Build 15 update...

[Brent J. Craig]

yeah, i quoted you because we are saying the same thing but in different terms.

Deanan, please continue translating Joofa for us. Ever since they named that unit of measurement after him it's like he's been speaking Greek or something!

 

[Dj Joofa]

Keep in mind that the full frame DSLR's (D3 for example) have the same pixel count (12MP) in a sensor that is much larger (about 3x the area). Which means the pixels are larger. Which means better low light capability and less noise. And much harder to do high frame rates...

Jim

True. You are right.

And this was our basic premise that as frame rate increases we start seeing noise problems in low-light and night-time shots, which, fortunately for a still image camera, they don't have to concern themselves with.

 

[Dj Joofa]

Your right... but presetting your camera to 6500asa "tells your camera" in wich portion of the sensitivity of your censor you want to grab informations. This is in the low light zone where usually you'll see a lot of noise and where usually, when preset to 200 asa, you should only get deep blacks.

So changing asa, even if it's only a relique of the ancient times where different film speed were used, does in the raw world actually affect the calculation of T stops/exposure times and, at the end ,the overall look of the picture.

Pat:sarcasm:

You are right. I agree. As CrewPix said, I am having trouble saying the same things ;-)))))

 

[Radoslav Karapetkov]

Jim you certainly know how to play with our emotions.
Jon Corcuera

The joy is greater that way.

If we get all the cookies at once, that's just one happy day.

But the modularity and upgradability of RED gives us an endless horizon of incoming cookies.

---------

EVF+LCD+Preview -> that's huge.

11 stops of latitude -> Yay! (+ a little more).

Nikon mount ready -> I'll take it.

RED Drive -> 2+ hours of 4K -> Endless extasy.

Progressively decreasing noise -> Yummy and Yummier.

(Windows QT codec + AVI Export from REDCine) -> I'm coming baby!

...

And that's about enough.

"I don't want anything more." William Parish.


P.S. Who needs Panavision. :bleh:

 

[I Bloom]

Don't worry jim, that wasn't a dig at the Mysterium. Neither Nikon nor Canon have a sensor that can do 24FPS so it's of little consequence if they have more aggressive noise control. Also we have no idea how that noise, however pleasing in a still frame would look when passing by at 24 FPS. I think we all have faith that the Mysterium will reach it's full potential.

Evin,
Do a little test for me.
Take that 6400 D3 image into photoshop. Switch the color mode into Lab color. Do normal levels modification on each channel and show us the L channel and the the a+b channels. I think that will reveal much about what Nikon has been up to, and where Red should not follow.:shiftyph34r:

IBloom

 

[Deanan]

The PL mount diagonal 30/31mm design constraint is significant. You can't cheat by making bigger pixels and bigger sensors and still be compatible with all the great lenses out there.

 

[mezmo]

If Build 15 and the next builds give us a quality close to the photos below

that were shot on Nikon D3 I would be really, really satisfied>>>

A narrow street in Florence at night.
Digital Photography by Dennis Hissink
Nikon D3 Test Photo
Resolution : 4256 x 2832
Aperture : f/6,3
Exposure time : 1/50 sec
ISO : 6400
Focal length : 24mm
Metering : Aperture Priority
Compensation : 0 step

Download RAW NEF file>>>

Ponto Vecchio at night, Florence.
Digital Photography by Dennis Hissink
Nikon D3 Test Photo
Resolution : 4256 x 2832
Aperture : f/4,5
Exposure time : 1/10 sec
ISO : 6400
Focal length : 18mm
Metering : Aperture Priority
Compensation : 0 step

Download RAW NEF file>>>

Also find more and download original NEF RAW files at

LINK>>>

Hi Sanjin,
These frames are a great examples for NOT going Digital
at night.
In the first frame the ambient light is still daylight.
About 7-8K Kelvin and the street lights are strange, look Photoshopped
with strange flares and starbursts.
The real colortemp of these lamps could be the same as the blueish lens
flare bottom right of frame and the reflection in the truck window.
This is a tricked up example of a night shot to help the D3 sales department IMHO.
The Ponto Vecchio shot is also a little dissapointing showing lack of Dynamic Range
as the camera has let go of most detail in the burned out windows along the
waterfront.
Mezmo

 

[Warren Kommers]

And this was our basic premise that as frame rate increases we start seeing noise problems in low-light and night-time shots, which, fortunately for a still image camera, they don't have to concern themselves with.

