Welcome to our community

Be a part of something great, join today!

  • 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 :)

The toe and the shoulder Dragon DR

The "rolloff" thingy is a consequence of the cineon curve. And is "right" for RLF.
It's also nice that it's extended into most of the other gammas, like the RGs.

For linear... Not sure.
I can see why I'd like that to be actually linear.
 
Integer linear mode is linear unless you gain up ISO when the FLUT curve kicks in to try to preserve highlights. If you don't want that, use "exposure" instead. That said, it's not the most appropriate linear mode any more, and has been superseded by the EXR exports to half-float which are always nice and linear. If you want linear use EXR please.

Graeme
 
Integer linear mode is linear unless you gain up ISO when the FLUT curve kicks in to try to preserve highlights. If you don't want that, use "exposure" instead. That said, it's not the most appropriate linear mode any more, and has been superseded by the EXR exports to half-float which are always nice and linear. If you want linear use EXR please.

Graeme

Aha, Thanks for that info did not know that. Have not tested yet, but the exr's keep timecode and clip name etc just as the dpx's I assume?
 
You've got to remember there's legacy support for older ways of working that we tend not to eradicate from the code because some people like / rely on them. But we do try to advance how things work forwards and for any kind of linear work, EXR export is the way to go. Integer linear formats are not really all that suitable for much other than the odd bit of debugging.

Graeme
 
I am a little disappointed in OLPF2. You need to rate the camera at 400ISO. Yes - highlight roll off is awesome (and the extended range at that end). So Dynamic range is sorted. Now what about sensitivity? For low light applications I would not pick Dragon OLPF2.
 
So please excuse my ignorance, but is this thread now just another extension of the ongoing discontent of where Dragon has actually landed? really 2 OLPF's.,250asa, , noisy blacks so no real dynamic range advancement. Please tell me im wrong!
 
Doubling the amount of light to make up for the limitations that OLPF2 imposes on the sensor. Fantastic...
The decision to switch to this OLPF might help sink this camera. That would be a shame. I hope I'm wrong.

AGREE 100%
 
So please excuse my ignorance, but is this thread now just another extension of the ongoing discontent of where Dragon has actually landed? really 2 OLPF's.,250asa, , noisy blacks so no real dynamic range advancement. Please tell me im wrong!

Sergio, have been shooting dragon? The dynamic range increase is very real. I find it great at about 800 ISO. Same as the old days. Plenty fast. 250 is just a hyperclean magical point.
 
Integer linear mode is linear unless you gain up ISO when the FLUT curve kicks in to try to preserve highlights. If you don't want that, use "exposure" instead. That said, it's not the most appropriate linear mode any more, and has been superseded by the EXR exports to half-float which are always nice and linear. If you want linear use EXR please.

Graeme

Graeme, can you please write up a description of what all the dials in RCX do, and how they effect the image? I have been using it for 5 years now, and really other than ISO and Whitebalance dont really know what each dial does, in the definition sense.
 
Sergio, have been shooting dragon? The dynamic range increase is very real. I find it great at about 800 ISO. Same as the old days. Plenty fast. 250 is just a hyperclean magical point.

Sorry for veering off topic David - but I need to reply.

Yes. The DR is real. The low light ability is not. I shot a"night for night" scene. 1.4 on Cooke S5's. City lights in background. Blacks were very noisy. And no - I couldn't "light" the blacks of a city skyline unfortunately. I am hoping to do a side by side (C300/Alexa/F55/V1OLPF and V2 OLPF) in this kind of scenario.
 
I apologize, my audio background and limited RED background keeps pulling me back to this point on MX and now Dragon -- why are people using histograms that are impacted by the ISO setting? Isn't the RAW histogram -- and especially the stoplights and goalposts -- the only way to properly understand what DR your camera is offering/capturing at any given time?

Unless this is another legacy issue -- to use Graeme's term -- I really can't see why there is so much interest in ISO. If it is truly a post-capture data modifier then it has nothing to do with the DR of the camera.
 
I apologize, my audio background and limited RED background keeps pulling me back to this point on MX and now Dragon -- why are people using histograms that are impacted by the ISO setting? Isn't the RAW histogram -- and especially the stoplights and goalposts -- the only way to properly understand what DR your camera is offering/capturing at any given time?

