Mysterium-X dynamic range...

[Andrew Walker]

I know it's something that we'll find out soon but when people say that the low light capability of M-X is a game changer what do they mean exactly?

Just so we have a reference point for a comprehensive comparison... Can anyone that attended yesterday say how it compares to the low light capability of 7D?

All i cared about in order to upgrade my RED1 was more dynamic range and better colour so i am already sold but it would be cool to understand how low we can go.

Well what I saw, the new M-X sensor smokes the 7D in the sensitivity department. We could talk about how its vastly superior to the 7D but there are already far too many threads about that on here.

 

[Miltos Pilalitos]

Well what I saw, the new M-X sensor smokes the 7D in the sensitivity department.

ok, Got it! :thumbsup:

 

[Rick Darge]

Last night was amazing. 2000ISO is so damn clean it blows my mind. To my eye it looks like 640ISO on the current R1. I was planning on just going balls to the wall Epic-X but now am seriously seriously considering this new sensor upgrade. 6400ISO looked incredible. I can't wait to shoot some downtown night exteriors with a set of T1.4 lenses. Rock

 

[Jannard]

I know it's something that we'll find out soon but when people say that the low light capability of M-X is a game changer what do they mean exactly?

Just so we have a reference point for a comprehensive comparison... Can anyone that attended yesterday say how it compares to the low light capability of 7D?

All i cared about in order to upgrade my RED1 was more dynamic range and better colour so i am already sold but it would be cool to understand how low we can go.

Getting a useable image at ISO 6400 is cool... but really. What is practical is shooting ISO 2000 and not having to use noise reduction for me. DSLRs typically use F 2.8 lenses for professional use. That means the a 2.8 lens at ISO 6400 is about the equivalent to T1.8 at ISO 2000.

We showed footage of Leo DiCaprio lit by David Fincher with a match (only key light) at ISO 2000. I'm not exactly sure how much lower light you really want to shoot in...

My impression of the M-X sensor is that any practical ISO advantage of a DSLR is now gone. All that is left is line-skipping, H.264 and price. But that is just me...

Jim

 

[Andrew Luczenczyn]

We showed footage of Leo DiCaprio lit by David Fincher with a match (only key light) at ISO 2000. I'm not exactly sure how much lower light you really want to shoot in...Jim

Who needs a whole lighting crew when all you need is a box of candles and a fire marshal?

 

[Miltos Pilalitos]

We showed footage of Leo DiCaprio lit by David Fincher with a match (only key light) at ISO 2000. I'm not exactly sure how much lower light you really want to shoot in...

In my imagination this shot could look like many different things. I hope we could see it posted here soon.

Anyway, i was planing to go the Epic-X way and be happy about it but now i think i will upgrade my RED one, buy the Epic-X and a Scarlet to carry with me.

Dammit!

 

[Rick Darge]

Getting a useable image at ISO 6400 is cool... but really. What is practical is shooting ISO 2000 and not having to use noise reduction for me. DSLRs typically use F 2.8 lenses for professional use. That means the a 2.8 lens at ISO 6400 is about the equivalent to T1.8 at ISO 2000.

We showed footage of Leo DiCaprio lit by David Fincher with a match (only key light) at ISO 2000. I'm not exactly sure how much lower light you really want to shoot in...

My impression of the M-X sensor is that any practical ISO advantage of a DSLR is now gone. All that is left is line-skipping, H.264 and price. But that is just me...

Jim

I think ISO6400 would be incredibly useful for Dogma95-esque features and documentaries. Just having that option brings me joy

 

[Rob Ruffo]

Bits do not have to equal stops. A stop is an abstract unit of measure used to assign value to a quantifiable range. A stop measures light transmission similar to how an inch or centimeter measure length.

A 12bit value can represent an integer range from 0 to 4095. Or 4096 integer steps between two points. I doesn't matter if the sensor can register 2 stops of DR, 8 stops or 20 stops -- a 12bit value will represent 4096 total steps between the lowest and highest value recorded.

