
Last edited by Michael Sandiford on Sun Oct 28, 2012 8:28 am, edited 1 time in total.
Offline

Dennis Nomer

  • Posts: 113
  • Joined: Fri Aug 24, 2012 11:54 pm

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSun Oct 28, 2012 7:55 am

Marco delivers again. Very nice intro to the idea of grading within a 12-bit log space. For many of us (including me) this is relatively new territory, and we need to wrap our minds around it. Us shooters are a 'show me' crowd, and that is exactly what you do.
Dennis Nomer
Offline

Jesuan Soriano

  • Posts: 245
  • Joined: Wed Aug 22, 2012 4:17 am

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSun Oct 28, 2012 11:37 am

Gotta love you Marco!!!! :)

Jesus
Jesus Soriano
Offline
User avatar

rick.lang

  • Posts: 17156
  • Joined: Wed Aug 22, 2012 5:41 pm
  • Location: Victoria BC Canada

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSun Oct 28, 2012 3:40 pm

Andrew Hunter wrote:
CaptainHook wrote:
John Waldorff wrote:It corresponds a possible 8-bit camera shot and shows the limitations to bring a badly exposed shot back to life.

But what about an 8-bit camera shot in log or some kind of flat profile? I wonder if Marco would upload that DNG of the room for people to do their own tests?

And yes, thanks Marco! :)


A log curve is used to compress higher bit depth values into a lower bit depth container.

In an interview, Grant Petty referred to the log curve applied to the raw sensor data. While you may go "but it's raw!", the sensors ADC units probably operate at 14 bits and use a log curve to compress those values into a 12 bit container.

I also want to point out that there is a fallacy inherent in saying that higher bit depths mean greater latitude. In practice, this is often the result but it is not tightly coupled.

For example, one could have an 8 bit, 12 bit or even 1 bit integer code value representation of the exact same scene luminance values. Each has the same dynamic range relative to scene min and max values. What is extremely different is the ability of an observer to perceive the changes between code values and banding artefacts are the result.

In terms of it's practical applications, the average human observer's tolerance to just noticeable differences (JND) dictates the trade off between scene refered dynamic range and bit depth. In other words, higher bit depths allow for a greater dynamic range without introducing visual artefacts.

Once floating point numbers are introduced, the range of scene luminance values one is able to encode in a given bitdepth increases greatly.


The BMCC sensor captures 16bit linear data output by the camera as uncompressed raw 12bit log or 10bit ProRes or 10 bit DNxHD. All options will support better quality than 8bit anything.
Rick Lang
Offline

Theodore Prentice

  • Posts: 591
  • Joined: Mon Sep 03, 2012 3:56 pm

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSun Oct 28, 2012 4:09 pm

rick.lang wrote:All options will support better quality than 8bit anything.


The latest "camera spec war" theme... :roll:

Comparing 12 bit raw to 8 bit anything is cool academically speaking, but in practice the argument is moot considering it is the perceived DR that the client/audience sees, understands, and interacts with.

Put another way, the specs of a Ferrari become a moot point against a turbocharged Honda civic given that the perceived power of the average driver in both cars from 0-100mph is the same.
The actual comparison of the available power of the Ferrari vs the souped-up Civic is indeed silly, as there really is NO contest, but try telling that to the person that just equally enjoyed the 0-100mph experience in both cars.

Marco, thanks for taking the time to do the comparison video!
Offline
User avatar

rick.lang

  • Posts: 17156
  • Joined: Wed Aug 22, 2012 5:41 pm
  • Location: Victoria BC Canada

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSun Oct 28, 2012 4:14 pm

Marco Solorio wrote:HOT OFF THE PRESS! Here's the new follow-up video on Vimeo to our last Cinema Camera comparison video. Get your popcorn ready! This one goes deeeeeeep!

And a HUGE thanks to all those that watched and made comments on the first video! Always appreciated!



Cheers!


Such an elegant overview of the benefits of shooting raw 12bit. An excellent tutorial to illustrate your point.

It's unfortunate you weren't able to compare to an Alexa, an Epic, a Scarlet, a GoPro Hero3, a C100, a C300, a C500, and especially the new Sony Fk (since we all know Sony gave you a new Fk camera last year)! I don't know how you can sleep at night with these obvious slanted perceptions of the world according to principles of limited time and resources. Surely that fat pay check you receive monthly from BMD should help you put a little more effort into your free contributions to the never-satisfied-and-smarter-than-everyone-else BMCC community here!

