PDA

View Full Version : Dragon sensor noise, fix yet ?



Shermen L
08-27-2014, 01:57 AM
When I compare the Dragon sensor of Scarlet Dragon and the MX sensor of Red One both rated at iso 800 with same aperture, I found the noise in Dragon sensor is more apparent than MX sensor in dark area. I found this issue has been discussed for a while before, just wonder if this problem has been fix yet with any updated firmware in these months ?


Since I will shoot with both camera, the noise issue will be a big obstacle for me to match the cut and so do the postproduction : (

Álex Montoya
08-27-2014, 02:51 AM
I really doubt that noise will be an obstacle matching Dragon and MX at 800.

Shermen L
08-27-2014, 03:14 AM
The quality is not what i expected if I shoot at low light because of the noise in Dragon sensor, even it does have good DR , hope this issue can be solved

Björn Benckert
08-27-2014, 03:47 AM
I dont have my MX around any more as it's now a dragon. But I doubt what you are saying is right. and with ADD and the other (forgot what the name) fix for lowlight noise I think the difficulty would be to match the MX to the better dragon quality.

Christoffer Glans
08-27-2014, 04:16 AM
The tools for using D.E.B and ADD will come to other softwares as well (like resolve and After effects) so it will not be as much of a problem later on. As for low light, I think people unintentionally underexpose with the dragon because they think it's a low light sensor, which it's not.

Björn, we could do a test between my MX and your Dragon to check just how it performs in low light.
Also I can do regular post enhancement on it to see which gives the best result, doubt it's going to be my MX even in low light.

Shermen L
08-27-2014, 05:22 AM
Since I probably hurry for the production, I don't think I can wait for these tools(software) officially released , here's the attachment of the test, I'm not sure if the D.E.B can solve the issue since the difference is quite obvious, in the worst situation, I might have to return the camera since it's not fit what I expected if I film in low light situation

[
IMG]http://i1278.photobucket.com/albums/y508/Shermen833/RedoneMX_zpsb629cdcc.jpg (http://s1278.photobucket.com/user/Shermen833/media/RedoneMX_zpsb629cdcc.jpg.html)[/IMG] http://i1278.photobucket.com/albums/y508/Shermen833/ScarletDragon_zps2db1287f.jpg (http://s1278.photobucket.com/user/Shermen833/media/ScarletDragon_zps2db1287f.jpg.html)

Lakis Amarantithis
08-27-2014, 07:27 AM
I noticed that if you shoot with some lenses with the Dragon you have a lot more noise than the MX and with other lenses much less noise than the MX!

I know it does not make sense but I noticed just that. I decided to use more lights to eliminate the noise with some lenses. I think it has to do with the olpf but what I know? Not an expert.

David Mullen ASC
08-27-2014, 07:33 AM
Since I probably hurry for the production, I don't think I can wait for these tools(software) officially released , here's the attachment of the test, I'm not sure if the D.E.B can solve the issue since the difference is quite obvious, in the worst situation, I might have to return the camera since it's not fit what I expected if I film in low light situation

[
IMG]http://i1278.photobucket.com/albums/y508/Shermen833/RedoneMX_zpsb629cdcc.jpg (http://s1278.photobucket.com/user/Shermen833/media/RedoneMX_zpsb629cdcc.jpg.html)[/IMG] http://i1278.photobucket.com/albums/y508/Shermen833/ScarletDragon_zps2db1287f.jpg (http://s1278.photobucket.com/user/Shermen833/media/ScarletDragon_zps2db1287f.jpg.html)

The two shots don't seem to be exposed or processed to be the same brightness. If you darkened the Dragon frame to match the MX frame, the difference in noise would be less, or vice-versa, brightened the MX frame to match the Dragon frame.

Jan Reiff
08-27-2014, 07:40 AM
please have a look here

http://www.reduser.net/forum/showthread.php?120906-Dragon-noise-do-i-smth-wrong-!

Brian Merlen
08-27-2014, 10:18 PM
not to question you, but you sure the settings matched and were processed the same? post the r3ds and metadata...maybe its a potential hiccup along the way? cuz damn that looks pretty bad...did you blackshade both?

