More HDRx...

[Bill Anderson]

Bring it on.

 

[D Fuller]

David... agree with you (again). I think half the battle will be deciding when to use HDRx™™™ and when not to. The variable in the equation is that EPIC will have even less noise and FPN (when pushed to the extreme) than you have ever seen.

It is great to have lots of tools. This time, they will all be sitting inside the same camera and you get to decide which and when you will choose what to use.

That's a great observation. Like any new tool -- swing/shift lenses, the mesmerizer -- HDRx™ will go through the growing period where we all use it too much as we try to incorporate it into the toolkit we know how to control. But what a tool!

 

[JD Holloway]

Is variable sensor sensitivity up and running?

I posted a couple of times earlier but I think it got buried by RED 1 HDR talk

 

[Jack Jones]

RED is really pushing the boundaries like nothing else.

No better time to be a Colourist, as our job becomes a more integrated and vital part of the post production process.

Colourists are becoming a mix of onliners, finishers, post prod. supervisors, depth adjusters (SS3D) as well as our more traditional function of image enhancement.

I fully recommend DOP's choosing a colourist, they trust, to work with the images they shoot to realise the full look they were shooting for. If you, as a DOP, have time to get a days 'basic grade training' in with that colourist, all the better as you'll really enhance and understand the limitations and more importantly the possibilities of the grade.

Colourist, likewise, should get onset more often and learn from DOPs about onset lighting and in-camera functions.

 

[Shane Betts]

Well put Jack. Thanks very much for your viewpoint. And welcome aboard!

 

[Michael Stanmore]

I've just discovered that the guys who were with Jim getting this Barn shot... are Aussies... so that pretty much kills the test results because it's common knowledge that Australians are awesome.

So can Red do another test with someone else doing it?

 

[Todd M.]

Maybe my brain is broken but I just see a blurry hand unless I follow it with my eyes.

I think the easy way to understand what Jim means is the old bending pencil optical illusion (this thought experiment is what helped me see/understand what is happening with MM). When you hold a pencil in the center with two fingers and quickly move your fingers up and down the pencil will appear to bend up and down at the ends. But it's not just a yellow blurry mess, you can still clearly see distinct edges of the pencil. The illusion that the pencil is bending, I believe, is caused by the fact that our eyes/brain can process any single moment and see the edges of the pencil. But we are also perceiving a blur caused by the motion. This "trail" of image perception is what causes to interpret that pencil as bending. This similarity of MM to our own eyesight is clearly visible in the three stills Jim posted of the spinning pole.

 

[Gavin Greenwalt]

I think the easy way to understand what Jim means is the old bending pencil optical illusion (this thought experiment is what helped me see/understand what is happening with MM).

On the other hand if you hold a pencil tightly and simply move it up and down quickly it just looks like a blurry pencil (except where it slows to change direction

Different optical illusion. The rubber effect is the product of the motion blur forming an arc by rotating and moving up and down as opposed to straight rigid motion blur from either rotation or translation on its own. If you were to do a long exposure of the pencil the motion blur would be very similar to that of an actual rubber object.

If I had to assign a hypothesis as to why MagicMotion looked more desirable I would attribute it to eye tracking. When you look at an object your eyes follow it and that reduces the motion blur for that subject. If you have a pan it's motion blurred. But if your eyes follow something in the background the motion blur in real life would go away since relative to your vision it's no longer moving. Magic Motion is both motion blurry and sharp at once. So if I had to guess why that's desirable it's because you don't notice the sharp when it's moving (since motion on screen is blurred in perception anyway even with short exposures) but if your eyes do follow they partly get that sharp motion blur image they expect.

Eyes like to follow things it's really hard to wave your hand in front of your face without following it with your eyes. If you track your hand with your eyes it's motion blur free. If you don't it's just motion blur. MM gives you both. Or at least that would be my guess.


An interesting experiment that someone could perform without HDRx™ would be to do a pan tracking a subject with both 180* shutter and 5* shutter (but exposed the same) and then have people follow either the background or the subject. My prediction is that people would prefer the 180* when tracking the subject and the 5* when tracking the background with their eyes.

 

[Jason Sinclair]

Is the magic motion a by-product of the amount of HDR used?