Enough said, I certainly don't want to hijack your thread in any way because I am truly very grateful for what you have done here. And grateful that you will continue you meager efforts in the future even though they likely will be criticized and shown to be inferior to what anyone else here could post if they only had a BMCC and your aforementioned gravy train.

I can understand it though if you are too ashamed to ever post a video here again. You seasoned pros are so alike demanding respect and all that just because you have the camera before the King of England got his! After reading thousands of posts here and elsewhere, I've learned two things: first, people that post in forums stand head-and-shoulders above people that contribute original content that provokes all those posts. And secondly, you, John, Dan, Frank, Philip, and the others... sure know how to keep your perspective, if you know what I mean. And a special thanks to Jonathan Swift for his solution to the Irish potato famine. Whew! It just had to be said and now I'll shuddup knowing that it won't make one iota of difference to the quality of posts here. May your gods strike me senseless for adding to the quantity of posts.
Rick Lang
Offline

Costa Louvieris

  • Posts: 343
  • Joined: Wed Aug 29, 2012 8:06 am

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSun Oct 28, 2012 4:58 pm

rick.lang wrote:
Marco Solorio wrote:HOT OFF THE PRESS! Here's the new follow-up video on Vimeo to our last Cinema Camera comparison video. Get your popcorn ready! This one goes deeeeeeep!

And a HUGE thanks to all those that watched and made comments on the first video! Always appreciated!



Cheers!


Such an elegant overview of the benefits of shooting raw 12bit. An excellent tutorial to illustrate your point.

It's unfortunate you weren't able to compare to an Alexa, an Epic, a Scarlet, a GoPro Hero3, a C100, a C300, a C500, and especially the new Sony Fk (since we all know Sony gave you a new Fk camera last year)! I don't know how you can sleep at night with these obvious slanted perceptions of the world according to principles of limited time and resources. Surely that fat pay check you receive monthly from BMD should help you put a little more effort into your free contributions to the never-satisfied-and-smarter-than-everyone-else BMCC community here!

Enough said, I certainly don't want to hijack your thread in any way because I am truly very grateful for what you have done here. And grateful that you will continue you meager efforts in the future even though they likely will be criticized and shown to be inferior to what anyone else here could post if they only had a BMCC and your aforementioned gravy train.

I can understand it though if you are too ashamed to ever post a video here again. You seasoned pros are so alike demanding respect and all that just because you have the camera before the King of England got his! After reading thousands of posts here and elsewhere, I've learned two things: first, people that post in forums stand head-and-shoulders above people that contribute original content that provokes all those posts. And secondly, you, John, Dan, Frank, Philip, and the others... sure know how to keep your perspective, if you know what I mean. And a special thanks to Jonathan Swift for his solution to the Irish potato famine. Whew! It just had to be said and now I'll shuddup knowing that it won't make one iota of difference to the quality of posts here. May your gods strike me senseless for adding to the quantity of posts.


Wow you have a lot of hate in your heart man. Let it out and then Let it go.
Offline
User avatar

Peter J. DeCrescenzo

  • Posts: 2407
  • Joined: Wed Aug 22, 2012 6:53 am
  • Location: Portland, Oregon USA

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSun Oct 28, 2012 5:04 pm

rick.lang wrote:
Marco Solorio wrote:HOT OFF THE PRESS! Here's the new follow-up video on Vimeo to our last Cinema Camera comparison video. Get your popcorn ready! This one goes deeeeeeep!

And a HUGE thanks to all those that watched and made comments on the first video! Always appreciated!



Cheers!


Such an elegant overview of the benefits of shooting raw 12bit. An excellent tutorial to illustrate your point.

It's unfortunate you weren't able to compare to an Alexa, an Epic, a Scarlet, a GoPro Hero3, a C100, a C300, a C500, and especially the new Sony Fk (since we all know Sony gave you a new Fk camera last year)! I don't know how you can sleep at night with these obvious slanted perceptions of the world according to principles of limited time and resources. Surely that fat pay check you receive monthly from BMD should help you put a little more effort into your free contributions to the never-satisfied-and-smarter-than-everyone-else BMCC community here!

Enough said, I certainly don't want to hijack your thread in any way because I am truly very grateful for what you have done here. And grateful that you will continue you meager efforts in the future even though they likely will be criticized and shown to be inferior to what anyone else here could post if they only had a BMCC and your aforementioned gravy train.

