Epic and Scarlet release in 2010

Status
Not open for further replies.
EPIC Tattoo is out-

EPIC Tattoo is out-

The Good News is that the Epic Tattoo is out, and if they only have a 2 month prep window, then that would be around the first of october-ish, which in turn could possibly mean that Scarlet has a chance of being out by years end. Of course, as we all know everything is subject to change ...

At least one more mile stone has been crossed towards the end goal with Epic:
http://reduser.net/forum/showthread.php?t=48278
 
And the somewhat bad news is how many months are they behind on their projects to this point. I'd take a serious bet that you won't see the Scarlet out until this time next year at the earliest. I feel that the buzz and hype generated is somewhat of a tactic and a lacking level of accountability. I know that they say things change, but they've dashed the enthusiasm of how many people..? I'ts also not going to be the 3k for 3k first proposed, and ultimately it's going after the big boys lunch money and that might press on the big camera companies to release cameras in advance that fullfill most of the hopes and dreams of the Scarlet.
 
Justyn, I'd bet RED has learnt its lesson.

They are quiet for how many months until now?

Also Jim already said that we're going to see Fixed Scarlet by the year's end.

Anyway... Let's wait and see.
 
Bit of news. Epic prototype is reportedly hitting 18 stops of DR using the HDR mode in 24fps 1/48 shutter motion test, but is likely limited by the DSC chart it was tested on.

Michael Cioni, of LightIron Digital, was on set for the test to say:
Today I saw more stops from a captured R3D than I could with my own naked eye. No exaggeration. I've long talked about a world in which digital technology-driven dynamic range exceeds what the human eye can render. I am dangerously close to being speechless.
Click to expand...

There have been no posted charts or images (still or otherwise) for public viewing.

The HDR mode will evidently be graduated, so one doesn't always have to capture in "full HDR" mode.

All of this is expected to be released after the first hardware starts shipping.

http://reduser.net/forum/showthread.php?t=48494
 
Last edited:
if the human eye can't resolve the extra stops, then what good is it. Did I mention I can play Bach on my dog whistle. Although no one can hear it, so it really is irrelevant. Im just kidding. Im obviously missing something here. So if this out does the human eye, How is this benificial. just curious as to why it's great news if you can't see it.
 
maranfilms said:
if the human eye can't resolve the extra stops, then what good is it. Did I mention I can play Bach on my dog whistle. Although no one can hear it, so it really is irrelevant. Im just kidding. Im obviously missing something here. So if this out does the human eye, How is this benificial. just curious as to why it's great news if you can't see it.
Click to expand...

basically, if you are shooting a scene, you can pick how much is exposed, be it the highs or lows... everything with me exposed, so you can crank darker from there
 
maranfilms said:
if the human eye can't resolve the extra stops, then what good is it
Click to expand...

That argument works for resolution, less so for exposure. It can be quite the opposite.

As you know, a screen (television, computer, theater) has a maximum contrast ratio. Usually, by the way, it's less than what the salesman pitches. Not 1,000,000 to 1. More like between 100 and 500. This is the same as saying 7 or 8 f-stops.

So if you have a camera that records more f-stops than people's screens will show, where do the extra f-stops go? This is where curves come in, and film has been doing it from the beginning. With film's S-shaped curve, the darkest f-stops in life do not translate to full stop changes on the film. They move the film only about half a stop. Then, when you get to the middle brightnesses --- where human faces usually fall --- film starts to move more in step with the changes in brightnesses in real life. Then, as it nears the brightest f-stops, it tapers off, and starts responding in half stops or so. You might say that film is low contrast in the shadows, high contrast in the mids, and low contrast again in the highlights.

You might call this compression. The only other choice is clipping. Most people feel that compression, especially S-curved compression, looks the most pleasant. I don't think you would even want an image as bright as in real life. Imagine watching a movie where the shot of a Brooklyn sidewalk at noon is as bright as is real life. Or a shot of the sun. You want your TV to tame the contrast range.