I feel like these comparisons between RED and DSLR are being drawn with the understanding that the exposures being the same. The frame rate, shutter speed, iris, etc. We all know that SLR's can do exposures times that RED can't and that RED can do more frames a second. However with all that in mind I think most of us are saying that there are many great digital sensors that have shown low noise at high asa's with all things equal. I was surprised with the low sensitivity of RED. Perhaps there is more to the processing than the hardware than we expected. At least that is what it is starting to sound like.

Drop 15 on us Jim!

 

[Dj Joofa]

Hi Sanjin,
These frames are a great examples for NOT going Digital
at night.
..

The Ponto Vecchio shot is also dissapointing showing lack of Dynamic Range
as the camera has let go of all detail in the burned out windows along the
waterfront.
Mezmo

Interesting observations.

If that image was not acquired digitally, as you suggest that is not a wise step at night time, would you think that a DPX film scanner would have recovered all that detail if that image was acquired on a film negative?

We debated this issue to death on a different thread entitled "Are digital sensors there yet?", where I was of the view that if we use non-linear CMOS sensors then we might have a better response in night-time images such as these. Graeme Nattress opposed me, and his concerns were valid at many places, such as additional noise added to CMOS transistor circuits before ADC. However, I have a feeling that is a route that needs to be explored for recovering detail in images such as the ones Sanjin has posted. Daytime shots are too easy for any high profile digital camera, such as Red, but one of the challenging test cases are scenarios as presented in Sanjin's pictures.

 

[Dj Joofa]

I feel like these comparisons between RED and DSLR are being drawn with the understanding that the exposures being the same.

Drop 15 on us Jim!

Actually the situation is as follows: Suppose you have a night time shot, and you are already at a very big aperture size, but the image acquired still looks underexposed with those horrible artifacts of temporal noise. You are then forced to increase the only other free variable left, viz., the integration time (exposure time). But there is a limit to it if you want to sustain your frame rate -- heck, you are making a movie, it better be rock solid 24 fps. So sooner or later, a point reaches, where you have to either (1) drop frame rate, or (2) drop exposure time with increased perception of temporal noise.

 

[Warren Kommers]

Actually the situation is as following: Suppose you have a night time shot, and you are already at a very big aperture size, but the image acquired still looks underexposed with that horrible artifacts of temporal noise. You are then forced to increase the only other free variable left, viz., the integration time (exposure time). But there is a limit to it if you want to sustain your frame rate -- heck, you are making a movie, it better be rock solid 24 fps. So sooner or later, a point reaches, where you have to either (1) drop frame rate, or (2) drop exposure time with increased perception of temporal noise.

Yup. That's how exposure works. I'm not sure what your point is I guess. Most of us know that the higher your frame rate the less exposure time. There is also officially less light at night.

Provided the DSLR and motion picture camera(growing closer every day) have the same exposure time independent of frame rate(1/48), aperture, and asa. How do they compare? Isn't that what we are talking about? So we can compare processing and sensor characteristics. Isolating the variables. Perhaps add sensor size and mega pixel count to that list of things to isolate as Jim mentioned.

 

[Dj Joofa]

Yup. That's how exposure works. I'm not sure what your point is I guess. Most of us know that the higher your frame rate the less exposure time. There is also officially less light at night.

Provided the DSLR and motion picture camera(growing closer every day) have the same exposure time independent of frame rate(1/48), aperture, and asa. How do they compare? Isn't that what we are talking about? So we can compare processing and sensor characteristics.

Nope, exposure time on a digital sensor has no direct relationship with frame rate. As I said just a while ago it is only bounded by the frame rate. Typically the frame rate will be fixed, say the normal 24 fps, because you do not want to change that unless you are going for special effects, which we are not talking about. Now 24 fps is fixed. Hence, it provides us an upper bound on exposure time. Now during daytime the exposure time could be quite small but still be within that constraint of 24 fps. However, to sustain 24 fps at night time, after exhausting the useful range of the aperture size, you are forced to increase your integration time, and it has a limit enforced by 24 fps.

 

[Dj Joofa]

I'm not sure what your point is I guess. .

BTW, isn't that what my reputation is coming to on this forum that what is the point I am harping about. So sad. So sad. So sad. So sad. So sad.

I think I need to take a sabbatical.

 

[Evin Grant]

Evin,
Do a little test for me.
Take that 6400 D3 image into photoshop. Switch the color mode into Lab color. Do normal levels modification on each channel and show us the L channel and the the a+b channels. I think that will reveal much about what Nikon has been up to, and where Red should not follow.:shiftyph34r:

IBloom

By normal levels do you mean "auto" levels? Or something else?

 

[Evin Grant]

Keep in mind that the full frame DSLR's (D3 for example) have the same pixel count (12MP) in a sensor that is much larger (about 3x the area). Which means the pixels are much larger. Which means better low light capability and less noise. And much harder to do high frame rates...

Jim

This is a very important point and one I should have remembered to make.

 

[Warren Kommers]

Nope, exposure time on a digital sensor has no direct relationship with frame rate.

I think we are running in circles here. I of course understand this but a 1/48 is a 1/48th for comparing sensitivity and processing of the image. At 24fps you won't be able to do better than 1/24.

 

[Dj Joofa]

I think we are running in circles here. I of course understand this but a 1/48 is a 1/48th for comparing sensitivity and processing of the image. At 24fps you won't be able to do better than 1/24.

That is an automated message. I am on sabbatical. Please contact xxxx for your inquiries meanwhile.

 

[Bing Bailey]

we need to sensors , one rated for daylight and one for night on a little carousel so they can move into position at the click of a button

 

[Dj Joofa]

I think we are running in circles here. I of course understand this but a 1/48 is a 1/48th for comparing sensitivity and processing of the image. At 24fps you won't be able to do better than 1/24.

Okay I am back from my sabbatical. The point is that if you opt to not drop frame rate for difficult night time shots, and increase exposure time and aperture size to a maximum, then on a film camera you only see an underexposed image. On a digital camera in addition to an underexposed image you see those horrible artifacts of temporal noise. One can still live with the corresponding film image, by the famous trick of DPs that they stick a bright light source in a distant corner of the film to psychologically lure audience into believing that the image was deliberately underexposed for artistic reasons -- the DP can get away with this because on film (whose noise characteristics are inverse of digital; more noise in highlights; less in dark) the image though underexposed was surprisingly clean of noise. Where as the digital image because of temporal noise has been rendered useless. Unless you really drop frame rate on a digital camera to a low number the temporal noise will be sucking the pixel-blood out of the image.

Therefore, to develop a digital camera that has even reasonable frame rate in difficult night shots is a challenging task.

 

[Warren Kommers]

Okay I am back from my sabbatical. The point is that if you opt to not drop frame rate for difficult night time shots, and increase exposure time and aperture size to a maximum, then on a film camera you only see an underexposed image. On a digital camera in addition to an underexposed image you see those horrible artifacts of temporal noise. One can still live with the corresponding film image, by the famous trick of DPs that they stick a bright light source in a distant corner of the film to psychologically lure audience into believing that the image was deliberately underexposed for artistic reasons -- the DP can get away with this because on film (whose noise characteristics are inverse of digital; more noise in highlights; less in dark) the image though underexposed was surprisingly clean of noise. Where as the digital image because of temporal noise has been rendered useless. Unless you really drop frame rate on a digital camera to a low number the temporal noise will be sucking the pixel-blood out of the image.

Therefore, to develop a digital camera that has even reasonable frame rate in difficult night shots is a challenging task.

The last thing I think about, as a DP/director who has shot my fair share of 35mm and digital, is...."I'll put a light in the corner of the frame so that the audience will think I'm artistic and that I haven't underexposed this shot".
If they are thinking about that at anytime then I have failed in my opinion. They should be in the world of the story.

Although one is a positive image and one is a negative that renders different characteristics and techniques I can tell you that I worry about noise in the shadows on both far more than noise in the highlights. It's far more troublesome.