I can understand it though if you are too ashamed to ever post a video here again. You seasoned pros are so alike demanding respect and all that just because you have the camera before the King of England got his! After reading thousands of posts here and elsewhere, I've learned two things: first, people that post in forums stand head-and-shoulders above people that contribute original content that provokes all those posts. And secondly, you, John, Dan, Frank, Philip, and the others... sure know how to keep your perspective, if you know what I mean. And a special thanks to Jonathan Swift for his solution to the Irish potato famine. Whew! It just had to be said and now I'll shuddup knowing that it won't make one iota of difference to the quality of posts here. May your gods strike me senseless for adding to the quantity of posts.


Commentary? Humor? Sarcasm? Beer?

I've re-read it a few times and I can't tell which it is, regrets.

Cheers.
Offline
User avatar

dennysb

  • Posts: 81
  • Joined: Tue Oct 02, 2012 5:28 am
  • Location: Hudson, WI - USA

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSun Oct 28, 2012 6:32 pm

Thank you Marco for sharing all these information, it sure makes it a bit easier on the waiting for production units to arrive in "actual" qty.

Thanks!!!!
Offline
User avatar

dennysb

  • Posts: 81
  • Joined: Tue Oct 02, 2012 5:28 am
  • Location: Hudson, WI - USA

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSun Oct 28, 2012 7:09 pm

rick.lang wrote:
Marco Solorio wrote:HOT OFF THE PRESS! Here's the new follow-up video on Vimeo to our last Cinema Camera comparison video. Get your popcorn ready! This one goes deeeeeeep!

And a HUGE thanks to all those that watched and made comments on the first video! Always appreciated!



Cheers!


Such an elegant overview of the benefits of shooting raw 12bit. An excellent tutorial to illustrate your point.

It's unfortunate you weren't able to compare to an Alexa, an Epic, a Scarlet, a GoPro Hero3, a C100, a C300, a C500, and especially the new Sony Fk (since we all know Sony gave you a new Fk camera last year)! I don't know how you can sleep at night with these obvious slanted perceptions of the world according to principles of limited time and resources. Surely that fat pay check you receive monthly from BMD should help you put a little more effort into your free contributions to the never-satisfied-and-smarter-than-everyone-else BMCC community here!

Enough said, I certainly don't want to hijack your thread in any way because I am truly very grateful for what you have done here. And grateful that you will continue you meager efforts in the future even though they likely will be criticized and shown to be inferior to what anyone else here could post if they only had a BMCC and your aforementioned gravy train.

I can understand it though if you are too ashamed to ever post a video here again. You seasoned pros are so alike demanding respect and all that just because you have the camera before the King of England got his! After reading thousands of posts here and elsewhere, I've learned two things: first, people that post in forums stand head-and-shoulders above people that contribute original content that provokes all those posts. And secondly, you, John, Dan, Frank, Philip, and the others... sure know how to keep your perspective, if you know what I mean. And a special thanks to Jonathan Swift for his solution to the Irish potato famine. Whew! It just had to be said and now I'll shuddup knowing that it won't make one iota of difference to the quality of posts here. May your gods strike me senseless for adding to the quantity of posts.


** I am taking this to the U-Minnesota/Mayo Clinic Psychology department for research study on Bi-polar disorders :D
Offline
User avatar

Marco Solorio

  • Posts: 118
  • Joined: Wed Aug 22, 2012 3:40 am
  • Location: Alamo, California

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostWed Oct 31, 2012 2:07 am

THANK YOU to everyone that posted such wonderful comments! I've read ALL of them!

I've been getting some comments (mostly on Vimeo itself) as to why I didn't use a LOG curve on the 8-bit source footage (along with some other related questions from some nice people, and some not so nice people!). As such, I decided to spell it out in a blog post that'll hopefully quell some of the questions or implications of error on my part. I'm actually quite surprised there are some people that think they can obtain a fully rendered image (like the one in my *graded* 12-bit version of the interior room with exterior sky) from an 8-bit camera. I'd love to know what 8-bit camera they're able to achieve that kind of latitude on! :D

Anyway, if your interested in my "Part 2.1" blog post, here it is:

http://www.onerivermedia.com/blog/?p=750

As always, enjoy! And thanks again!
Offline

Andrew Deme

  • Posts: 501
  • Joined: Wed Aug 22, 2012 9:52 am

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostWed Oct 31, 2012 3:31 am

Excellent vid...!

Reminds me of the day I saw what a RAW stills photo looked like in Lightroom compared to JPEG.
Offline
User avatar

dennysb

  • Posts: 81
  • Joined: Tue Oct 02, 2012 5:28 am
  • Location: Hudson, WI - USA

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostWed Oct 31, 2012 4:27 am

Marco Solorio wrote:THANK YOU to everyone that posted such wonderful comments! I've read ALL of them!

I've been getting some comments (mostly on Vimeo itself) as to why I didn't use a LOG curve on the 8-bit source footage (along with some other related questions from some nice people, and some not so nice people!). As such, I decided to spell it out in a blog post that'll hopefully quell some of the questions or implications of error on my part. I'm actually quite surprised there are some people that think they can obtain a fully rendered image (like the one in my *graded* 12-bit version of the interior room with exterior sky) from an 8-bit camera. I'd love to know what 8-bit camera they're able to achieve that kind of latitude on! :D

Anyway, if your interested in my "Part 2.1" blog post, here it is:

http://www.onerivermedia.com/blog/?p=750

As always, enjoy! And thanks again!


Marco, very important technical question ................ where can I buy the shirt at the end of the above blog posting :)
Offline

CaptainHook

Blackmagic Design

  • Posts: 2054
  • Joined: Wed Aug 22, 2012 4:50 am
  • Location: Melbourne, Australia
  • Real Name: Hook

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostWed Oct 31, 2012 6:34 am

Marco Solorio wrote:http://www.onerivermedia.com/blog/?p=750

Thanks Marco, would you post the DNG frame so we can do tests as well? Always a good idea to test yourself if possible, then you can feel the limits first hand. :)

Also, in your post show you some testing with an inverted "s" curve, but not a "log" (or log-type) curve. Thanks! :)
**Any post by me prior to Aug 2014 was before i started working for Blackmagic**
Offline

Bill Rich

  • Posts: 292
  • Joined: Fri Sep 07, 2012 10:19 pm
  • Location: Los Angeles, California

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostWed Oct 31, 2012 2:29 pm

Great Blog! I also love the shirt!
Bill Rich
PhotoJournalist/Editor/Producer
Los Angeles, California
Offline

rawCAM35

  • Posts: 30
  • Joined: Wed Aug 22, 2012 4:31 pm

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostWed Oct 31, 2012 5:30 pm

rick.lang wrote: It's unfortunate you weren't able to compare to an Alexa, an Epic, a Scarlet, a GoPro Hero3, a C100, a C300, a C500,
Surely that fat pay check you receive monthly from BMD should help you put a little more effort into your free contributions to the never-satisfied-and-smarter-than-everyone-else BMCC community here!


Try to flex some muscles and get Alexa, an Epic, a Scarlet, a GoPro Hero3, a C100, a C300, a C500 and test it yourself, since other people do not have the resources or the time to do it.

I have no idea that you have access to Marco Solorio's bank accounts and you know for fact that he is getting paid by BMD ???

Thanks to you Marco and others for your contributions small and large to this community
Offline

Theodore Prentice

  • Posts: 591
  • Joined: Mon Sep 03, 2012 3:56 pm

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostWed Oct 31, 2012 9:08 pm

lol rawcam35, I fully believe rick lang was being totally sarcastic.

Im sort of lost on the "fake" youtube example in the video marco, with all sorts of examples on the web, it is quite hard to tell the difference between something that was lit, exposed and shot correctly on an hdslr, versus a Red Epic. lol.
In no way am I disputing the difference between the 12 bit raw, vs 8 bit compressed (as I said in my earlier post, there is really no comparison) , just the notion that web delivery isnt crap, or somehow made better by the bmcc.

I still think the 8bit uncompressed 5d footage will give the 10 bit prores bmcc footage a run for its $ all other things being equal. (*disclaimer- yes, yes, I know, not the point of this test..AND im totally willilng to eat crow if wrong :mrgreen: )
Offline

Jules Bushell

  • Posts: 1026
  • Joined: Wed Sep 05, 2012 3:06 am
  • Location: London, England

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostWed Oct 31, 2012 10:02 pm

Theodore Prentice wrote:I still think the 8bit uncompressed 5d footage will give the 10 bit prores bmcc footage a run for its $ all other things being equal.


Hi Theodore,

I'm still not sure if you are getting the point of the video? I didn't understand, until this thread but willing student all the way.

If you take, e.g. Canon C300 which is an 8 bit camera, and shoot, it can look gorgeous straight out of the camera; as clearly evidenced by the many vimeo/youtube clips.

