Canon 5D Mark III - 3.5K and 4K raw video with Magic Lantern

[Andrew Reid]

The thing is Magic Lantern's option for 8bit RAW at ISO 6400 makes perfect sense as the sensor isn't delivering more than 8 stops dynamic range at that ISO any way

So to cut the data rate and save some card space in low light it is a good option.

[Shirozina]

3 minutes ago, Andrew Reid said:

The thing is Magic Lantern's option for 8bit RAW at ISO 6400 makes perfect sense as the sensor isn't delivering more than 8 stops dynamic range at that ISO any way

So to cut the data rate and save some card space in low light it is a good option.

Genius!

[tweak]

I think 8-bit Raw would be pretty crappy to be honest.

[Andrew Reid]

Choose 10bit or 12bit then

This isn't North Korea.

[Mattias Burling]

36 minutes ago, Andrew Reid said:

Choose 10bit or 12bit then

This isn't North Korea.

Actually...

[Andrew Reid]

All the choices in camera as long as it's Arri!

How long as he been shooting film for? Since he was kidnapped?

[Mattias Burling]

10 minutes ago, Andrew Reid said:

How long as he been shooting film for? Since he was kidnapped?

Totally worth it

[maxotics]

People talk about only "8 stops of dynamic range" as if it's unusable.  I love following Andrew's and everyone else's analysis of how to maximize various camera's potential, but we sometimes let things get out of perspective.  The order of important in my experience is 1. Lighting 2. Focus/stabilization 3. Lens 4. Sensor 5. Ergonomics 6. CODEC.   Yes, we can't control the lighting all the time, but we also have to compare cameras in how they do in properly lit scenes; that is, scenes at/under 8 stops of DR.  Here's my take on why we must never forget that shooting scene with more than 8 stops of dynamic range is something we really want to avoid.  No camera can truly save it.  It becomes a calculus of atrocities.  

 

[Andrew Reid]

Well, getting back onto the subject of the 10bit and 12bit lossless 3.5K raw on the 5D Mark III, I can safely say it's the best QUALITY dynamic range I've ever used in post. I can correct for huge exposure differences and there's no crushing of the shadows / blacks at all.

So I agree in that I'd rather have a high quality 10 stops or 12 rather than a compressed to hell 14 stops that's for sure (*cough cough* S-LOG 3 *cough*)

[Shirozina]

33 minutes ago, maxotics said:

People talk about only "8 stops of dynamic range" as if it's unusable.  I love following Andrew's and everyone else's analysis of how to maximize various camera's potential, but we sometimes let things get out of perspective.  The order of important in my experience is 1. Lighting 2. Focus/stabilization 3. Lens 4. Sensor 5. Ergonomics 6. CODEC.   Yes, we can't control the lighting all the time, but we also have to compare cameras in how they do in properly lit scenes; that is, scenes at/under 8 stops of DR.  Here's my take on why we must never forget that shooting scene with more than 8 stops of dynamic range is something we really want to avoid.  No camera can truly save it.  It becomes a calculus of atrocities.  

 

5-8 stop DR of the human eye is just plain wrong. Everything else therefore doesn't add up.

[Mattias Burling]

What the eye can do doesn't really matter. Because the brain can modify it any way it pleases.

Our eye has variable dynamic range and HDR. We can expose for highlights, mids, and shadows all at one and the brain shows us a perfect 20 stop image.
The brain is super fast and adaptable. If I remove one of my contacts I never have the chance to see the decrease in resolution and sharpness. My brain is way to fast in applying the proper corrections.

So trying to mix DR from a camera with that of screens is already a bit weird. I know some say that 13 stops isn't needed if a screen is only 10.. just wrong. With the 13 in the camera I can adjust more info into the 10 for the screen. Simple as that.

To add the eye into the mix is just silly imo.

[Andrew Reid]

It's great to start off with as much dynamic range as possible because post is important... you can do so much... and even if your grade is only going to deliver a punchy, contrasty 8 stops, your highlight and shadow rolloff is much smoother if you shoot with more stops than you finally need.

To make crushed-footage look like it has more dynamic range, many people raise the black level to a milky grey like in a lot of Philip Bloom's drone videos...

With the 5D Mk II 3.5K you never find yourself needing to do this... the shadows already look filmic, milky and creamy with lots of tonality and yet you get to keep true black when and where it's needed.

It's an even better image than the GH5 in 10bit, it is up there with an Alexa or RED.

Also the live-view hack is not bad... It is nice and sharp for focus and the framing is 100% bang-on and easy to do... They just need to work on increasing the frame rate.

[tweak]

There's a few threads on ML forums where 8-bit Raw is discussed, mostly coming to the conclusion that it isn't a very good idea.

[maxotics]

1 hour ago, Mattias Burling said:

What the eye can do doesn't really matter. Because the brain can modify it any way it pleases.

Our eye has variable dynamic range and HDR. We can expose for highlights, mids, and shadows all at one and the brain shows us a perfect 20 stop image.
The brain is super fast and adaptable. If I remove one of my contacts I never have the chance to see the decrease in resolution and sharpness. My brain is way to fast in applying the proper corrections.

