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 :)

MX problems with vertical lines/pans

Gunleik Groven

Gunleik Groven

REDuser Sponsor
Joined
Dec 28, 2006
Messages
13,465
Reaction score
5
Points
38
Location
Norway
Website
quine.no
Have been looking at some MX footage from different cams on 30.5.0. Vertical lines stutter a lot in playback, on the SDI monitor, the onboard monitor and on dcp projections in theatre.

We have tried to pan ultra slow, but with lots of lines and deep focus (think wooden houses, gardens with fences + trees), and we see the same thing.

at first we thought it might be due to cameramotions. Now we don't.

Anyone recognize this?

Cheers

Gunleik
 
Yes, is at its worst with bright vertical objects on a dark background. Seems to do this at a wide range of pan speeds. Looks horrible.
 
Have you tried the 30.6.0 build? It fixes a few issues regarding noise / image quality. I would start there. Then post some samples. I'm not completely sure what you're talking about without seeing it and I haven't seen it myself. Or maybe I have seen it and I'm thinking of something else..
 
Panning across vertical lines.

Panning across vertical lines.

This has been discussed before and it's nothing to do with the type of sensor or version of firmware.

Nor is it a RED specific issue, rather it's a high contrast / panning / 24 fps issue.

But please post a clip ...
 
Will a faster sensor reset time better this issue?
Is it connected to rolling shutter? (looks like it on more cams than the MX... -:)
 
Hm, that was nice and informative. Thanks.

My guess it is it is related, but it is not the vertical going diagonal thing I am thinking about, it's more like long vertical lines seemingly not following a pan, but jumping a bit as you pan. If you look at individual frames, you see only normal motion-blur, but in motion it iss less than fluent.

The worst example I have seen is not uploadable for a number of reasons, but I can post a short example. Have to I guess...

BTW, I understand this is not an MX issue, so if a mod needs to change the threadtitle, please do...
 
Gunleik Groven said:
Will a faster sensor reset time better this issue?
Is it connected to rolling shutter? (looks like it on more cams than the MX... -:)
Click to expand...

Nope. It's 24fps stutter and you see it on film, ccd, cmos, rolling shutter, global shutter, spinning mirror, etc.

Contrast between the background and moving object is what makes it stand out more or less. The only small difference between film and digital is the sensitivity response curve makes the blur slightly different.

That is if you're talking about the same stutter we've discussed here before.
 
Deanan, would a 360 deg shutter fix that? I know what the OP is talking about, and it looks terrible (on all cameras).
 
I remember seeing this on Genesis footage that was being projected.

It panned passed some boat masts and they started jumping back and forth like crazy.

The room contained some world class DOPs and operators.

I asked if it bothered anyone. They all thought it looked great. I couldn't believe we were talking about the same footage.
 
Erich Ocean said:
Deanan, would a 360 deg shutter fix that? I know what the OP is talking about, and it looks terrible (on all cameras).
Click to expand...

Following the ASC guidlines for panning speeds is the first place to start.
In leiu of that, wider shutter angles can help if you can break
180deg convention.

It's basically all about cadence, blur and contrast.
 
I would second Deanan. The effect is a byproduct of high contrast, cadence, motion blur and viewing fps regardless of acquisition format.

It can be reduced on RED by treating your debayer-sharpening accordingly btw (obviously not eliminated, but less post-sharpened images will exhibit less strobing)

Is also a problem in 25fps PAL-land as well.
 
You cannot pan slow enough for this to go away...

I am aware of the panning guidelines.

It is different on differnet cameras.

It doesn't really matter, but we shoot a lot of 25 @ 180 here, and (of course) the issues are similar...
 
Stick to the rules gentlemens. There are no other options for now... :smash:


24FPS with 18-20mm lens should take 9 seconds to pan 90° sweep.

24FPS with 25-28mm lens should take 15 seconds to pan 90° sweep.

24FPS with 50mm lens should take 23 seconds to pan 90° sweep.

24FPS with 100mm lens should take 60 seconds to pan 90° sweep.

and etc...


This could be extenuate with HDRx... Maybe.


Is there any big difference between 24fps and 30fps ?
 
Philip Orlandic said:
Stick to the rules gentlemens. There are no other options for now... :smash:


24FPS with 18-20mm lens should take 9 seconds to pan 90° sweep.

24FPS with 25-28mm lens should take 15 seconds to pan 90° sweep.

24FPS with 50mm lens should take 23 seconds to pan 90° sweep.

24FPS with 100mm lens should take 60 seconds to pan 90° sweep.

and etc...


This could be extenuate with HDRx... Maybe.


Is there any big difference between 24fps and 30fps ?
Click to expand...

I can't see that HDR would have any effect on this phenomenon at all. Why would it? It's a strobing artifact caused by the movement of the subject relative to the sensor between frames. It's the temporal equivalent of moire -- the same effect that causes wagon wheels in old westerns to turn backward from time to time. It's the same principal that makes kinescopes work. As Deanan says above, it is a product of the combination of 24fps and a 180-degree shutter, and it's present in all formats that shoot that way: film, digital, CMOS, CCD, whatever.

Higher contrast exacerbates it a bit, but any sharp image of vertical lines will be at risk. The panning guidelines will sometimes help, but if the verticals are close together, only blurring them will really have a significant effect (and that may not be possible for a given shot). Increasing the shutter angle can help (because that will increase motion blur). And a 360 degree shutter will pretty surely eliminate the effect, but at the cost of far more motion blur than you're likely to be happy with.

Shooting 30i or 60i or 60p will also lessen the effect, because motion resolution is increased (there is less time between recorded images) but at the cost of giving up the more "cinematic" look of 24fps.
 
D Fuller said:
I can't see that HDR would have any effect on this phenomenon at all. Why would it? It's a strobing artifact caused by the movement of the subject relative to the sensor between frames. It's the temporal equivalent of moire -- the same effect that causes wagon wheels in old westerns to turn backward from time to time. It's the same principal that makes kinescopes work. As Deanan says above, it is a product of the combination of 24fps and a 180-degree shutter, and it's present in all formats that shoot that way: film, digital, CMOS, CCD, whatever.
Click to expand...

You will probably have less strobing artifacts because you won't have overexposed elements on the picture.

It should be a little better in some situations then a shooting with the standard dynamic range (10-12)
 
You must log in or register to reply here.
Back
Top