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 :)

Red Epic Dragon #5232 "Kozmo" - Los Angeles

  • Thread starter
  • Moderator
  • #64
Gratitude

Thank you again guys. Nice to have the support in the back of my mind late at night while testing :)

dio Zafi said:
no dragon yet..but why not kodak stock in the meantime
http://www.youtube.com/watch?feature=player_detailpage&v=GpTwgRk2aUU
Click to expand...

I've got about 15 years of film experience behind me. At this point in my career and considering all things industry-wide, I don't know if film will exist in my future workflows. That said, that's a rad spot. Film will always be my "bar" to judge all things, but in many ways we're past it now. Aesthetically speaking it's still rad.


Peter Lyons Collister said:
Really happy for you Phil.. After the new year and after I get back from Paris let's compare notes. Maybe I should take one of my Dragons to Paris!!!!
Click to expand...

Enjoy the holiday! You've been a busy guys lately. Send a flare and we'll Dragon-chat.


The last couple days have been busy, very, very busy for tests. I've had Dragon footage in several grading programs and suites to see how nicely they play with both .R3D material and 16 bit DPX, TIF, and EXR for workflows that demand that. Also, did a bit more in terms of lens coverage tests. More later this week on that.

However, there are some bigger overall concepts that needed to be tested out.......


Initial IR Tests

Well, this will be one category that separates Dragon and Mysterium-X in a HUGE way. I haven't gone through all of the filters I'll be testing, but I wanted to show you guys this:




That's an ND 1.8, 6 stops. Cheapo brand. Just one I had laying around in a drawer. The lens is at T10. Camera is at ISO 800, exposure compensation on the clean image is using shutter speed/angle. REDcolor3/REDgamma3 at Daylight 5600K.

Okay, so the immediate thing here is I'm not seeing any nasty IR contamination. Which means Dragon's new OLPF IR Cut tech is good. Real good.

Some quick words on ND filters. Moving back to normal NDs is on paper "simple". However, there's crazy variances in mixtures, densities, and accuracy of rating specific filters. This inexpensive ND sort of demonstrates that. Lower contrast is to be expected to a certain degree, especially at such strong filtration, however this brand isn't going to win the beauty contest in terms of best ND out there.

Testing 6 stops of protection and having great results made me fish around for my 9 stopper:




So this ND filter is considerable more expensive and it shows why. Much better contrast is being retained. This shot was at T1.3 in broad daylight. There is a color shift, but what we aren't seeing is any nasty purple-ish contamination coming into the black fabrics. FYI, the thing below the mask is a somewhat fuzzy hat that used to be my "danger fabric" in situations like this.

I can't stress enough how good this performance is. I'll test tungsten tonight, but based on these results Dragon doesn't look like it needs any sort of IR filtration. So now Hot Mirrors, Hot Mirror/ND Combos, IRNDs, or Hot Mirror IRNDs. This is bloody fantastic news. Things are simple again.

Some Quick Words on Color, Dynamic Range, and the New Calibration.

Dragon in many ways will make you rethink how you expose your image. This is mostly due to the additional Dynamic Range. This is my 2nd time seeing the final calibration of the sensor and it looks good. It has it's limits, but damn you can shoot recklessly and I love the highlight roll-off on this camera. ISO performance is an interesting thing and I've got some tests lined up to visualize the ability, but depending on how you shoot I'd say between ISO 250-800 is very safe. ISO 2000 is fine too, but I can see extreme color pushes that could cause some unwanted noise, but under most conditions it's gravy. In terms of usability against Mysterium-X, well ISO-wise I'd say it's a safe 1.3 stops improvement. Sensitivity gains I'll still need to measure for those who need a hard number. It's just not as simple as that really as Dragon has so much more Dynamic Range than Mysterium-X.

There's going to be some room to experiment. I found myself shooting mostly at ISO 500 and 800 so far. Did a bit of 250 and 2000 too. You just need to make decisions really.