By having a larger range or more stops applied to this 12bit value, you only lose precision between the end-points of your range. In a linear system, the precision loss is uniform across the range. In a logarithmic system, you have a curve and can weight the curve to retain precision in one area of the range more so than another.

Another way to look at a 12bit system, if it accounts for 12 stops of dynamic range, then you have 4096 values ranging from 0 on up to 12 stops in 0.0029 stop increments (341.33 values between full stops). If this same system were to account for a range of 16 stops, then you would still have the same 4096 values, but this time each value would correspond to 0.0039 stop increments or 256 values between full stops. So greater range, but less precision.

This is an excellent explanation, but innacurate in one way: Stops are not abstract. 1 stop difference, = half as much light, exactly. +1stop but -50%ISO = same exposure. So ISO100 f8 = ISO200 f11, for example.

 

[Jimmy Odom]

Ok so I have a question, will the Scarlets' seeing as they will also have the M-X within their frames be just as capable at grabbing, apparently jaw dropping footage of DiCarprio being key lit by a match.

Or will they have slightly lower thresholds. I don't think we've talked about this yet.

I mean are these capabilities of the M-X going to be a sensor wide thing, or will size factor into it's over all capabilities?

 

[Rob Ruffo]

You can cram more stops into fewer bits, the danger just starts to become visible contouring steps in brightness on a smooth surface with a gradient.

I 50% agree. If you do not grade much, this is never a problem. For example we have hacked our 8bit EX1 via our own firmware to deliver 12 full, real stops into its 8 bit codec. It works, but grading is almost impossible - you can only shoot what you need to get, at least in terms of luma (same-luma hue shifts still work fine). In this case all we were doing is reprogramming how much light it takes to deliver a 100% white pixel vs a black one (we increased the amount it takes, there is no cheating base-ISO), but 8 bit has enough values to give a smooth gradient between full white and 0,0,0 black, no matter how large the area of that gradient.

SD DVD is something like 6 bit in practical reality, but you can display high-range film-transfered images on DVD without problems. BUT... I don't think many people have ever tried to grade DVD footage....

 

[Andrew Thomas]

This is an excellent explanation, but innacurate in one way: Stops are not abstract. 1 stop difference, = half as much light, exactly. +1stop but -50%ISO = same exposure. So ISO100 f8 = ISO200 f11, for example.

Uhhh, what you posted is basically the definition of abstract :smiley:

 

[Martin Stevens]

SD DVD is something like 6 bit in practical reality, but you can display high-range film-transfered images on DVD without problems. BUT... I don't think many people have ever tried to grade DVD footage....

I have graded old analog Hi-8 video, and I was amazed at what could be done. It did not look good, but it was far more gradable than I thought possible.

 

[Jeff Kilgroe]

Uhhh, what you posted is basically the definition of abstract :smiley:

Ummm.... Yep.

This is an excellent explanation, but innacurate in one way: Stops are not abstract. 1 stop difference, = half as much light, exactly. +1stop but -50%ISO = same exposure. So ISO100 f8 = ISO200 f11, for example.

Actually stops are a level of abstraction. What you're actually saying is that the scale is not linear, which you're right, it isn't. A better way to look at it is that if you have 4096 (12bit) luma values to play with and 12 stops of range, you assign 0 to the lowest value you wish to work with and 4095 to the highest and distribute the other values throughout the range. They can be linear or they can be logarithmic. As of right now, on the current Mysterium, the data is quantified in a linear fashion. As for ISO, that is yet another factor and within this system, ISO rating is mostly a function of where mid-grey is placed within the luma range.

My point was that bits and stops do not correlate directly, the bit depth of the imaging system only determines the level of precision, not the number of stops in its range.

 

[Todd M.]

This is an excellent explanation, but innacurate in one way: Stops are not abstract. 1 stop difference, = half as much light, exactly. +1stop but -50%ISO = same exposure. So ISO100 f8 = ISO200 f11, for example.

Ummm.... Yep.