Now if you want to grade then 8 bit still can't compare with 10 bit. The problem is the number of bits. With 8 bits you have only 256 values per colour to cram in all that picture information from the very brightest to the darkest. 10 bits there are 1024 values to play with

If you store huge dynamic range (keeping lots darkness and brightness) then you compromise with less precision in the colour. That is the jump between one colour value to the next shade of the colour will be large. So when grading in 8 bit, with colours that should be different, they can get lumped together as one colour and is seen as the 'banding' problem.

So from the information I have understood so far, when grading... 8 bit RAW is not as good as 10 bit ProRes).


Jules
Jules Bushell
url: www.nonmultiplexcinema.com
url: www.filmmeansbusiness.com
url: www.blurtheline.co.uk
Offline
User avatar

rick.lang

  • Posts: 17156
  • Joined: Wed Aug 22, 2012 5:41 pm
  • Location: Victoria BC Canada

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostWed Oct 31, 2012 10:37 pm

Theodore Prentice wrote:lol rawcam35, I fully believe rick lang was being totally sarcastic.


Thanks, Theodore! It was a satirical response to those people that are never satisfied by what others post in the spirit of trying to be helpful. Several hints like the assumption that I would have any clue about anyone's bank account, access to unreleased Sony cameras, and so on. Trying to make the satire very clear referring to the King of England when there's no such person. And finally a reference to the world's best known satire of all time (Jonathan Swift's "A Modest Proposal" to solve the Irish potato famine by eating the children), should have been a dead giveaway.

Way over the top and yet people were upset with my post even though my track record of posting messages is hopefully always trying to be considerate and helpful. Apologies to those who took it seriously.

Just having some fun... and the Mayo Clinic visit went very well... I had to promise never to try that stunt again though before they let me out.
Rick Lang
Offline
User avatar

Nick Bedford

  • Posts: 352
  • Joined: Wed Aug 22, 2012 3:56 am
  • Location: Brisbane, Australia

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostWed Oct 31, 2012 11:25 pm

Theodore Prentice wrote:I still think the 8bit uncompressed 5d footage will give the 10 bit prores bmcc footage a run for its $ all other things being equal. (*disclaimer- yes, yes, I know, not the point of this test..AND im totally willilng to eat crow if wrong :mrgreen: )


I'm really skeptical as to this "uncompressed HDMI", mainly because I expect Canon to soften up the image before sending it through to the HDMI port. I have a 5DIII and frankly, the detail of the video mage has nothing on the BMCC and RED cameras, let alone the bit depth. They are simply sharp through the entire 1080p image. The 5DIII can barely save a 720p video image to save its life.

If Canon actually sends a PROPERLY scaled full sensor image then hell yes it will give the BMCC a run for its money, at least in shear sharpness stakes.
Nick Bedford, Photographer
http://www.nickbedford.com/
Offline

Theodore Prentice

  • Posts: 591
  • Joined: Mon Sep 03, 2012 3:56 pm

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostThu Nov 01, 2012 10:50 am

Jules Bushell wrote:
Theodore Prentice wrote:I still think the 8bit uncompressed 5d footage will give the 10 bit prores bmcc footage a run for its $ all other things being equal.


Hi Theodore,

I'm still not sure if you are getting the point of the video?


Im still sure that I have gotten the point of the test (even said so right after what you quoted above).
Offline

CaptainHook

Blackmagic Design

  • Posts: 2054
  • Joined: Wed Aug 22, 2012 4:50 am
  • Location: Melbourne, Australia
  • Real Name: Hook

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostThu Nov 08, 2012 4:09 am

I finally had time to do my own testing. I did this with a DNG raw frame from Evans DNG exposure test (i did ask for the frame Marco used but no response).

I tried to follow Marco's testing patterns except i used Resolve instead of Lightroom (i personally don't like the raw conversion from Lightroom and use Capture One 99% of the time for stills anyway). Here's what i did:

• Used frame one of Evan's exposure test since the outside "appears" clipped.
• Exported a jpg of this 'as is' for reference.
• Exported an 8Bit H264 version of the frame which came out at just over 37Mbps (less than Marco's 50Mbps).
• Lowered the exposure on the DNG raw file and did some quick curve and Lift/Gamma/Gain adjustments to represent something i might aim for (not looking to create an HDR type image since i'm not a fan). Created a jpg from this.
• Attempted to do the same with the 8bit H264 37Mbps version. Exported JPG.
• Then i reset the exposure/etc (Base Mem) on the DNG to start from scratch and applied the BMDFilm log preset. Exported as 8Bit H264 and this time the frame was at 12Mbps. MUCH less than 50Mbps which should be a disadvantage here.
• Attempted to do a similar recovery as i did for the DNG raw. Exported JPG.

