real native iso of dragon?

[David Mullen ASC]

Well yeah, people could just learn and have the discipline to expose properly, regardless.

And no, I wasn't suggesting just swapping one set of labels and numbers for another in defiance of established practices and standards.

I do think though the reliance on the use of 'ISO' is problematic, even though the way exposure information is presented in-camera is totally workable and fine by me personally.

Was just looking at the subject and asking questions in a more abstract way. Call it silly and impractical if you want, but I find it interesting.

It's not a bad idea but since the issue is really noise vs. headroom on most cameras, you'd almost need a numerical value for noise in combination with a second numerical value for exposure rating. As long as there is only one rating system and value, one person will object to the resulting noise level while another will think it is fine.


The funny thing is that still photographers rarely discuss what the "native" sensitivity is of, let's say, a Canon 5D sensor, they just talk about how fast they can rate it before the noise looks bad. But otherwise, most don't try to shoot everything at what they consider its native sensitivity is, they just pick an ISO setting that gives them the noise characteristics that they can live with. But that's partly because if the camera generally creates a low-noise image, people feel they can rate it all sorts of ways without thinking too much about it.

 

[Robert Ruffo New]

It's not a bad idea but since the issue is really noise vs. headroom on most cameras, you'd almost need a numerical value for noise in combination with a second numerical value for exposure rating. As long as there is only one rating system and value, one person will object to the resulting noise level while another will think it is fine.


The funny thing is that still photographers rarely discuss what the "native" sensitivity is of, let's say, a Canon 5D sensor, they just talk about how fast they can rate it before the noise looks bad. But otherwise, most don't try to shoot everything at what they consider its native sensitivity is, they just pick an ISO setting that gives them the noise characteristics that they can live with. But that's partly because if the camera generally creates a low-noise image, people feel they can rate it all sorts of ways without thinking too much about it.

Yes but, also in part, I think, because photographers rarely shoot with a grip truck full of lights. They must make do with a lot less than we do, and, further down the line, Photoshop can fix a lot more issues quickly than what is possible with moving footage except an gynormous expense - and sometimes even then.

 

[Abe Espinoza]

^^^ True, I hardly ever notice people discussing lighting here on the forum. Conversations usually consist of geeking out about the camera and it's specs.

 

[David Battistella]

There's quite an important distinction between photographers and cinematographers when it comes to lighting..

a cinematographer lives at a minimum of 24fps or above and a photographer lives at 1fps with often with burst flash lighting.

i think that although red is a DSMC system, if you want to compare lighting then compare it at equal FPS. The reason the stills guys don't have a truck load of lights is because they are capturing at 1fps and that is what makes a big difference.

Just flip your red camera ove to 1FPS and notice how "sensitive" it becomes.

the grip truck (and I just don't believe you need a grip truck unless you need to cover large areas) or to keep a set moving. Even then one can be quite imaginative and creative wit their lighting choices based on so many other factors.

yes, large productions have more resources, but those needs are more in the context of shooting a number of pages and keeping a large crew on schedule.

Battistella

 

[David Battistella]

^^^ True, I hardly ever notice people discussing lighting here on the forum. Conversations usually consist of geeking out about the camera and it's specs.

And they are often asking for "unicorn specs" because they want a "low light look" at 3200 ISO. they have now clue how many grip trucks and water trucks it takes to create a really nice looking low-light, night scene.


battistella

 

[Abe Espinoza]

I completely agree, David. We've still got to put effort in to creating the image and not let the camera alone dictate our outcome.

 

[Les Hillis]

It's not a bad idea but since the issue is really noise vs. headroom on most cameras, you'd almost need a numerical value for noise in combination with a second numerical value for exposure rating. As long as there is only one rating system and value, one person will object to the resulting noise level while another will think it is fine.