That brings me to something else. Dragon's Color. It's gorgeous and you can push it big time. Especially if you expose more in the conservative fashion at lower ISOs. I did about two hours of shooting yesterday with the PL Motion Mount and then did some quick color experiments in the evening. Not going to lie, I'm real happy with what I'm seeing. While it could be considered a subtle change, for those who deal with color all the time it's night and day really when compared to Mysterium-X's color and we don't even have the new color science in there yet.

Just a few grabs from various color tests, also shooting into the sun here and there to get some highlight roll-off going:



Yep. Dragon is fun. It's going to make people think and think differently. It's made some things much easier to deal with while adding some crazy flexibility. Today I put the PL glass down and shot mostly with the DSMC Canon Mount. Dragon has sort of breathed new life into these lenses actually. That 6K frame is just noticeably bigger and that really helps when searching for that intangible 3D depth feeling I'm after.

I'll say this to wrap up my first couple days here. There's a bit of a learning curve. It's not a huge one, but the gains are superb if explored.

That's what I got for now, but there's more coming.

 
Awesome Phil.
 
  • Thread starter
  • Moderator
  • #67
Oooh. Almost forgot to add. I don't use Variable ND filtration much, but I know they are popular among shooters. In the process of testing NDs I did put up a Light Craft Fader ND, which goes up to 2.4 ND (8 stops). No contamination. At full cut max (that area after 8 stops) the image interestingly turns and extremely dark blue, but everything before that cliff looks "as desired".
 
Phil thanks for your efforts. :)
I'm curious to know how many stops of dynamic range you can find in DRagon.
For the rest be really impressive.
have also tried the motion mount how it behaves?
 
  • Thread starter
  • Moderator
  • #73
luigivaltulini said:
Phil thanks for your efforts. :)
I'm curious to know how many stops of dynamic range you can find in DRagon.
For the rest be really impressive.
have also tried the motion mount how it behaves?
Click to expand...

Just posted this bit up on Dynamic Range:

Phil Holland said:
After a few days of testing the new OLPF I found myself thinking about something.

Mainly this image:



Now I know how that was created and I think I have a bead on a Xyla 21 down here in LA, so this might happen in the near future.

The above graphic as far as I know was produced sans color science and calibration tweaks and made back in late 2012. Red stated that we would be getting "part" of that possible dynamic range once the color science and calibration was worked out. Now that we have the new calibration and we have the new OLPF. What does the camera "see?"

That's essentially what's been jumping around in my mind.

Been shooting weird charts now and have tested Dynamic Range with the new OLPF and Calibration in a few different ways. Finding some interesting variances along the way.

Sekonic EPTII Chart.



I'm focused on what we can actually see and use. And also measure. Recently I've seen people discuss "captured dynamic range" and "usable dynamic range". That's an interesting conversation in itself.

In these following graphics you should get a general idea for both though.

Stipulations here first. This is with the latest calibration on with the new OLPF, which has changed things a bit. These were created using REDcolor3/REDlogFilm with and without curves applied as I've been finding the most success of extracting the most out of Dragon this way currently. Linear also has a lot of chocolaty goodness, but most don't work that way.

Each square is one stop apart. Exposure was achieved through one stop increments. Base exposure derived from 18% Gray. Note images will have noise outside of Black Shading Calibration. That's not the point of this particular test. What you are looking for though is clear tonal separation. And you can count it.


Red Dragon REDcolor3/REDlogFilm 21 Step - No S-Curve.




Red Dragon REDcolor3/REDlogFilm 21 Step - With S-Curve.



That's clipped in the first square btw.

Essentially this is what Dragon sees and how it sees it under the current/newest calibration that cameras will be receiving with the new OLPF is installed.

Personal note. Charts don't make compelling motion pictures. Shooting controlled wedge patches also doesn't exactly happen in real life either, which is one reason why I think the Xyla test will be interesting as other chaos can occur utilizing that method. However, what these charts can do is illuminate how the camera system, sensor, and calibration works. Which can inform an individuals exposure choices down the line.
Click to expand...