With an 18-stop camera, you get more details in the shadows and highlights, instead of them being totally black or totally white. Or you could have grading flexibility in post. Only show a certain 12-stop slice of the image.
 
Last edited:
One should also note that HDR is not the same as a tone-mapped image, which is what all those rather garish "Photoshopped HDR" images are.

Rather, the image will remain natural looking, it will just have more latitude as combatentropy notes above.
 
18 stops? How much of a range is that? From Big Bang to complete darkness? Wonder what it really means - especially what's the amount of noise in the shadows. Still, amazing number. It's hard to imagine its impact on color grading, lighting, shooting. Who needs fill light anymore? ;) Who needs film with its fixed toe-shoulder response (S-curve) when we can get 18stops of linear response, allowing any curve to be applied in post? This cannot be true, of course. Like somebody mentioned, TVs have contrast ratios of millions already, which is over 20 stops. Unfortunately they loose 10 or 12 during shipping... Or maybe it's because my room is never dark enough to see them all? ;)
 
combatentropy said:
That argument works for resolution, less so for exposure. It can be quite the opposite.

As you know, a screen (television, computer, theater) has a maximum contrast ratio. Usually, by the way, it's less than what the salesman pitches. Not 1,000,000 to 1. More like between 100 and 500. This is the same as saying 7 or 8 f-stops.

So if you have a camera that records more f-stops than people's screens will show, where do the extra f-stops go? This is where curves come in, and film has been doing it from the beginning. With film's S-shaped curve, the darkest f-stops in life do not translate to full stop changes on the film. They move the film only about half a stop. Then, when you get to the middle brightnesses --- where human faces usually fall --- film starts to move more in step with the changes in brightnesses in real life. Then, as it nears the brightest f-stops, it tapers off, and starts responding in half stops or so. You might say that film is low contrast in the shadows, high contrast in the mids, and low contrast again in the highlights.

You might call this compression. The only other choice is clipping. Most people feel that compression, especially S-curved compression, looks the most pleasant. I don't think you would even want an image as bright as in real life. Imagine watching a movie where the shot of a Brooklyn sidewalk at noon is as bright as is real life. Or a shot of the sun. You want your TV to tame the contrast range.

With an 18-stop camera, you get more details in the shadows and highlights, instead of them being totally black or totally white. Or you could have grading flexibility in post. Only show a certain 12-stop slice of the image.
Click to expand...


ahh, thanks guys. i have a much better understanding of it now. very cool indeed.
 
18stops is probably achieved with dual exposures. This method has major artifacts in motion but it can work when there is no motion in frame. Does anyone know if it's a single exposure?

I have processed RAW 16bit from 10 to 22 true stops and there is practically no difference in the output. You can't fit more than 10 or 11 stops without an impact in the midrange contrast and the skintones. It can work for technical images and microscopy but not for cinema. What the extra range does provide is room to correct exposure in post if the recording has enough precision to avoid alias. But a good cinematographer wouldn't expose with an error of 5 stops.

Some dSLRs are better than 15stops and they are not better than 10 stop dSLRs in 8 bit output (best quality JPEG). More than 2 stops in highlight range is useless because it will look like clipped white anyway if you preserve the midrange contrast required for filmic color, and more shadow range is not very useful, because most cameras are reasonably clean there.

A high dynamic range camera in low light wouldn't be any better either, because the noise would be photon limited as it is in most good cameras today. If only 200 photons are hitting your pixel, the noise limit is in the actual light, it's not much use recording it with higher precision. With smaller pixel cameras like Scarlet 2/3" with 3.3um pixels, that would be only about 60photons, so that would be even worse.

HDR is full of limitations like this and it's not the answer to a filmic look. What it can provide at some point is the range of film negative which will allow a freedom in exposure error. It will be great when HDR is achieved in one linear or log exposure (without slopes or multiple exposures), but the final images will not be much better than the 9 to 11 stops cameras we have been using for years.
 
Status
Not open for further replies.
Back
Top