Welcome to our community

Be a part of something great, join today!

Register Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

M-X at ISO 2000...

David Battistella said:
I don't think you'll see a huge change in lighting styles anytime soon. Well lit images are well lit images and that won't change.

When fast film stocks came around that added a tool to the arsenal, it did not change lighting forever.

David
Click to expand...


I disagree- when fast film stocks came around, lighting styles changed quite a bit over time. Look at movies shot on old stocks- you see a lot of hard light and multiple shadows everywhere. Lighting a large stage for ISO50 or 100 could not have been easy. Those films have a specific look due to the need to use hard light to get exposure on slower stock. As stocks have gotten faster, we've seen way more soft light, more natural light, more night exteriors, single sources, etc. Faster stocks, in conjunction with faster lenses, DID help change lighting styles forever. I don't think lighting will change for most projects, but a new arena of low light photography will be opened with clean ISO1000 or 2000 and T1.3 lenses like Master Primes. Will this drive a whole new aesthetic? Who knows, but there are a lot of new creative opportunities available with this kind of sensitivity.
 
I don't think the jump from 500 ASA to 800 ASA is enough to radically change lighting techniques in general, which are based as much on stylistic trends as they are technology -- after all, look at Storaro's work on "Reds" or "Apocalypse Now" shot on 100 ASA film, the lighting is still "contemporary". Or look at Ridley Scott's "Alien" and "Blade Runner", also shot on 100 ASA stock.

What will change over time with greater sensitivity and dynamic range is the look of night exterior and super low-light photography, but even today, we have a lot of tools to shoot in those conditions, so what these new sensors will mainly get us is less noise in the same low-light image.

The trouble with actually lighting your average interior scene to ASA ratings in the 1000's with fast lenses is that it can actually be hard to get your light levels that low, when even a 100w Dedolight is too powerful. Plus natural ambience really starts to take over and lower the contrast and "pollute" the shadows with odd colors and multiple sources. Have you ever been out at night and noticed all the faint shadows on a face caused by all the different sources out there, though quite dim? These days I find myself having to flag off a lot of stray lighting outdoors at night when working at really low levels. Otherwise, you discover faint camera and mic shadows in post that you didn't catch on the set all the time.

Even in that test of Leonardo DiCaprio lighting the cigar and having his match light his face... it's not like we have never seen such an image before in a movie. Even "The Grapes of Wrath" from 1939 has a scene with a face convincingly lit by a match in the hand.

2000 ASA at T/1.3 is not outside the realm of 35mm 5219 500T stock if you push it two stops, it just would be grainier.

So certainly low-light scenes are going to get easier to do, with less noise, but I don't see that causing fundamental lighting changes; in fact, most night work today attempts a reasonable level of realism, ever since the 1970's (look at "Taxi Driver"), and should it be even easier to get realistic night photography, so that anyone can do it... for all you know, the next stylistic trend will be away from realism just because the style pendulum always starts to swing the other way at some point. If we see thousands of movies shot in city streets at night in available light, you know at some point someone will be hailed as the next cinematic "genius" for actually artificially lighting the streets with some unusual look.

But I agree that creativity can be expanded with tools that have greater abilities. Whether that expansion is meaningful or significant is often debatable though. Art often thrives on limitations, so fewer limitations doesn't necessarily increase the level of artfulness.
 
David Mullen ASC said:
I don't think the jump from 500 ASA to 800 ASA is enough to radically change lighting techniques in general, which are based as much on stylistic trends as they are technology -- after all, look at Storaro's work on "Reds" or "Apocalypse Now" shot on 100 ASA film, the lighting is still "contemporary". Or look at Ridley Scott's "Alien" and "Blade Runner", also shot on 100 ASA stock.

What will change over time with greater sensitivity and dynamic range is the look of night exterior and super low-light photography, but even today, we have a lot of tools to shoot in those conditions, so what these new sensors will mainly get us is less noise in the same low-light image.

The trouble with actually lighting your average interior scene to ASA ratings in the 1000's with fast lenses is that it can actually be hard to get your light levels that low, when even a 100w Dedolight is too powerful. Plus natural ambience really starts to take over and lower the contrast and "pollute" the shadows with odd colors and multiple sources. Have you ever been out at night and noticed all the faint shadows on a face caused by all the different sources out there, though quite dim? These days I find myself having to flag off a lot of stray lighting outdoors at night when working at really low levels. Otherwise, you discover faint camera and mic shadows in post that you didn't catch on the set all the time.

Even in that test of Leonardo DiCaprio lighting the cigar and having his match light his face... it's not like we have never seen such an image before in a movie. Even "The Grapes of Wrath" from 1939 has a scene with a face convincingly lit by a match in the hand.

2000 ASA at T/1.3 is not outside the realm of 35mm 5219 500T stock if you push it two stops, it just would be grainier.

So certainly low-light scenes are going to get easier to do, with less noise, but I don't see that causing fundamental lighting changes; in fact, most night work today attempts a reasonable level of realism, ever since the 1970's (look at "Taxi Driver"), and should it be even easier to get realistic night photography, so that anyone can do it... for all you know, the next stylistic trend will be away from realism just because the style pendulum always starts to swing the other way at some point. If we see thousands of movies shot in city streets at night in available light, you know at some point someone will be hailed as the next cinematic "genius" for actually artificially lighting the streets with some unusual look.

But I agree that creativity can be expanded with tools that have greater abilities. Whether that expansion is meaningful or significant is often debatable though. Art often thrives on limitations, so fewer limitations doesn't necessarily increase the level of artfulness.
Click to expand...

Wasting space here with my reply, but, great post, David.
 