So i have 4 JPGs to post. The reason i'm doing this is because i think Marco's demonstration was EXTREME and i believe he made a mistake by not using a log curve as well and in his follow up post by using an inverted "s curve" instead of a proper log-type curve (like so - http://mikemost.com/?p=251). I'm VERY aware of the advantages of 10bit log and 12bit raw but i still think GOOD things are possible with a nice 8bit log file (not on par with 10bit or 12bit raw, but still..).

DSLRs are currently the worst examples of 8bit, yes. But all 8bit != DSLRs. Given the choice, of course 12bit raw or 10bit log is highly preferable to 8bit, and part of the 'success' of my example is the very nice way BMCC handles it's image/dynamic range and the strength of the resolve output/compression. I'm just posting because i wanted to test this for myself (as we always should if we have time/resources) and to provide perhaps a bit of 'balance'.

Click on the JPGs for the large versions if desired:

The DNG raw with no adjustments in Resolve:

Image

The 8bit NO LOG H264 37Mbps recovery attempt (no denying it's awful):

Image

The DNG raw "recovery" in Resolve (to my preference what i would use as a starting point for grading):

Image

The 8Bit Log (BMDFilm) H264 12Mbps attempt (not as nice as the raw but respectable and useable IMHO):

Image

Hope some find that useful. I'm not a DSLR "apologist" and i don't think they're currently capable of this demonstration either, but i DO think 8bit Log didn't get a completely 'fair' run by Marco either. But i understand his enthusiasm for the BMCC, 12bit raw, and 10bit log since they ARE fantastic to work with in post. What i took away from this, is that a GOOD camera with a solid (uncompressed/un-lineskipped etc) 8bit log output is pretty good. Not 12bit/10bit log good, but not necessarily miles away either.
**Any post by me prior to Aug 2014 was before i started working for Blackmagic**
Offline

SylvainS

  • Posts: 7
  • Joined: Sun Nov 04, 2012 5:45 pm

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostThu Nov 08, 2012 12:24 pm

Thanks for this test ! :oops: Please anybody who has a BMCC, could you be kind to go out 5 min in the night and to shoot the sky at 1600 ISO, shutter 180° & 360°, at the maximum aperture your can, and to post the 12 bit raw file here (5 seconds are enough), or here : viewtopic.php?f=2&t=2333 ? Thank you !
Offline
User avatar

Jason Greene

  • Posts: 140
  • Joined: Mon Oct 01, 2012 12:03 pm
  • Location: Illinois

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostThu Nov 08, 2012 12:43 pm

CaptainHook - Thank you for doing this test. Frankly, I was surprised at all of the gushing compliments for Marco's video. I didn't find it to be very helpful. Is his conclusion correct that 12-bit raw is superior to 8-bit compressed? Yes. Absolutely. But, was his method correct? I didn't think so. Yours is.

What Marco's method would have done is blow out the sky, just as in your "with no raw adjustments". Had Marco done "recovery" as you did, I suspect that his window detail would have been recoverable. Again, would it have looked as good grading it from an 8-bit file rather than in 12-bit raw? No. But it seems to me that the ill effects from the 8-bit compressed file would have been banding in gradients and otherwise less color detail, not such significant loss of dynamic range. The loss of dynamic range came from his failure to recover the highlights prior to compressing into the 8-bit space. Is my understanding correct?
Offline

CaptainHook

Blackmagic Design

  • Posts: 2054
  • Joined: Wed Aug 22, 2012 4:50 am
  • Location: Melbourne, Australia
  • Real Name: Hook

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostThu Nov 08, 2012 8:20 pm

I had similar thoughts as you Jason, which is why i wanted to test myself. Log in an 8bit output can help with dynamic range recovery (providing the camera/sensor actually has it) but at the cost of tonal/colour information that will show up as banding etc especially more so if the grade is pushed.

I think a better stress test for this comparison would be pulling 'tough' keys or qualifiers and/or shifting colours/hue quite far. IMHO anyway.
**Any post by me prior to Aug 2014 was before i started working for Blackmagic**
Offline

melissa lancaster

  • Posts: 26
  • Joined: Sat Oct 27, 2012 11:53 pm

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostThu Nov 08, 2012 8:26 pm

What a fantastic comparison. Thank u so much! Very insightful. Anyway someone could show a side by side comparison of the 10 bit pro res and 12 bit raw files? I wanna see how big a difference there is there in case I want to shoot more in pro res to save room on hard drives. Thanks so much again!
Offline
User avatar

Nick Bedford

  • Posts: 352
  • Joined: Wed Aug 22, 2012 3:56 am
  • Location: Brisbane, Australia

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostThu Nov 08, 2012 11:39 pm

Technical his method is correct when comparing the impact of higher bit depth log (or 12-bit DNG) over 8-bit baked tone curve DSLR type footage where you don't actually have any option of capturing all the DR into a log image.

