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 :)

Scarlet Dragon Resolution vs. compression decisions

We just did tests with scarlet mx vs epic dragon. 4k 6:1. And 6k 5:1 I will try and find somewhere to host the tiffs for you all to see. The detail level was basically indistinguishable. We were very surprised
 
Ryan Hamblin said:
We just did tests with scarlet mx vs epic dragon. 4k 6:1. And 6k 5:1 I will try and find somewhere to host the tiffs for you all to see. The detail level was basically indistinguishable. We were very surprised
Click to expand...

Based on the lenses you list in your signature, you were really testing the quality of your lenses, not the quality of the sensors/compression.
 
We used Nikon glass. Though surprisingly I've had the rokinons play nice with ultra primes once stopped to a 4. The dragon still was marginally better at color and highlight retention
 
Here is a link to the tiffs

https://copy.com/iTtrw8OJWzTpWBok

we tried to match frames as best as possible between the two cameras. They were both set to 640ISO at a 5 or 5.6 depending on the setup. both redgamma 4. the scarlet is red color 3, the epic dragon is dragon color. the scarlet was 4k at 6:1 ratio, and the dragon 6k at 5:1 ratio. The A roll files are Dragon the S roll are scarlet. The first frame of each set up is unsharpened, the second is with sharpening. Both cameras were white balanced in redline x off of the white shirt. Both shot on the same nikon 24-70mm zoom and with a .9 ND filter on it.

When viewing these on a 4k monitor its really tuff to tell the difference. The dragon is still a bit better when you go and pixel peep it. I just don't think its by a ton.

Look forward to seeing every one's thoughts.
 
This is kind of backwards though from what the OP was asking about. The question was whether 8:1 5k was better than 5:1 4k. You both increased the resolution and lowered the compression. You also changed sensors between MX and Dragon? Pretty hard to make too many insights with that many variable changes.


The detail level was basically indistinguishable. We were very surprised
Click to expand...
EDIT: Also your framing is totally different. That makes comparing sharpness pretty much impossible as well.

I'm not at all surprised since you were effectively just using a 1:1 crops of both. To really get a good sense of resolution you have to match the framing and then supersample the 6k down to 4k or the 4k undersample up to 6k and compare at a constant resolution. It's true, given any crop of a 6k and a 4k image... I would expect similar sharpness between pretty much all camera systems to be extremely similar.
 
The framings were quite similar. Assimilate as we could get without a monitor and we did view all tests on a 4k display with the 6k down sampled. The color and dr was not nearly as major as we had figured we would see. The main thing I walk away with is no matter what red your using your really dealing with big quality material
 
I've been playing with some R3Ds from another thread and I'm quite impressed by the noise and compression on Dragon. Highlights are really better too. Here are some enlargements of MX vs Dragon noise (it was shot by Werner Jauch on scarlet dragon in 4k 8:1 compression), i pushed iso and contrast (and sharpeness for 2 of them). There is a highlight comparison too (that is obvious on the side of the swimming pool, everything was shot at the same aperture).
https://dl.dropboxusercontent.com/u/90344707/MX DRAGON.zip http://www.reduser.net/forum/member.php?32217-Werner-Jauch
 
Do you think a hypothetical third party, "unsoftened," OLPF in the new interchangable OLPF system would do much to improve the image quality of cropped resolutions, such as 2K, disregarding the potential moire?
 
You must log in or register to reply here.
Back
Top