Gavin Greenwalt
08-27-2014, 11:48 PM
Did you let your sensor warm up? I was shooting my first full project on Dragon today and when I turned on the camera it was bright red noise everywhere. Terrible, unusable. Once it got up to 47° the noise went away. I almost never worried about blackshading with MX but with Dragon it seems critical. No wonder RED included a sensor heater to ensure it maintains temp.

Andrew Usher
08-28-2014, 01:16 AM
Yeah i am noticing how much more important it is to black shade and have camera to temp than i did with my epic.

Timur Civan
08-31-2014, 10:32 AM
Keep it on adaptive mode for the fan.

Let the camera run recording, for about10 minutes. Basically fill up a card @6kFF at3:1 till the camera hits 65.

This bring everything up to recording temprature. Then do your blackshade with the adaptive mode set to 65. This is its optimal point.

You should be good

Charles Bergquist
08-31-2014, 11:09 AM
Keep it on adaptive mode for the fan.

Let the camera run recording, for about10 minutes. Basically fill up a card @6kFF at3:1 till the camera hits 65.

This bring everything up to recording temprature. Then do your blackshade with the adaptive mode set to 65. This is its optimal point.

You should be good

How are you getting 3:1 6K FF ?

Never thought about rolling for a bit to warm the camera up, I'll start implementing that method.

Stacey Spears
08-31-2014, 11:21 AM
One thing I will add is that the iPad sucks for critical viewing. It's hard to see the noise in the posted examples on my iPad. When I go to my desktop, the noise is much more visible. I bring this up for those trying to view sample noise images online.

As the camera warms up you can really see the image on the LCD change until it hits calibrated temp.

David Battistella
08-31-2014, 11:28 AM
This one is easy.

When you turn the camera on and it is capped turn on the exposure tool as well. You will see the screen turn purple from left to right. When the screen is completely purple you will be at operating temperature.

As for Black Shading. There is no "optimal point" as the optimal point is the conditions you are in and the calibration to go with that calibration. There is a goos working range and adaptive mode does do a great job keeping your sensor at a consistent temperature The best thing to do is to check to see that your temp indicators are

GREEN/GREEN

and not

GREEN/YELLOW

GREEN/RED

Re telling you that your calibration is not matching your exposure time.

Red has a great page on this stuff, best to go to the source.

http://www.red.com/learn/red-101/black-shading-calibration


Battistella

Timur Civan
08-31-2014, 10:22 PM
How are you getting 3:1 6K FF ?

Never thought about rolling for a bit to warm the camera up, I'll start implementing that method.

Sorry 5:1

Björn Benckert
09-01-2014, 03:54 AM
This one is easy.

When you turn the camera on and it is capped turn on the exposure tool as well. You will see the screen turn purple from left to right. When the screen is completely purple you will be at operating temperature.

As for Black Shading. There is no "optimal point" as the optimal point is the conditions you are in and the calibration to go with that calibration. There is a goos working range and adaptive mode does do a great job keeping your sensor at a consistent temperature The best thing to do is to check to see that your temp indicators are

GREEN/GREEN

and not

GREEN/YELLOW

GREEN/RED

Re telling you that your calibration is not matching your exposure time.

Red has a great page on this stuff, best to go to the source.

http://www.red.com/learn/red-101/black-shading-calibration


Battistella

Just a little correction. Not shutter angle, but exposure time.

David Battistella
09-01-2014, 03:59 AM
Just a little correction. Not shutter angle, but exposure time.

Corrected. I get caught in terminology sometimes. :)
It's a very important distinction.

Fixed it.

Battistella

Dave Summers
09-05-2014, 03:50 AM
not to question you, but you sure the settings matched and were processed the same? post the r3ds and metadata...maybe its a potential hiccup along the way? cuz damn that looks pretty bad...did you blackshade both?
Like he said. The Scarlet Dragon footage is quite a bit brighter(higher ISO?) and therefore noise will be more apparent. Also the Scarlet footage looks more green, maybe wrong color temp? You need to do this test again and make sure all conditions are exactly the same for each camera(same ISO, same color temp, same lens, same iris setting, etc.)