Actually stops are a level of abstraction. What you're actually saying is that the scale is not linear, which you're right, it isn't. A better way to look at it is that if you have 4096 (12bit) luma values to play with and 12 stops of range, you assign 0 to the lowest value you wish to work with and 4095 to the highest and distribute the other values throughout the range. They can be linear or they can be logarithmic. As of right now, on the current Mysterium, the data is quantified in a linear fashion. As for ISO, that is yet another factor and within this system, ISO rating is mostly a function of where mid-grey is placed within the luma range.

My point was that bits and stops do not correlate directly, the bit depth of the imaging system only determines the level of precision, not the number of stops in its range.

I think you are both right, just attacking this from different angles:

Stops are finite when referring to light levels as Rob points out.

Stops are abstract, so to speak, when assigning data units to a given stop-range in the form of integers, be they 8-bit, 10-bit, 12-bit, etc. as Jeff rightly mentions.

 

[Sarah C.]

You can cram more stops into fewer bits, the danger just starts to become visible contouring steps in brightness on a smooth surface with a gradient.

Just coincidently: I thought the White R8 footage also looked really good...saw it a second time over in the next room on the Pablo.. "RAW" without grading and Ian(?) showing us what it looked like through grading very specifically via the highlights on the side of the car. imho.. seeing the demo reel "played with" on the Pablo told me a lot more about what the M-X was providing than just the demo reel alone.

@Rod, Jeff @ Todd - f-stops are abstract by nature! They are based on mathematical formula vs. measured light like T-stops. ((giggle))

~s

 

[Todd M.]

@Rod, Jeff @ Todd - f-stops are abstract by nature! They are based on mathematical formula vs. measured light like T-stops. ((giggle))

~s

See, I was just trying to wrap it up and you had to go and add more fuel to the fire, huh Sarah. :001_tt2::wink:

 

[Bill Anderson]

"Getting a usable image at ISO 6400 might be cool" for some,
but it's going to be much more than that for a few others-
at least if their definition of "usable" coincides with Jim's.
As the technology progresses then so will our vocabulary
as filmmakers etc.., interpretations- expressiveness.
We can already feel the odd ripple from the likes of Soderbergh.
With this in mind: the practice and art of lighting can be born out
of a particular desire rather than stark necessity.
But that's just the tip of the Soderbergh. (;

 

[Enrique Blein]

Red Dynamic Range is what this project is all about for me!
Why is it so difficult to Google the exact dynamic range of Sony prosumer cameras? It is such an important factor and yet it is like a family member nobody wants to talk about because of some obscure illness. I guess HDV and XDCAM must be around 6 stops in my opinion.

 

[David Mullen ASC]

Because dynamic range is a marketing tool, so don't believe what anyone tells you, it's nearly useless information in the abstract. If one company or person uses a method of measuring dynamic range that is different than another, then the figures are useless as a point of comparison. Plus there is a difference between recorded range and useable range, and everyone decides what is "useable" differently.

A practical example would be the recent comparisons between the Sony F35 (considered probably the king of dynamic range right now) and the new MX sensor and the new ARRI Alexa sensors -- what seems to be the result of these tests is that the dynamic range is similar but the noise is lower in the new sensors. And less noise gives you more flexibility in color-correction, so more of that range is useable. But the recorded measurable range may be rather similar.

So the only time you are going to get a realistic view of the differences between two cameras is to shoot your own tests and color-correct them, because that can also be a factor in what information is useable and able to be manipulated. In other words, the recording format can have an effect on what the sensor can output.

So Sony releasing "X" figures for dynamic range and Red releasing "Y" and ARRI releasing "Z"... well, it has little value.

An even when an outside party does their own comparison test, even that has to be taken with a grain of salt. Their testing methodology could be potentially flawed or simply represents their own personal way of working with digital images. But at least a side-by-side comparison does make certain differences more immediately visible than specs released by manufacturers, which is just a different form of information to be weighed.

On the other hand, common sense tells you that it is unlikely for an 8-bit prosumer HDV camera to capture the same dynamic range, with the same ability to be manipulated, as better cameras and recording formats working at a higher bit-depth and with better compression schemes.

 

[Jason Diamond]

I was at RED Day last weekend and Ted and I were swinging the camera around on the tripod and if the rolling shutter was present i did not see it. It is pretty astounding all the way around.