So trying to mix DR from a camera with that of screens is already a bit weird. I know some say that 13 stops isn't needed if a screen is only 10.. just wrong. With the 13 in the camera I can adjust more info into the 10 for the screen. Simple as that.

To add the eye into the mix is just silly imo.

Why is "adding the eye" into the mix silly?  It's the only way I can see photographs.  How do you see them?

Our eye, like our camera, does not have variable dynamic range.  It is around 6 stops, that's what they've measured. however, in low light the eye can do amazing things in those 6 stops and the same with bright light; HOWEVER, the pupil must resize for it do it.  The brain doesn't so much modify what the eye sees as in create a real-time composite of all the information the eye has looked at.  

Sure, in real life we can take wide DR for granted.  How often do we change brightness levels, say go from inside to out, or outside to in.  In VIDEO however, there are quick cuts and to change brightness to the point where the eye would need to keep adjusting its pupil every few seconds would create a very disorienting experience.  Stand in your doorway and look inside, then outside on a bright day, back and forth and you tell me how good your brain really is at switching.  It's not because, believe it or not, the pupil doesn't change size instantaneously.

I'm not trying to mix DR from a camera with that of screen.  Sorry, but I feel you're really not paying attention to what I'm explaining and why I'm doing it.  You're looking to disagree with what I'm saying.  Fine.  Assume you can always mix your 13 in the camera to the 10 for the screen (don't know where you get that EV spread).  Most of your photographs I see are processed to very high contrast.  Such post processing hides many defects of the subtle DR you have to work with.    

And if you believe the quality of your videos has remained the same since you stopped shooting with a RAW based camera, well, you should explain why Andrew just wasted his time with that guide

(I LOVE your stuff Mattias, just having a little fun here!  Get a 5D3, Andrew's guide, and come back to real digital film (RAW)   Which by the way, is not about DR, it's about the fact that RAW doesn't chroma-sub-sample and degrade the image.  Noise is organic.  It doesn't contain the various artifacts of a digital processor)

[Bioskop.Inc]

What ML has done with Canon cameras is simply amazing - they've taken a camera with great colour science, a really shitty delivery codec & given it a greatly needed shot in the arm!

But, and it is a big but, RAW can be a real pain in the arse - with my set up, I need to transcode the DNG files with RPP (it's great as it can batch stuff, has 3 different LOG settings & loads of JPEG/TIFF settings), then I need to wrap them in a QT PR .mov file & then i can edit them.

Personally, H264 is dead to me & all i want is a sturdy codec with which to play with in post - PR422 HQ or above is more than good enough for most things. RAW is only useful when you come across a difficult scene that needs some extra care/attention in post - DR to you guys.

However, if I was given the choice between the 5D3's internal codec & RAW (regardless of bits), then I'd certainly choose RAW everytime - so some of the ML flavours might not have as much DR, but come on, you don't or won't need that amount of DR all the time. I know it might be a bit old fashioned, but the rule of thumb should be to get it right, as close as possible, in-camera & then all you'll need to do is tweek in later - personally if you're spending more time in post than actually filming, you're doing it wrong.

What Andrew shot (see vid above) really shows the strengths of what ML has brought to the 5D3 table - if you can't see it, then i feel really sorry for you. It isn't an ideal situation, but if you have a 5D3 you should be jumping up & down with joy - so many choices are now available to you!

If you're still using a camera with an H264 delivery mode, I'd be pulling what's left of my hair out!

[Shirozina]

13 minutes ago, maxotics said:

 

Our eye, like our camera, does not have variable dynamic range.  It is around 6 stops, that's what they've measured.

Where and who has measured this because I'm telling you it's just factually wrong. You make some valid points in other areas but your whole argument falls down if you keep on insisting that our eyes are limited to a very narrow DR and therefore any more than this in a capture or display device is not needed.

[maxotics]

1 minute ago, Shirozina said:

Where and who has measured this because I'm telling you it's just factually wrong. You make some valid points in other areas but your whole argument falls down if you keep on insisting that our eyes are limited to a very narrow DR and therefore any more than this in a capture or display device is not needed.

Scroll down to "Dynamic Range"  https://en.wikipedia.org/wiki/Human_eye  

[Shirozina]

9 minutes ago, maxotics said:

Scroll down to "Dynamic Range"  https://en.wikipedia.org/wiki/Human_eye  

Wikipedia is clearly wrong then as anybody with a pair of working eyes can easily prove.

[maxotics]

3 minutes ago, Shirozina said:

Wikipedia is clearly wrong then as anybody with a pair of working eyes can easily prove.

I know, those scientists, right!!!!  Global warming isn't real either.  You're confusing apparent DR AFTER the brains composition of an image with what the eye can resolve at any given moment.  Mattias is making the argument "who gives a fudge, as long as we end up with 20 stops".  And I'm answering, you'd give a fudge if you ever tried to look at dramatic changes in DR.  

[Shirozina]

1 minute ago, maxotics said:

I know, those scientists, right!!!!  Global warming isn't real either.  

How about doing some proper research and getting your data from several sources -  just like a scientist?