Unless this is another legacy issue -- to use Graeme's term -- I really can't see why there is so much interest in ISO. If it is truly a post-capture data modifier then it has nothing to do with the DR of the camera.

This is getting way off topic - feel free to PM for further discussion - But we need RC3 and ISO (non RAW "looks") to help us gauge how things are looking on set. We always "check" the RAW to be sure. But DR has everything to do with it. I would tend to rate the camera at 400 if I was in low light and 800 if I was outside during the day. Yes - the cameras DR is the same in either case - but it forces me in to protecting important information in each scenario (400 protects for noise floor in dark scenes - 800 protects highlights in bright daytime scenes). I will always check the RAW from time to time to see where I'm at. Not sure if that helps?
 
Sorry for veering off topic David - but I need to reply.

Yes. The DR is real. The low light ability is not. I shot a"night for night" scene. 1.4 on Cooke S5's. City lights in background. Blacks were very noisy. And no - I couldn't "light" the blacks of a city skyline unfortunately. I am hoping to do a side by side (C300/Alexa/F55/V1OLPF and V2 OLPF) in this kind of scenario.

v1 is obviously the better in lowlight...
 
As things stand today I assess the particular shooting situation and adopt one of two strategies:

1) rate the Dragon V2 at 500ISO and try to keep a little dead space between the lowest histogram value and the underexposure goalpost. This approach has so little noise I can process footage with RR-X and ignore DEB (I can always re-process a couple of shots in software with DEB if need be).

2) rate the Dragon V2 at 800 and account for the time it will take to process with DEB in software at conform/grade/finish.

Kodak 5219, arguably the reference film stock of the last decade, has an EI of 500 and IMO has a similar "texture" in the shadows to Dragon V2 at EI500. Yes, digital noise is not the same as film grain, but that's a whole other topic - my point is, even with OLPF V2, the Dragon has good signal to noise ratio - just not "clean 2000ISO".

Cheers - #19
 
Sorry for veering off topic David - but I need to reply.

Yes. The DR is real. The low light ability is not. I shot a"night for night" scene. 1.4 on Cooke S5's. City lights in background. Blacks were very noisy. And no - I couldn't "light" the blacks of a city skyline unfortunately. I am hoping to do a side by side (C300/Alexa/F55/V1OLPF and V2 OLPF) in this kind of scenario.

Footage of your Scene ? Raw? Thx

Night for Night here too ...
https://www.youtube.com/watch?v=NhGjQ2j7oP8&list=UUANSkXF6yhB2Lfke_G2149A
 
As things stand today I assess the particular shooting situation and adopt one of two strategies:

1) rate the Dragon V2 at 500ISO and try to keep a little dead space between the lowest histogram value and the underexposure goalpost. This approach has so little noise I can process footage with RR-X and ignore DEB (I can always re-process a couple of shots in software with DEB if need be).

2) rate the Dragon V2 at 800 and account for the time it will take to process with DEB in software at conform/grade/finish.

Kodak 5219, arguably the reference film stock of the last decade, has an EI of 500 and IMO has a similar "texture" in the shadows to Dragon V2 at EI500. Yes, digital noise is not the same as film grain, but that's a whole other topic - my point is, even with OLPF V2, the Dragon has good signal to noise ratio - just not "clean 2000ISO".

Cheers - #19
Shooting at 500 while making sure to keep dead space on the left of the histogram and still getting noise (even if very little) sounds like a long way off from "clean 2000"? and why is there noise at all if i am protecting the shadows (leaving space on the left)? I shoot my Epic in a similar way (always protect the shadows) and get absolutely no noise, zero. What would i have to shoot Dragon at to get no noise, texture, anything?
 
Yes. The DR is real. The low light ability is not.

Exactly my findings.

Feed the Dragon with light and you get superb pictures. I set the ISO to 400. 250 is super clean. 800 a tad to rough, especially when grinding on the edge. 320-400 makes you happy for most situations and is something I was used to a while back when changing a mag took longer than 5 seconds.


Hans
 
Back
Top