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 :)

Sweet...

1292903_10151877131395126_25205200_o.jpg


RLF/800
 
Just thought this would be good to show, if you react to my (very quick) grade (done in a workflow I would not recomend, but the way I could get it finnished before IBC), which actually works kinda cool as a DCP, but gets a bit chopped off in the h.264.

There is "kinda" a bit of luma info and a few stops in there....

Actually there is other stuff that fascinates me more with the Dragon than the DR... ;) OR the resolution.

The highlight rolloff being one of them
Improvement in whitebalanced images being VERY important
Fidelity, kinda the generally mostly impressing thing.

But those things take a bit more effort and examples to discuss, so... right now.

That will have to wait a little bit.

G
 
I meant to comment on this a couple moons ago... Gunleik this is amazing. The sheer fact that you did what I guess is you is great.
This is not your ordinary nice and cliche demo. But this is no ordinary camera either...thx for this and for having the balls (pun intended) ;)
 
THX!

I have a little something for todays analysis, too...

The beginning of one of the things that really impressed me with the Dragon.

Whitebalancing...
 
Frankly I have had no major issues with RED and skintones generally. On the original R1M they really shone (despite "limited" DR compared to todays cameras..)
On the R1 MX and Epic, it was also not really hard, but there were a few "but"s involved. You had to think about what you did. At least I felt like that to some extent.

If I have interpreted Cionis posts on post-workflow right (and I may well have gotten them wrong), starting off from a "flat" linear or log image has been the way to go.
Now... If I did NOT misunderstand that, that corresponds very well from my workflow findings.

Which could possibly be summed up as.

On set:
Shoot camera flat, to avoid grading in light on a small monitor, lights that are not really there.
WB became a lot better with build 30 (on the R1), still... stay within a "safezone" when you expose and start post to avoid too low saturation in some colors. That safezone (last time I tested for it) was kinda between 4400 kelvin and 6500 kelvin in metadata, independently of "real" temp.
IR filtering sucks for skin tonality...
Of course you need it, but do not put it on as a "default" if you do not.

Avoid heavy underexposure. Especially with "strange" lightsources.

Now...
Let us go Dragon... :)
 
This is "as shot" with a master @ 1.3 with 360 degrees shutter to get more light in but at 5600/800

1266743_10151879463115126_1078377797_o.jpg


Looks kinda yellow/greenish, would you not say?

And a slight bit underexposed.
 
Just for fun, I pipetted the image, and got this result

1262453_10151879462845126_1530118853_o.jpg


That is not really bad for a lowlight low contrast shot....

Skin is pretty much alive.

And this is actually just with a fluorescent lamp a floor above at night....
 
Now. Let us push it a bit...

with these settings

1238000_10151879463165126_1886006545_n.jpg


Result is this:



No CC applied here.

Just WB and a 3 stop ISO push from 800 to 6400.


1270396_10151879463040126_1182349344_o.jpg


To my total surprise, skin is still breathing and alive

I would have expected quite a bit of grading time to make this work
 
And finally, at 3200 ISO, where I landed it...

1265565_10151879463045126_774758397_o.jpg


this is "out of the box" stuff and before any grading...

AND with non-Dragon optimized colrscience and gamma...

Still works kinda ok...
 
And I dunno if there is the bitdepth, resolution, or better compression or whatnot.

But this is pretty much where I think the Dragons fidelity shines.
Well...
In addition to highlights and lowlights rolloff...
 
Back
Top