Welcome to our community

Be a part of something great, join today!

Register 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

Explain me something.

Why on the chart do I see -8 (where it starts to lift the curve) to + 6.5 (where it stops to read an increment)? What are those 14.5 stops?

Thanks

Pat
 
In the sekonic software you can set your own preferred clipping points within the overall dynamic range, the meter uses these and your own preference of mid grey too to customise its response in use. I prefer it to allow me some headroom, that way, given the broad workable range of dragon I'm able to waste a stop into my choice of highlight roll off characteristic if I need to and likewise at the bottom.
 
So now, with the spotmeter (reflected light) function John can take a quick reading of a scene's highlights and ( since the reflected light reading denotes mid gray) open up from there - in John's case, 6.5 stops - and know the highlights values are well contained. A very quick and accurate way of keeping your highlights in play. And the beauty of the spotmeter is that you can read where all the other values are falling on the curve, once the highlights have been placed.
Very nice to have such a big room to play in. ( :
 
John,

You chart maps the image with no roll-off in the highlights, yet the Dragon imagery looks like
they built some roll-off in the highlights. ???
 
John Marchant said:
Broke out my Sekonic exposure chart and calibration software to get my meter dialed in for Dragon. Nice to see the DR validated another way :) This is the most pessimistic I could get the software to read... I could have posted the version where the software announced 17.1 stops DR... mid 16s was most common with the various methods I checked.

I've actually found I'm active much more with the meter on Dragon. The broad exposure latitude (broader than many typical scenes), and the friendly noise floor and clipping mean that I no longer live in mortal fear of the goalposts... and the meter is a nicer way to work all told.

sekonic%20dragon%20800%20iso.jpg


( Hey, look, its Dragon info that isn't sexy girls ;) Normal service will resume shortly... )
Click to expand...

Gunleik's Dragon vs M-X side by side test shows 16 stops of DR as well.

Thanks John for well-proven RED's claim of 16 stops.
 
Martin Stevens said:
John,

You chart maps the image with no roll-off in the highlights, yet the Dragon imagery looks like
they built some roll-off in the highlights. ???
Click to expand...

The charts were posted RLF to best measure the available range as the sekonic software only takes 8 bit input images, hence the apparently hard top end. Other more viewable gamma curves (preset or your own) typically roll off more nicely. Another appealing part of the Dragon top end is also the way the channels align nicely in overexposure, which the chart doesn't really illustrate...
 
Very useful information to share.

never let anyone tell you where or where not to point your camera. (Unless they are holding a firearm)

battistella
 
Well graph or not, there's four images right there labelled up with a 16 stop spread and data in the lightest and darkest. A 13.5 stop camera may have trouble there..

can't see your link sadly...
 
Gavin Greenwalt said:
So am I reading this correctly in that middle gray with equal dynamic range on both ends would be closer to a rated exposure of iso 484?
Click to expand...

More like ISO 2400
 
Martin Stevens said:
More like ISO 2400
Click to expand...

How'd you reach that conclusion Martin?

According to John's graph and accompanying pics with 8 stops over and 8 stops under, 800 provides a fairly well balanced detail in shadow and highlight images…. Maybe a tad skewed to highlights :but surely close, as Gavin suggests.
 
Just as a sad comparison using my D800: I place specular highlights 3 stops above (my very accurate spot meter's) middle grey. 3 lousy stops, and I like to live dangerously. A proverbial hair more and it's all gone Pete Tong.
 
+1.

I'd like to see comparison with an MX curve.

AndreasOberg said:
That is great to see! What would such a curve look for the MX sensor? Landing around 13.5 stops?
/Andreas
Click to expand...
 
Very interesting. Thanks for posting John! Be curious also to see at lower and higher ISO's as well (250, 1600, etc).

Thanks for taking time to do this...very generous.
 
John Marchant said:
Broke out my Sekonic exposure chart and calibration software to get my meter dialed in for Dragon. Nice to see the DR validated another way :) This is the most pessimistic I could get the software to read... I could have posted the version where the software announced 17.1 stops DR... mid 16s was most common with the various methods I checked.

I've actually found I'm active much more with the meter on Dragon. The broad exposure latitude (broader than many typical scenes), and the friendly noise floor and clipping mean that I no longer live in mortal fear of the goalposts... and the meter is a nicer way to work all told.

sekonic%20dragon%20800%20iso.jpg


( Hey, look, its Dragon info that isn't sexy girls ;) Normal service will resume shortly... )
Click to expand...

Amazing result, John!!!!

Everything beyond 14 stops of DR I call it a flexibility.

And that's the first step where I've found Dragon beats Alexa.

DR together with colour (skin & sky) is called the image quality.

The second is a spatial resolution.

The third is a smaller body factor.

The fourth is affordability.

So with Dragon we would need DSC Labs latest high dynamic range test system - the Xyla-21 with 20 f-stops of dynamic range (21 steps)!!!

Xyla-21.jpg

the Xyla-21 with 20 f-stops of dynamic range (21 steps)!!!

http://dsclabs.com/xyla.htm
 
Why does the curve seem to fully flatten to horizontal by around -7 to -8? Would be nice to see that part of the curve in more detail. What is the software seeing that we can't? If all I saw was that graph I'd say it looked like it indicated about 14 stops.
 
What the software is looking at for the toe of the graph is the final darkest image.you should be able to make out a number of 1/6th stop chips in that picture, even in the squished screen grab. They're clearer to the eye in the image than they are on the graph. If you bump ISO you lift the toe of the graph and it becomes more readily apparent. I totally agree that the graph is questionable down there, that's why I included the source images. If you feel inclined you can count the stops the long long way on those :)
 
took the data from the site of Ryan Walter.

Finally,
Sekonic on Arri Alexa, are between 12 and 13 stop

alexa_zpsd1c3617d.png
[/URL][/IMG]
 
luigivaltulini said:
took the data from the site of Ryan Walter.

Finally,
Sekonic on Arri Alexa, are between 12 and 13 stop

alexa_zpsd1c3617d.png
[/URL][/IMG]
Click to expand...


"Common knowledge" is that Alexa is about 13,5-14

Epic is about 12,5 - 13

From when I checked the MX vs Alexa, noise was pretty similar at similar exposures within their common range. That Alexa has "more" in the top and that they are pretty similar in the bottom, BUT that the lower exposures are mapped to a bit higher luma-values on the Alexa vs the MX.

Seems that that mapping-difference is still present in Marks shots, according to Sanjins posts. Does not really say a lot about sensitivity, though, jut where the signal is mapped.

Look forward to AB them myself...
 
Last edited:
You must log in or register to reply here.
Back
Top