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 GOALPOSTS and Exposure Tool

Phil Holland said:
Let's start checking things off though.

- Black shade again and get camera up to operating temperature.

It's hard for me to evaluate this via the internet, but let's check a few variables off the list.
Click to expand...

Hi Phil,

I noted the same , as regards L Goalpost, with 6K FF, T/E Green/Green, 65 Adaptive, RC 5:1, 23.976. My Dragon CF, has been at RED, for a few days, as there was a communication error, per log, showing manifesting as a different sign. I don't know if these signs are related, but will report back, when I get updated, from RED tech.
 
Maybe there is a sensor recal coming when the beta makes it into full production mode. Having 4 black shading temperatures is great.

I've noticed that there is generally much higher goal post noise on dragon over MX in camera, but when checking the files it seems ok. I find the recal (or decision to change the raw view to something friendlier) cough! I'm sure mark was lobbying hard for that. :) ). to raw at 800ISO and 5600k is a big switch as seeing a brighter image when looking at raw view takes some getting used to. I suspect it's just so that RED log view sorta falls into line with other manufacturers and I am not sure what it has to do with the sensors rating.

It seems like files I've seen on the old calibration were cleaner but had the flaring problem, now with the flares gone, it still feels like things need to be tweaked because the images seem a bit noisier in general compared to what was floating around months ago on the old calibration.

I'm still running a bunch of tests to see where (my) sweet spot is.

Battistella
 
The goalposts and exposure check exhibit the same behavior when camera is blackshaded at 65 degrees and brought up to 65 to shoot. That rules out temperature being the cause of this.

The bottom line is that I am 99% sure that the left goalpost fills up much quicker on Dragon. I've shot Epic MX for 2 years and have become very familiar with the metering tools. Within 30 seconds of powering up Dragon and looking at the histogram and goalposts it felt wrong on the low end. Maybe it's the firmware and maybe a side by side with MX will sort it out. But the lowlight goalpost doesn't make sense as per the manual and I can't quite sort out what is going on. Familiar shots and histogram levels are yielding FILLED UP LOWLIGHT goalposts. It is way too easy to fill up the left goalpost.

Hoping to compare with an Epic MX later today...I remember on the MX if the left goalpost was a third full and you toggled exposure check you would see about a third of the image represented as purple.
 
If you cap the camera when you boot up and then turn exposure check on, you can literally watch the purple go away as the sensor heats up. It's a little habit I got into a while back as a visual check to see where I was at before filming.

@Eric. You are right, I have been shooting in this one location for quite a while on MX and when I was in there last week with dragon for the first time I was really noticing how full the noise goal post was. Maybe that particular tool just needs to be tweaked to be a bit more representative or maybe it's accurate and there needs to be some sensor tweaking with the new OLPF/calibration combo.

Battistella
 
David Battistella said:
I've noticed that there is generally much higher goal post noise on dragon over MX in camera...
Click to expand...

Well, there *are* about 5.59 million more pixels produced by the Dragon sensor to potentially be noisy... :)
 
I noticed my goal post filled on Dragon too when on MX it wouldn't of been... I think with higher sensitivity comes larger warnings of clipping. I had everything but the 2k on at my set, and was wide open and it still was fully clipping, which surprised the hell out of me, but I am going to do more tests tomorrow. I think the more sensitive it is the more it picks up clipping too, so its just a warning I didn't consider my shots hopelessly under exposed, and they look pretty fine to me...Its just a really fat negative I think, so fat even MX shooters haven't gotten used to it yet... Try lighting the under exposed parts and crushing in post, I am sure it won't clip then... also consider running a production build and not beta, just to rule that out as well :)
 
If the bottom goal post is full of red but you're not seeing any purple when checking your exposure, then clearly there's a bug or technical issue going on. Black shading or temperature would be irrelevant. The exposure tools aren't working how they're supposed to, plain and simple.
 
I just joined up with Eric and another DP at his house to see the Epic MX vs Dragon side by side tests on this. Stills & r3D's coming in another thread.

On the Dragon, either the goal posts are off or the exposure check tool is off. They're not showing anywhere close to the same thing. Something is wrong here.

But the biggest surprise was how much significantly cleaner the noise is on the MX, at any ISO. My fear is the noise goal posts on the Dragon is actually correct, because there's that much noise.

dragon_mx_noisedifference.JPG


Dragon on the left, MX on the right.
 
Eric Haase said:
No. Dragon is on the left.
Click to expand...

Okay, I know this is about the noise in the shadows etc., but the highlights on the right
looked more like Dragon. Oh well.

Perhaps the camera needs to be sent in for repair etc.?
 
Please do this test at 60 or 50p with a 180 degree shutter!!!

CJ Roy said:
I just joined up with Eric and another DP at his house to see the Epic MX vs Dragon side by side tests on this. Stills & r3D's coming in another thread.

On the Dragon, either the goal posts are off or the exposure check tool is off. They're not showing anywhere close to the same thing. Something is wrong here.

But the biggest surprise was how much significantly cleaner the noise is on the MX, at any ISO. My fear is the noise goal posts on the Dragon is actually correct, because there's that much noise.

dragon_mx_noisedifference.JPG


Dragon on the left, MX on the right.
Click to expand...
 
Last edited:
CJ Roy said:
This is the goal post issue. This is different than the noise and is regarding the tools, which seem to be not working properly. Merging may get this problem lost.
Click to expand...


Cool. I see your point.
 
You must log in or register to reply here.
Back
Top