What is great about high useable iso's is we may not have to open the lens up a much for low light situations. A more manageable dof. Deeper focus low light.
 
Here's a fundamental question I keep discussing with people. If you had a Red 1 at 320iso sitting right next to a Mysterium-X Red set at 800 iso, both pointing at the same thing and with the same lens setting, will the Red M-X clip at any point where the Red1 did not? Or is it that they'd both clip identically but the M-X get's 1.5 stops deeper into the shadows?
 
David, It sound like you're looking over your shoulder for
fear of something great being left behind.
No matter which way the "pendulum" swings,
it will have a wider arc and it will include every
increment of the past.
 
Shawn Nelson said:
Here's a fundamental question I keep discussing with people. If you had a Red 1 at 320iso sitting right next to a Mysterium-X Red set at 800 iso, both pointing at the same thing and with the same lens setting, will the Red M-X clip at any point where the Red1 did not? Or is it that they'd both clip identically but the M-X get's 1.5 stops deeper into the shadows?
Click to expand...

I don't think any tests at that level of detail have been released yet.

The MX sensor seems to be (1) faster; (2) with more dynamic range and (3) less noise. But by how much compared to the old sensor? Well, we'll find out as more and more people get their hands on the MX sensor.

Questions yet to be answered includes if 800 ASA is really the MX setting that splits highlight and shadow detail equally, or is it more like 500 or 640 ASA. If it really is 800 ASA, that could be problematic in sunlight -- we're all going to have to start carrying an ND1.2, 1.5, and maybe even an ND1.8.

But the great thing about greater dynamic range is that exposing is a little less critical, you have more leeway. Eventually we'll have something like color negative film in terms of range, making exposing a lot more idiot-proof and problematic shots a lot more fixable in post.
 
Shawn Nelson said:
Here's a fundamental question I keep discussing with people. If you had a Red 1 at 320iso sitting right next to a Mysterium-X Red set at 800 iso, both pointing at the same thing and with the same lens setting, will the Red M-X clip at any point where the Red1 did not? Or is it that they'd both clip identically but the M-X get's 1.5 stops deeper into the shadows?
Click to expand...

They's both clip (but not identically) and the M-X will see deeper into the shadows.

I asked Jim pretty much this exact question at the Red Day session. What he said was (you might want to read this a couple of times, because it is all I've been thinking about for the past two days as I've tried to understand how this can be) "There is no penalty at lower ISO's." He sad to Adam Wilt at Pro Video Coalition, "This sensor is EXACTLY like the original Mysterium sensor but with a much lower noise floor."

Based on those two bits of information, here's how I think the magic works. They haven't really produced a more sensitive camera, they've produced a cleaner camera. We all know that noise is the limiting factor for dynamic range, and what RED has done is to produce a sensor system with so much less noise that the footage can be processed using a much higher ISO ratings with very acceptable noise levels. That's what we saw in the DiCaprio shot.

At the other end of the spectrum, highlights don't clip any sooner than the current sensor, because the sensor is not "more sensitive," it's just cleaner. But there's another bit of magic at work here, and that's FLUT. Graeme's got algorithms in there that seem to me to be adjusting gamma in non-lenear, non-computer way that more closely resembles the curves of film, so the roll-off at the high end is softer and more film-like. Additionally, these curves seem to put the midrange of the response (18% gray) where a light meter says they should be.

I suppose this is more conjecture than I should be making based on the things that were actually said to me, but if I'm close to right, it's pretty damn sweet.
 
I have to say guys I saw this on Saturday projected in 4k on Sony's new projector and it looked absolutely amazing. There was NO visible that could be seen from a 'normal' theatrical viewing distance. When I was able to get quite a bit closer to the same image which was being piped out of a Pablo with a new 4K output board and projected on the older Sony 4k projector there was some visible noise but really, this may have been the projector introducing some artifacts because the images Jim posted look very similar to how I remember the initial projection on the new Sony 4K.

Red Day was excellent btw and I highly recommend it to anyone who can make it the next one. Well worth the 5.5 hour flight from NYC.
 
With a new sensor, camera build, SDK and more lattitude, I am wondering two things:

1. Because of increased data, do the R3D's in shot with the M-X take more time to process than the ones that we shoot now with our R1's? (important consideration for post workflows)

2. How much of the extra sensitivity is the sensor vs. the post tools?


Stuff looks great.
Thanks,
David
 
David Mullen ASC said:
I don't think any tests at that level of detail have been released yet.

The MX sensor seems to be (1) faster; (2) with more dynamic range and (3) less noise. But by how much compared to the old sensor? Well, we'll find out as more and more people get their hands on the MX sensor.

Questions yet to be answered includes if 800 ASA is really the MX setting that splits highlight and shadow detail equally, or is it more like 500 or 640 ASA. If it really is 800 ASA, that could be problematic in sunlight -- we're all going to have to start carrying an ND1.2, 1.5, and maybe even an ND1.8.

But the great thing about greater dynamic range is that exposing is a little less critical, you have more leeway. Eventually we'll have something like color negative film in terms of range, making exposing a lot more idiot-proof and problematic shots a lot more fixable in post.
Click to expand...

Jim said at the RED Day that there would be no penalty in going lower in the ISO setting with the M-X sensor. So if you needed to go down to ISO 100 you could without losing detail like you would with the original sensor.
 
David Mullen ASC said:
2000 ASA at T/1.3 is not outside the realm of 35mm 5219 500T stock if you push it two stops, it just would be grainier.
Click to expand...

Sure, except film grain reminds the viewier they are watching something artificial. I think that a digital sensor that can create a less distracting image in low light really sells the shot and the camera.
 
You must log in or register to reply here.
Back
Top