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 :)

Horrible magenta flare caused by highlights

Hmph. So when you ordered, you made no mention of getting the "old" OLPF?

The other thing I recall is that some OLPFs were put in backwards, though I don't know what this did to the image.
 
Guys, this is the new olpf and this issue has been well documented here... I'll try to find the other threads but there has been a lot of people with this exact issue on the new olpf. Jarred said it's a CMOS thing and all cameras do it to a degree. He said all they can do is try to neutralize it because they can't stop it from happening. It's not the lens or the mount.

Edit: Here's a link to the other discussion.
http://www.reduser.net/forum/showthread.php?118019-Noise-on-the-low-end-but-even-worse-scary-megenta-lines&highlight=streaks

It has this issue as well: http://www.reduser.net/forum/showthread.php?118848-Dragon-tests-Crazy-alien-flare-and-ghosting&highlight=streaks
 
Sorry Mike I ment to say the new OLPF that was shipped with my camera.

Thanks Matt thats all I was looking for. I found a few threads briefly talking about the issue but was looking for more info. I appreciate the lead. At first I thought it was a motion mount issue because I wasn't having the same problem with my regular EF mount.

Anyway, Much thanks guys.
 
This is a BIG ISSUE for the Dragon sensor. It's not an OLPF-related thing.

I know what Jarred has said. He is right - this is indeed a CMOS artifact.

However, it's much, much worse on the Dragon sensor, compared with other high-end digital cameras - including Red's own MX sensor.

I don't know why it's so bad on Dragon. It needs to be fixed because - as it stands now - you can't shoot against bright areas. Look at the first image in this thread - I see this all the time in similar shooting scenarios (which are not rare on any narrative professional shoot).

Like I said in another thread - take the Dragon in a dark room/environment, shoot a bright LED light and pan around it to see what happens. You will see an incredibly visible magenta flare dancing around. You don't see such results on the MX sensor, or the Alexa, so something is off with the Dragon sensor.

Can somebody post a test like this for anyone to see?
 
I haven't had my dragon for long now and this is the first time I have noticed the issue. I will be keeping an eye out for it now as it has made me a little concerned when put in future scenarios that are similar.
 
Last edited:
But you said it only happens with the motion mount? It certainly looks like an internal reflection of some kind, which is why I gravitated towards the OLPF (which has had different varying issues).

What strength of ND was the motion mount was set to? Colour not withstanding, it looks like a reflection of the bright areas off of something internal (presumably the motion mount plane, though I was thinking OLPF originally).
 
Paul J Steinberg said:
While it is a CMOS issue, it seems like every other camera I use has compensated for it to the point you can't see it without pixel-peeping to find the small RGB value changes.
Click to expand...



Did Red say they were going to do the same? Was there any details given on what their approach was going to be?
 
Mike P. said:
But you said it only happens with the motion mount? It certainly looks like an internal reflection of some kind, which is why I gravitated towards the OLPF (which has had different varying issues).

What strength of ND was the motion mount was set to? Colour not withstanding, it looks like a reflection of the bright areas off of something internal (presumably the motion mount plane, though I was thinking OLPF originally).
Click to expand...

It's not a reflection. It's the sensor getting overloaded with light, and having it affect the readout of other pixels.

How prevalent is this problem? All dragons?
 
Mike P. said:
But you said it only happens with the motion mount? It certainly looks like an internal reflection of some kind, which is why I gravitated towards the OLPF (which has had different varying issues).

What strength of ND was the motion mount was set to? Colour not withstanding, it looks like a reflection of the bright areas off of something internal (presumably the motion mount plane, though I was thinking OLPF originally).
Click to expand...

This is NOT a Motion Mount issue. This is a Dragon sensor issue that needs fixing.
As mentioned above by other posters, it's a CMOS issue that is not so apparent in other cameras.
 
Any chance you could post .r3d snapshots of those frames that we can examine?
 
This is all Dragon cameras. This is the only issue I feel is truly important that it be fixed. Shot an interior bar scene today, issue came up multiple times. It's making it's way around set so hopefully they can do something quick. All other threads related to the issue have been shut down, which concerns me.
 
It is definitely not a motion mount issue, I was just stating that I had noticed it while the motion mount was on in order to cover that question from being asked. After doing much research I can agree it is a sensor issue. One that I believe also happens on the MX. I am Just not sure if it is to the same extent. Here are the .r3d snapshots!

I have bumped the ISO a bit to make the issue more visible to examine.

https://www.dropbox.com/s/64xtrzpih0ptntv/A001_C002_07295S.0000000F.R3D
https://www.dropbox.com/s/ugv7og6u2fhgxrr/A004_C023_0729GE.0000769F.R3D
 
Hi Kris,

I've seen even more extreme examples with smaller sources of light.

However, looking at your first example, I'm sure even RED would agree that something needs to be done as far as this nagging issue is concerned.
That shot is not usable in a professional project. And again, this type of shot comes out a lot - it's not something which only happens once in a while and can be avoided.

I haven't noticed RED shutting down similar threads - or maybe I wasn't paying enough attention. But I'd be very surprised if they did that.

Knowing Jarred and RED as a company, I think they are looking for ways to fix this issue. It might be a CMOS artifact, but it's far more obvious on the Dragon.
 
[FONT=Tahoma, Calibri, Verdana, Geneva, sans-serif]Medavoym,[/FONT]

I couldn't agree more about it not being usable. I can't tell you for sure it won't be seeing my final edit, and that is really unfortunate. Something as simple as shooting in a dark space with bright background highlights shouldn't be something I need to avoid. I am going to do some tests to see if I can pinpoint exactly what is causing the problem. I hope it is something more complex than a composition of a silhouette.

Lets just hope there is a fix pending and that they have addressed that there is definitely a problem.
 
You must log in or register to reply here.
Back
Top