 

[David Quinn Carder]

If I had to assign a hypothesis as to why MagicMotion looked more desirable I would attribute it to eye tracking. When you look at an object your eyes follow it and that reduces the motion blur for that subject. If you have a pan it's motion blurred. But if your eyes follow something in the background the motion blur in real life would go away since relative to your vision it's no longer moving. Magic Motion is both motion blurry and sharp at once. So if I had to guess why that's desirable it's because you don't notice the sharp when it's moving (since motion on screen is blurred in perception anyway even with short exposures) but if your eyes do follow they partly get that sharp motion blur image they expect.

Eyes like to follow things it's really hard to wave your hand in front of your face without following it with your eyes. If you track your hand with your eyes it's motion blur free. If you don't it's just motion blur. MM gives you both. Or at least that would be my guess.

That's a really interesting hypothesis! Perhaps you're right. I had not considered that.

 

[Jannard]

Is the magic motion a by-product of the amount of HDR used?

1. EasyHDR combines the images in camera outputting the preset amount of "Magic Motion" and extended dynamic range.

2. Shoot and output dual stream HDRx, which is later combined in post (or not). The amount of the "x" frame (dark exposure) used for combining is variable and tied to the amount of "Magic Motion".

3. Same as #2 only use The Foundry plug-in on the "x" frame to create a MNMB (more normal motion blur).

Side note... in #2, the less "x" frame you use, the more normal the motion blur is.

Jim

 

[Tom Wardrop]

Is the magic motion a by-product of the amount of HDR used?

The answer is yes.

Luckily for RED, the by product is a desirable attribute. It's desirable probably because of the following theory...

If I had to assign a hypothesis as to why MagicMotion looked more desirable I would attribute it to eye tracking. When you look at an object your eyes follow it and that reduces the motion blur for that subject. If you have a pan it's motion blurred. But if your eyes follow something in the background the motion blur in real life would go away since relative to your vision it's no longer moving. Magic Motion is both motion blurry and sharp at once. So if I had to guess why that's desirable it's because you don't notice the sharp when it's moving (since motion on screen is blurred in perception anyway even with short exposures) but if your eyes do follow they partly get that sharp motion blur image they expect.

Eyes like to follow things it's really hard to wave your hand in front of your face without following it with your eyes. If you track your hand with your eyes it's motion blur free. If you don't it's just motion blur. MM gives you both. Or at least that would be my guess.

I like this theory, but I believe there's more to it. How our eyes track moving objects in video playback is one thing, and I can definitely see magic motion helping there, however, I'd guess that the reason magic motion looks better as a whole, is because of how our eyes perceive motion as compared to how video/film cameras perceive motion.

Our eyes don't have shutters, and hence they're constantly receiving new light information. The motion blur our eyes see in real-life is essentially a trail of old information, as opposed to an averaging of light over a given time period (which is essentially what a video/film frame represents). For our eyes, it's less motion "blur" and more motion "trail". If you switch on a bright light for example, you see the full brightness of that light instantly, where as when you switch it off, you'll likely see a ghost image of the light for a short period of time after. Magic motion, to an extent, simulates this behaviour of our eyes, and so it's my guess that this is why magic motion looks more natural. It's just a theory.

I believe this is essentially what Jim was getting at in one of his earlier posts (I'm not sure if it was in this thread or some other thread).

 

[Scott Crawley]

Sounds outstanding! You guys really are making the camera system I've been dreaming about since I was a kid.

 

[Gerald Menclik]

This was shot with Peter Lik and Mick Plumridge in a ghost town near Las Vegas over the weekend. Both said "impossible" when we went into the barn. It was wicked dark in there and blistering noon desert-sun outside. They both fell over dead when they saw this. I called 911. They are OK now...

Jim

a little late .... can't find the right words ..... but ....

 

[Sanjin Jukic]

HDR is possible also with older RED1-M R3D files

HDR is possible also with older RED1-M R3D files

I tried out to get HDR from my older R3D files that were shot two years ago at the same location.

The images were shot at the same location as my first HDR example here

but turned on another or opposite side of the entrance gate at Belvedere Gardens in Vienna.

And HDR works also with M sensor R3D files.

You just have to choose a right exposure when you pull and push to get two files for HDR tone-mapping.

The image was pulled at ISO 200 and pushed at ISO 800 just because this is ISO value when noise and hard clipping starts with M sensor.

Shot on RED1-M with Angenieux 35-70mm f/3.5 zoom: HDR tone-mapped image using Photomatix Pro from HDRsoft company based in France.

Shot on RED1-M with Angenieux 35-70mm f/3.5 zoom: The basic image at ISO 320.

Shot on RED1-M with Angenieux 35-70mm f/3.5 zoom: The basic image at ISO 200 later used for HDR tone-mapping.

Shot on RED1-M with Angenieux 35-70mm f/3.5 zoom: The basic image at ISO 800 later used for HDR tone-mapping.

Snapshot sample image R3D file for download 1.2 MB >>>

HDR tone-mapped image looks much better than the image @ ISO 320 (rated M sensor exposure index).

Also don't forget that with adjusting curve you can't get more DR than in a comparison with generating HDR (tone-mapping).

That means HDRx™ enhanced recorded images with EPIC and Scarlet will have more DR

than any images recorded at Log C (Alexa) or S-LOG gamma curve (Sony F35/Panavision Genesis)...

 

[Shane Betts]

One of the things that really excites me about HDR is that we are reaching the point where, with the exception of extreme cases, we no longer will have to light for purely technical (exposure) reasons. So now lighting is all about modelling. Less work and more emphasis on creativity and art, IMHO.

Sure it's going to open the door wide to a whole bunch of twits who just shoot anything that moves - Mini-DV-itis in 5k - but the other side of that coin is what interests me. Crap will sink, cream will rise (but don't drink that particular milkshake :thumbup1

 

[Scott Crawley]

Sure it's going to open the door wide to a whole bunch of twits who just shoot anything that moves - Mini-DV-itis in 5k - but the other side of that coin is what interests me. Crap will sink, cream will rise (but don't drink that particular milkshake :thumbup1

People have been shooting anything that moves since the inception of the photo-chemical process... So what. The very first moving images ever shot were of common things and they were sought after the world around.

Why be intimidated by tech that makes imaging easier? I don't get the point. To me, it is so trivial that I don't understand the need to even discuss it.

What is remarkable is that not only will we be getting digital tools that really rival film, but we will also have the ability to to manipulate our moving images in much the same way as we do stills in Photoshop... Without a million dollar investment.

 

[Lauri Kettunen]

It involves knowledge (not speculation) of what each does. That knowledge is what separates a "pro" from a "prosumer".

My wish is that as much information as ever possible of all the tools inside the camera and Redcine-X will be eventually given. For the very reason that if one knows what is really going on, then there is no need to guess or speculate. Instead, on the field one becomes able to predict what is the outcome of different choices.

Would like to add, in my view knowing is about conceptual level understanding which is a rather different thing than testing, which is about creating a list of experimental results. Obviously, it's not a big deal to interpolate between experimentally obtained observations, but the real thing is to extrapolate and push the camera to its limits based on a conceptual level understanding of the device.

To give an example, personally I'm not completely happy with the explanations given so far about FLUT. Or, perhaps I've just missed that Graeme has given a proper explanation somewhere? Neverthelss, I've got the point that increasing the FLUT value does not clip the highlights, but beyond that all what I know about it is gained by experimenting. Still understand well that there are reasons why everything cannot be explained out and loudly, so don't want complain -there is no reason for that. But, hope, if possible, when the Epics are out, as much as ever possible of HDRx™ and eHDR™ will be explained. That is important in creating a proper conceptual level understanding which then helps to squeeze out as much as possible of the camera in various situations.

 

[Jason Sinclair]

1. EasyHDR™ combines the images in camera outputting the preset amount of "Magic Motion" and extended dynamic range.

2. Shoot and output dual stream HDRx™, which is later combined in post (or not). The amount of the "x" frame (dark exposure) used for combining is variable and tied to the amount of "Magic Motion".

3. Same as #2 only use The Foundry plug-in on the "x" frame to create a MNMB (more normal motion blur).

Side note... in #2, the less "x" frame you use, the more normal the motion blur is.

Jim

Thankyou.

Considering Magic motion is a certain look, in a sequence of shots that uses different levels of dynamic range, do we also contend with different motion looks or has your testing found this to be unproblematic?

 

[Hrvoje Simic]

Don't forget that there are many layers of options here. You can, depending on your skill or budget, use masks and do some pretty involved stuff with the "x" track, including motion and grading. You can adjust the amount of the "x" track you use in REDCINE-X.

Suggestion > keyframing

(+assigning metadata values to Redcine values)