The DSLR will crush shadows and blow highlights before you can even recover them because the tone curve is applied to the image before it's saved.
Nick Bedford, Photographer
http://www.nickbedford.com/
Offline

CaptainHook

Blackmagic Design

  • Posts: 2054
  • Joined: Wed Aug 22, 2012 4:50 am
  • Location: Melbourne, Australia
  • Real Name: Hook

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostFri Nov 09, 2012 12:10 am

Nick ~ it's my understanding that picture profiles are applied to the raw data in canon dslrs then compressed/lineskipped/destroyed/etc going through liveview to disc, this is how cinestyle etc 'work'. A log curve is applied to raw data via the picture profile. So my example is on par with what should be possible as i applied a log curve to a raw file, the same as what is happening in the canon DSLRs. The difference, problem, and blame with DSLRs like the 5D etc lies mostly with canon, not 8bit. This is partly my point and shown in my experiment and by other cameras doing 8bit well. By coincidence i came across this today, shot on an F3:

http://www.hingsberg.com/index.php/2012 ... dop-notes/
To the surprise of many the footage was recorded directly in-camera to Sony SXS media which means I used the XDCAM codec @ 35mbps at 8-bit depth in 420 color space.


And screen shots he posted (which look pretty nice to me - ignoring his preference for green):

Image
Image
Image
Image

I'm not sure if he even used slog because he hasn't posted that information yet but either way i don't think he's been hurt by 8bit 420 at all in this example. But it will be interesting to see his follow up posts and if he struggled at all in post with the files. I keep saying this, but 8bit != DSLRs. There's many camera's doing 8bit pretty well. I'd prefer 10bit log or 12bit raw though obviously.
**Any post by me prior to Aug 2014 was before i started working for Blackmagic**
Offline
User avatar

Marco Solorio

  • Posts: 118
  • Joined: Wed Aug 22, 2012 3:40 am
  • Location: Alamo, California

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostFri Nov 09, 2012 9:31 am

CaptainHook, the point of the video and my follow-up blog post is simple:

1. There isn't an 8-bit camera in the world that can capture a dark interior room with only practical lighting, retaining its shadow detail, while also shooting through an open window into the bright day's sky, while retaining its color and cloud detail, all in the same shot. It's mathematically impossible. Blown out highlights in sky regions is a common occurrence with 8-bit cameras.

2. Once you have that 8-bit footage (whether created from a 12-bit source or from an 8-bit format camera), once you push levels in post (no matter what kind of pre-LOG curves you apply to it), the sky (especially) will turn to gray, rather than bringing out its true blue color and cloud detail. Again, it's mathematically impossible.

Now, in my blog post, I supplied four examples of lifted shadows and reduced highlights for the 8-bit source. Whether or not you like the curve or HDR-ish look is moot... the point was to show that from the 12-bit source, we'd make an 8-bit "balanced" source, even though the outcome is unrealistic to what a real 8-bit camera would provide in its limited space and DR. But even so, once we get that magic 8-bit source, then what? Whether it's my 8-bit source, or the 8-bit source you provided in your examples, once we start pushing levels hard in the grade (again, the sky is a good example), it all turns to mush.

I'd also like to point something out in the four images you provided (namely the last three), which perfectly coincides with my video. Yes, the four images look nice. This is perceived dynamic range. In their own right, they look good. Whether or not the original footage has deep latitude is not known, and is moot at the deployment stage. But if we wanted to step into available dynamic range, I'll take a hefty wager the last three images would fail. Why? Look at the sky portions of them (top of frame-right). All instances of the sky are white/grey, and the fourth image has clipped cloud data in it (the third image especially appears to have a blown out "bloomy" appearance in the sky area). It's obvious they added a grad and/or vignette in post to help manage the brightness of the sky areas (not a bad thing, as I do that too when it helps a shot). But the grad causes the sky to be gray (and doesn't fix the clipped bloom in image 3), with no original blue from the sky. Had these been shot with any 12-bit RAW camera, those sky portions of the image would have produced a lot more deep color fidelity (and cloud detail) in them. The result? What we see in these images, is exactly what I show in my video when 8-bit exposure levels are lowered.
Offline

CaptainHook

Blackmagic Design

  • Posts: 2054
  • Joined: Wed Aug 22, 2012 4:50 am
  • Location: Melbourne, Australia
  • Real Name: Hook

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostFri Nov 09, 2012 11:05 am

Thanks for elaborating Marco, would you post the DNG raw frame you used so i could try my experiment on it to directly compare? Like i said many times, i don't dispute the advantage of 12bit raw over 8bit, just that i still believe a proper log curve would yield a better result on that image than the inverted "s" curve and would love to know if that's correct or not. I admit using a different DNG frame to compare to yours isn't ideal but i was confident it showed what i initially thought was true. Perhaps not. Again, if i could try on your DNG frame it would help me out.

I wont comment further on the F3 footage since i only really posted it to show 8bit 420 looking pretty nice after being graded (many people seem to think it's not possible), rather than to show dynamic range recovery since i have no idea if he even attempted that or what the circumstances were. And you're right that we don't directly know enough about that particular footage to really comment further.

Thanks. :)
**Any post by me prior to Aug 2014 was before i started working for Blackmagic**
Offline

CaptainHook

Blackmagic Design

  • Posts: 2054
  • Joined: Wed Aug 22, 2012 4:50 am
  • Location: Melbourne, Australia
  • Real Name: Hook

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostFri Nov 09, 2012 11:40 am

Marco Solorio wrote:1. There isn't an 8-bit camera in the world that can capture a dark interior room with only practical lighting, retaining its shadow detail, while also shooting through an open window into the bright day's sky, while retaining its color and cloud detail, all in the same shot..
I'm not posting this to be argumentative, but really just for anyone who hasn't seen it since it's pretty cool. :D There are other shots showing banding in the sky (typical 8bit tell) but this is sweet for an alpha release of ML. If the canon 422 uncompressed out in April is decent, this could be great for certain shots (without a lot of movement):

http://www.youtube.com/watch?feature=pl ... SfX8#t=85s
**Any post by me prior to Aug 2014 was before i started working for Blackmagic**
Offline

Seaborn

  • Posts: 73
  • Joined: Thu Aug 23, 2012 3:59 pm

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostSat Nov 10, 2012 3:36 am

http://www.youtube.com/watch?feature=player_detailpage&v=fjtGenTSfX8#t=85s


really nice! if i cant get a bmc before the end of the year, i probally get a 5d mk3 and wait for bmc v2.

there some really clean dslr hack, maybe will see a bmc hack one day!
Seaborn studio
https://www.studioseaborn.com
Offline
User avatar

Cesar Rubio

  • Posts: 13
  • Joined: Sat Jan 19, 2013 6:34 am

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostMon Feb 04, 2013 10:01 am

Hey guys:

Please check this test made with an Uncompressed RGB 8 bit file:

http://dna-rubio-s3d.blogspot.mx/2013/0 ... 2-bit.html

What do you think?

Thanks,
Cesar Rubio
It's the long term experience of problem solving what makes you successful! & what we do in life, echoes in eternity.

http://www.dna-rubio-s3d.com/
Offline

danap

  • Posts: 43
  • Joined: Tue Sep 25, 2012 1:13 pm

Re: Comparing the Cinema Camera: Part 2 The Impact of 12-bit

PostMon Feb 04, 2013 10:36 am

Nick Bedford wrote:
Theodore Prentice wrote:I still think the 8bit uncompressed 5d footage will give the 10 bit prores bmcc footage a run for its $ all other things being equal. (*disclaimer- yes, yes, I know, not the point of this test..AND im totally willilng to eat crow if wrong :mrgreen: )



After 2 1/5 with my BMCC and 1 1/5 months with my Canon 5DMkIII - I agree totally. For me there where is a BEFORE and an AFTER the BMCC. And nothing kind of 8 bits HDMI uncompressed will make any - any - difference. And given the arrogant, cynical and – in my honest opinion – suicidal approach chosen by Canon to "kill" resolution in the 5DMarkIII in order to protect their soon to collapse EOS Cinema business, I hope they will hear the bells ringing, goes into some kind of introspection, fire some marketing VIPs, and unleash the locked potential of their camera. But I do not believe it will happen.

The scepticism and animosity of video shooters towards Canon is just rising.
Next

Return to Cinematography

Who is online

Users browsing this forum: AlwaysWritePat, Bing [Bot] and 37 guests