The funny thing is that still photographers rarely discuss what the "native" sensitivity is of, let's say, a Canon 5D sensor, they just talk about how fast they can rate it before the noise looks bad. But otherwise, most don't try to shoot everything at what they consider its native sensitivity is, they just pick an ISO setting that gives them the noise characteristics that they can live with. But that's partly because if the camera generally creates a low-noise image, people feel they can rate it all sorts of ways without thinking too much about it.

Is an interesting point, that the stills side of the RED DSMC user base has yet another particular set of established practices, methods, needs & wants. Which could itself be broken down into film stills and digital stills approaches to image capturing and processing.

I imagine if RED were a much bigger company they might have several different models of camera's, all using the same or similar sensors, but each with their own menu structures and functionality and pre-set workflows aimed at the cine, video and stills markets. Which makes what RED is actually doing to accommodate all those different uses and approaches in the one camera all the more remarkable. Even if it does require the end users to put some effort into learning how to navigate all that in order to find a workflow that suits their particular needs.


About the need for a numerical value for noise and exposure rating, I think in an ideal world there would be such a scale. Something very similar to the standardized (I'm assuming it is actually standardized) signal to noise ratio (SNR) that exists in the audio realm.

A scale based on a standardized test set-up that can be used to objectively measure the intended visual signal and the unintended distortion or 'noise', and rate the results as a numerical percentage or ratio. People would then be free to interpret the results in a subjective way and set their own preferences for how much noise is acceptable, but it would still be based on a purely technical and unbiased measure.

Similar to the way MTF charts for lenses can show an accurate measure of resolution but can't account for other lens characteristics or peoples subjective taste, a 'noise' chart could show a sensors signal to noise ratio while still allowing for people's preference for a sensor that has a particular flavour or kind of noise (for example, one that seems more random or organic, or that only appears at extreme exposures). In that way, all sensors could be measured not only in relation to each other, but each individual sensor's exposure values could be measured in relation to each other. So, theoretically, if noise was the only factor in a persons sensor preference, they could choose which one to use and how to expose it based solely on the SNR results.

Anyway, that's all different to what I was mentioning before about how exposure/noise information could potentially be presented to the end-user of a RED camera, though it could help if it ever came into existence.

And again, this is all just thinking out loud, I'm not saying this should be done, or even could be done. It would obviously be a major undertaking, basically a whole new system of ('noise') measurement. But, with image-capture increasing in resolution and becoming predominantly digital, and with new technologies that allow for greater precision and consistency of measurement and capture, I wonder whether some of the old analogue standards are still the best option for current and future applications.

 

[Cassius sanders]

That makes sense

 

[Sabovic Adis]

Just oust another way of describing raw view...


"RAW" view is a guideline.

It is
800 ISO
RC2
redlog film

that is RAW Views default curve.

The stoplights and goalposts and gioscope are telling you more about what's going on at the sensor level.

RAW view is technically not RAW, but let's you see an RLF gamma curve at 800ISO.

Just for clarity.

Battistella

Hi, David!
Can I call ISO 800 a "netral" or "unity" setting (no midpoint shift down or up :smiley when developing using Redlog film?

 

[Christoffer Glans]

Isn't the rated ISO the same as the RAW exposure? In that case the rated value is ISO 320 or 400. Switching between looking at RAW and looking at a log representation of the image shows less shifts when set to ISO 320 or ISO400.

If you want to keep the image as good as the sensor can get it I believe that is the setting you are looking for. Now, Dragon will look good on most ISO settings, but I think that should be considered the base ISO to be fair technically.

 

[David Battistella]

Hi, David!
Can I call ISO 800 a "netral" or "unity" setting (no midpoint shift down or up :smiley when developing using Redlog film?

If if you look at the great series of charts Phil has posted you can start to see where the stops fall above and below middle grey. At 800 the stops are about equal in number above and below middle grey.

You can rate however you wish depending on where you want those stops to fall. Rating higher or lower will move middle grey.

800 is a very good starting point and it's there for a reason, because it's a very good place to start.

 

[Sabovic Adis]

Thank You very much, David!
:smiley: