Welcome to our community

Be a part of something great, join today!

Log in
  • 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 :)

Dragon dynamic range, profiling for Sekonic

The bidding stands at 16.3 stops, ladies and gentlemen, do I hear 17 from the big fellas in the Red seats? 18, who'll give me 18?

16.3 stops going once, 16.3 stops going twice,….. ( :
 
About Dragon DR - I'll say this:

I've spoken to two colorists recently who have played with Dragon footage, and they were impressed. Both of these gentlemen are well respected, and not particularly fond of RED. They both prefer Alexa. However, for the first time, both came to me and said "WOW, Dragon is pretty good!". They both loved how the highlights responded. That was their beef with MX all the time - they both said highlights clipped faster and harder than Alexa, and as colorists it really bothered them. At least the footage they got, if something was clipped...it was gone...whereas they both said that Alexa was more like film...if it was clipped...you could still get it back.

Now with Dragon...they are both seeing the same. Not just greater plasticity in the image, but fantastic highlights that are much harder to clip, and provide them much more flexibility in post. They were both very, very impressed.

On that note as well, I'll tell one more anecdote. About DIT's.

A good friend of mine is an Epic owner, and has had to suffer many a DIT complaining to him about Epic this...Epic that...they love to tease him and give him shit. This friend of mind called me last week and said: "Dude, I'm getting texts from all my DIT friends, same one's that have been giving me shit about Epic. They are fucking loving Dragon!! I've gotten a bunch of texts already while they are on set. Stuff like - sick skin tones, amazingly clean at 2000iso, Dragon is the real deal!". I've never heard my friend sound so, so happy.

So, I'm my little corner of the universe, where I'm starting to get trickled feedback from pros in the trenches, word on the street is Dragon is a home run. The ultimate curmudgeons, the colorists and DIT's who are first to see all the small flaws in footage, appear to be very excited with what they are seeing. So I'm genuinely very happy and optimistic about this. I'm still a long way off from my upgrade myself (about to upgrade to Epic first), but it makes me really happy to know RED has pushed themselves to the limit....and appear to have delivered an intriguing, spell binding new sensor.
 
Last edited:
Nick Morrison said:
...was more like film...if it was clipped...you could still get it back.
Click to expand...

I like your post, but this segment is a non-sequitur for me, as in, if it is clipped, you CANNOT get it back.

:)
 
Martin Stevens said:
I like your post, but this segment is a non-sequitur for me, as in, if it is clipped, you CANNOT get it back.

:)
Click to expand...

Haha. I hear you, fair point! I guess they're looking for more latitude in the highlights, and now with Dragon they got it!
 
Nick Morrison said:
Haha. I hear you, fair point! I guess they're looking for more latitude in the highlights, and now with Dragon they got it!
Click to expand...

Yes, indeed.

For me I have pretty much experienced Film Negative to have roughly 14 stops of DR, with 8 stops of highlight latitude and
6 stops of shadow latitude. And, now with Dragon we can get 8 stops on both sides if we wish.
 
Since we are ( or will be soon enough) heading into somewhat uncharted territory with these growing DR performances, where do you all think the practical breaking point lies: the point that does not warrant more DR just for the sake of catering to unique circumstances? When is enough enough? And what techniques/ solutions, including post, do you foresee making much longer scale images desirable and manageable? We can always crank the curves, so to speak, but when do we head into "reconstructing" values rather than retrieving them? Or is that not an issue?
 
Bill Anderson said:
Since we are ( or will be soon enough) heading into somewhat uncharted territory with these growing DR performances, where do you all think the practical breaking point lies: the point that does not warrant more DR just for the sake of catering to unique circumstances? When is enough enough?
Click to expand...

I would say 16 stops should be enough but humans are often tough to please.

Bill Anderson said:
And what techniques/ solutions, including post, do you foresee making much longer scale images desirable and manageable?
Click to expand...

Longer DR scale does not necessarily mean harder and less manageable for post work, it can also mean just more natural looking in falloff to darkness and whites.

Increased captured DR does not make images undesirable, improper picture manipulation used to represent it does.


Bill Anderson said:
We can always crank the curves, so to speak, but when do we head into "reconstructing" values rather than retrieving them? Or is that not an issue?
Click to expand...

When the image stops appearing natural, balanced, pleasing and/or out of context.
But that is a subjective, relative and variable LUT.

Yes it is an issue if miss-used. A bunch of those can be expected with newly available DR but as in similar cases that is likely to settle down once the infatuation with new creative potential passes.
 
Bill Anderson said:
Since we are ( or will be soon enough) heading into somewhat uncharted territory with these growing DR performances, where do you all think the practical breaking point lies: the point that does not warrant more DR just for the sake of catering to unique circumstances? When is enough enough? And what techniques/ solutions, including post, do you foresee making much longer scale images desirable and manageable? We can always crank the curves, so to speak, but when do we head into "reconstructing" values rather than retrieving them? Or is that not an issue?
Click to expand...

I would say 30 ish should be more than enough. If you want to emulate glare and lens flares you need to be able to record the detail on the elements. Also you get into new creative usage. For instance maybe you want a 2.8 Aperture on a sunny day. Normally you would need ND but with 30 stops you could just use ND as metadata. I would also like the opposite I would love to be able to have deep focus at night. Shoot F/12 but low noise.
 
This is awesome stuff. Thanks for doing these tests guys!
 
You must log in or register to reply here.
Back
Top