Welcome to our community

Be a part of something great, join today!

Register 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 :)

Black Shading Nightmares

Rick,
What you're seeing is the sensor at a very cold state compared to what it was black shaded at.
If the ambient temperature is constantly cold, then you'd want to black shade at the colder temp.

Martin,
The sensor in Epic is running at over 30x faster than a 5DII.
Running a sensor at high speeds makes things like black shading exponentially
more complex to both create the black shade as well as apply it in real time
at high frame rates.

Andrew,
1.7.10 and later has a more complex black shading mode that is necessary
for high speed sync. In upcoming builds we'll be adding more options for
black shading that will both improve the time it takes as well as flexibility.
 
Deanan, What's strange is that I had the Epic running for 3+ hours straight and then I noticed this issue after I cranked up to 2000 iso and had the lens cap on. I re-did the black shade, and the gradient effect was horizontally flipped. The cold / warm test was done late last night. It seems inconsistent. Any idea what this could be? I'm letting the camera roll right now for 40 minutes straight and am going to speed up the footage to see the gradient dissipate over time. Then I'm going to black shade again, let it cool down for a few hours and try the test again in the morning.
 
I'm not anywhere as knowledgeable as Deanan of the workings of Epic sensor, but things that I can think that can affect a sensor surface from side to side are:
How is camera stored? Upside up, or on side? Have camera been close to strong magnetics? Even if you are sure, have you tested blackshading with a black cloth over camera just to make sure this is no light leak (take care so it don't overheat)? Have you tried 2.02 firmware?

My guessing is that the sensor "bakground noise" gets louder the hotter is gets, and making a blackshade on a hot sensor removes more "noise" and make a cleaner image, but at same time make it tiny bit less sensitive and work less good at cold temps. But a hot-blackshaded camera will not show picture correct until it get hot. (Very apparent on some Phantom cameras)

I treat camera as unexposed film, because it's just what it is. You only have this sensor and this should be good foreveeeer*!
In any case I would get a support ticket for this.

*until you can upgrade to Dragon :)
 
Rick Darge said:
Deanan, What's strange is that I had the Epic running for 3+ hours straight and then I noticed this issue after I cranked up to 2000 iso and had the lens cap on. I re-did the black shade, and the gradient effect was horizontally flipped.
Click to expand...

That depends on what temp the earlier blackshading was done at.
It sounds like it was done at a higher ambient temp.

Email me a single frame r3d.
 
I will get you an R3D shortly

Question, since black shading calibration gives you two calibration maps, a factory setting and a user1 setting, could you theoretically make the factory setting setup for indoor shooting and user1 setup for colder outdoor shooting?
 
My guess is that the black shade may actually be trying to warm up the sensor to get more data points for hotter sensor conditions.
Otherwise, a black shade should take a couple of minutes max, even with a ton of frame averaging. I have done code to black shade sensors, I see no reason other than what I describe.
-Les Dittert
 
Rick Darge said:
I will get you an R3D shortly

Question, since black shading calibration gives you two calibration maps, a factory setting and a user1 setting, could you theoretically make the factory setting setup for indoor shooting and user1 setup for colder outdoor shooting?
Click to expand...

In theory yes but ideally, having the ability to do an additional setting would be ideal.
Our goal is to make that viable given the current constraints as well as other options
for more flexibility.
 
Les Dittert said:
My guess is that the black shade may actually be trying to warm up the sensor to get more data points for hotter sensor conditions.
Otherwise, a black shade should take a couple of minutes max, even with a ton of frame averaging. I have done code to black shade sensors, I see no reason other than what I describe.
-Les Dittert
Click to expand...

The process is very compute intensive to get the best results.
 
Hey Deanan, I sent you an R3D trim. Thanks again..

I noticed that when turning on false color, I don't see a discrepancy in the evenness of the sensor noise until I go over 2000 ISO, that's when the false colors begin to split. Does anyone else notice this with their Epics? Could someone put the port cap on their Epic, turn the body on from a cold start, crank up to the MAX ISO and turn on False Color and tell me if your image looks even?

Thanks! Hopefully I solve this issue soon!
 
The epic we were using on this past show did the exact same thing. Same side of the frame, same little curve-smear. At first I thought it was a light flare, but with the port cap on and false color turned on it was pretty evident.

one black shade and it seemed to go away, its just unfortunate that the shading takes so long since there is nothing you can do till you get a 20 min window of time.
 
Jens Jakob Thorsen said:
Had this problem last week..didnt realize until grading...This is a HUGE problem!!! I dont think this particular producer will use the Epic again...
Click to expand...

I believe the failure was in your DIT, not noticing the issue. We did have a similar situation with another brand Digital Cinema Camera but the DIT and I (DP) noticed the issue on set. Not need to blame the camera, just find a new DIT.
 
Christian Munoz D said:
I believe the failure was in your DIT, not noticing the issue. We did have a similar situation with another brand Digital Cinema Camera but the DIT and I (DP) noticed the issue on set. Not need to blame the camera, just find a new DIT.
Click to expand...

Um, that's going too far the other way - the camera is hardly blameless :) It's obviously a combination of DIT and camera. You can't leave the camera out of it when there are a lot of other cameras that don't suffer from this problem / usability issue.

Do you want RED to make it a priority to improve this issue in their cams - or for them to think that the current setup is fine for most folks? Every company has a list of priorities, and good companies (such as RED) listen to their users when deciding on how to allocate resources to fix things. Surely the current black shading setup isn't ideal for you?

My personal opinion is that threads like these are very much in RED's best interest. Even if it comes off as whiny. Besides, RED seriously wants to have a boxing match with Canon (Jim's words, not mine) in the years to come, not only the spec punch has to be strong (it already is) - but usability footwork has to be super solid too. So solid that even if a RED cam is standing on some kind of rickety indie film boxing ring with a crap DIT, it'd still be the winner any day of the week. How about that for an analogy?

It takes a certain amount of courage to point out to a boxer where you think they should improve though! Especially if you want to stay their friend. I think/hope RED takes this as a useful expression of user priorities.

Though if all of these issues go away with the Dragon sensor, I say stop worrying about the MX and just bring on the Dragon ASAP :)

Bruce Allen
www.boacinema.com
 
@ Bruce.
I 100%agree with you. You just misinterpreted my post. I never said the camera is blameless.
 
You must log in or register to reply here.
Back
Top