I'll be shooting with the Motion Mount tomorrow on the new OLPF to test that out. Some of the first footage I shot and released off of Dragon (those leaves blowing in the wind) was all Motion Mount material. Behaves nicely and looks great. Pretty confident the new calibration/OLPF works well with it, but I'll know for sure tomorrow.


Kemalettin Sert said:
Phil can you update thread with new photos of Kozmo :) would like to see how cool it looks
Click to expand...

Kozmo looks about this cool right now while I test out the new OLPF, shot said Dragon with the new OLPF.

Notice that nice clean flare :)



REDcolor3/REDlogFilm, new calibration with new OLPF + one of my custom LUTs.

Looking good.

Here's a few of my ugly mug during today's Dynamic Range testing with various LUTs I've made for a few films.



If you look closely, inside my ear, you can see my thoughts.
 
  • Thread starter
  • Moderator
  • #75
Hi Phil,
thank you for the great tests. in case you have time could you please shoot some footages with people and colorful subjects. (like supermarket full of stuff, streets with people and cars, playground etc .)
i really would like to play with these files to see new calibrated sensor and color response. especially to get hi contrast saturated look.
 
  • Thread starter
  • Moderator
  • #79
Ergin Ozturk said:
Hi Phil,
thank you for the great tests. in case you have time could you please shoot some footages with people and colorful subjects. (like supermarket full of stuff, streets with people and cars, playground etc .)
i really would like to play with these files to see new calibrated sensor and color response. especially to get hi contrast saturated look.
Click to expand...

Just got back to my HQ with a freshly upgraded/New OLPF and Calibration from the camera doctors over at Red. Everyday from here on out you'll be seeing material from the new tech. Which I'm real happy about.

I would stress those who have Dragon now to show the images found here:
http://www.reduser.net/forum/showthread.php?112230-the-New-Dragon-OLPF

to those making decisions about a project's camera selection to ensure them that the New OLPF tech solves the IR reflection/glare issues and in the process the New Calibration provides a approximately 1.3 stops more highlight retention than what existed prior to right now.


JS Fendelander said:
I think the upgrade is worth it just to be able to see your thoughts... :smile5:
Click to expand...

Heh. I'm equally as fascinated by these tools as I am fond of using them. Happy to hear there's a bit of entertainment to be had there :)


Robert Ruffo New said:
Thank you so much for all your very interesting posts. If we upgrade to Dragon, it will be in great part because of you and Gunleik.
Click to expand...

Thank you Robert. Dragon was essentially a no-brainer upgrade for me. Glad this stuff is useful in making purchasing decisions.


I'm going to be a busy shooter for the next 1.5 months leading up to NAB. 3 shoots booked now leading up to a trip across the pond just before the end of March. Also have a few bids on features shooting late Spring/early Summer 2014. Most landing in the 18-35 shooting day schedule. Hoping to land one of those and shoot one more before the end of year if possible. We'll see. I have a couple wildcards that might influence progress on my own projects, one of which I've been working on for a few months in terms of getting it off the ground and running. Once things are a bit more aligned I'll share more about that for sure. Producer meetings soon on that bugger. I've reached out to two good companies and they are on board. This will showcase some intense 4K visuals in a unique way if all goes well.

I'll leave you this evening with this enticing image from my Instragram:



While my camera was getting fondled by the black rubber gloves at Red I've been doing some interesting experiments with various Superresolution techniques and algorithms. I'll post some images soon. I feel A.D.D. might allow for even more insanity and I want to see what that can net me this week once it's released in Redcine-X Pro. The method I'm mostly exploring at the moment involves SR to 12K and downsampling for 8K and 4K finishes as desired. Not a quick process, but the results thus far have been interesting. Very, very interesting.

I'll have a bit of time to shoot some quick material to share this week. So stay tuned.
 
Looking forward to your 12k tests Phil, was trying something similar for very large prints... I do know that by the sound of it A.D.D. will help exponentially in this task already, and we'll make prints to test, but for really large prints, what you are working on here sounds very interesting indeed... ;)
 
You must log in or register to reply here.
Back
Top