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

Philip Orlandic said:
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)
Click to expand...

Why do you think fewer overexposed elements would have an effect on strobing?
 
High contrast and overexposure are not the same thing at all. And higher contrast is really a bit player in this phenomenon. I'll agree that a high-contrast scene will be more annoying, but this effect is really more about sharpness. Sharp verticals with well-exposed highlights between them will strobe just as much as sharp verticals with clipped highlights between.
 
It does indeed seem to be accentuated by contrast/thin lines/plentitude of lines.

Where it started screaming at me was from a cameratestshoot from a feature which is to be shot in an idyllic environment consisting of wooden houses (vertical lines) with white wooden fences (allmost overexposed on a sunny day), lamp-poles, cottage windows with white squares and trees.

Even at very slow pan-speeds it looked like a cluster-mess.

I cannot post these images as they are NDA for the production tests, but I will try to recreate.

One single high-contrast thin line is more obvios than low contrast thin lines.

It is also obvious on other cams, but it is not equal, just different bad.

Fot the procuction @ hand, this is what the script demands...

I'll follow up....
 
Gunleik Groven said:
It does indeed seem to be accentuated by contrast/thin lines/plentitude of lines.

One single high-contrast thin line is more obvios than low contrast thin lines.

I'll follow up....
Click to expand...

Yes, all of that seems right. I didn't mean to suggest that contrast had no part in this, only that the thin lines/plentitude of lines are the root cause. Contrast, as you say, makes it more obvious. On a recent shoot, I had some luck with increasing shutter angles, but found that the additional motion blur was also a problem on some shots where the actors were moving across the frame.
 
Gunleik Groven said:
It does indeed seem to be accentuated by contrast/thin lines/plentitude of lines.

Where it started screaming at me was from a cameratestshoot from a feature which is to be shot in an idyllic environment consisting of wooden houses (vertical lines) with white wooden fences (allmost overexposed on a sunny day), lamp-poles, cottage windows with white squares and trees.

Even at very slow pan-speeds it looked like a cluster-mess.

I cannot post these images as they are NDA for the production tests, but I will try to recreate.

One single high-contrast thin line is more obvios than low contrast thin lines.

It is also obvious on other cams, but it is not equal, just different bad.

Fot the procuction @ hand, this is what the script demands...

I'll follow up....
Click to expand...

Try going to a 270 shutter and see if that helps.
 
Hey Gunleik,

Obviously I hacen't seen your footage so I can only guess at what you are seeing. However, I would venture a guess that the reason that different camera's and exhibiting different results is probably down to the amount of detail, sharpness and other characteristics that are different between each camera and viewing pipeline that you've tested (In other words, not necessarily because RED is doing anything *wrong* - it could be more to do with the fact that RED is capturing more detail an this the effect is more pronounced?

I know you mentioned that you are seeing this in the viewfinder not just in post, but in my experience, it can definitely help to make sure the least amount of sharpening possible is being added in your processing chain for this. Sharper edges on verticals can really start to flicker and pop.

It may also be worth looking at your downscaling algorithm.. this will also have an effect.


Good luck, I hope you find a solution!
Seth.
 
I'll second that trying a 270 shutter is a good idea. You could also try adding a little motion blur in post to the problematic moves and see if that helps...

What length lenses were you shooting on?

Cheers,
Dom.
 
I have had a similar problem on a commercial shooting in studio - people against a white background, very white background almost blown out.The client is upset about the images they are seeing in the edit saying that they are strobing and they are.I suggested to the DP that he shoot at a higher frame rate when the strobing was noticed on the first day, but he declined.I was the DIT on that job.The client is not happy.I think it is the fact that the RED is shooting such sharp images against a high contrast background.I can't post any images because they don't belong to me.I wonder what film would have done in this situation, I get the feeling that it would be more blurred or less jarring but I will never know.Can you fix it in Post? Does anybody know? I have seen some amazing footage shot on 5D which had some work done in post and it looks like 1000 fps really amazing...this software created frames, does any body know what it is called?
The DP was a stills photographer and had never shot on RED before.
In December we will be doing the same thing again and I'm going to ask him to shoot the whole thing at 50 fps and this time insist.
 
If the DP was a stills photographer, it is possible that he went for a high shutter to get the image crisp (like you would maybe would like, if you shot a still...) That would accentuate the effect. Just a hunch here, but you could check the r3d's for shutterspeed to narrow it down...

I guess (GUESS) from what you say, that one could pull out a sort of matte on the edges and soften them a bit.

There is also software that emulates motionblur. There's even a function in FCP, but it ain't really topnotch... -:)

No... Just rambling here....
 
Gunleik, as I shoot outdoors all the time have run in same sort of issues. At some point picked from Steve Gibby the idea to use 1/60s with 25fps and 1/120sor 1/125s with 50fps and although this goes in the other direction what Deanan suggests, Steve's approach seems to work out.

While have tried to understand why this is the case, came to a preliminary conclusion that all this boils down to fact that the eye is very good in recognizing and catching anything which is regular in the spatial or temporal space. For example, when it comes to debayering and aliasing, believe Graeme has added some irregularity, such as a stochastical component, into the debayer algorithm to avoid aliasing.

Correspondingly, the 180 degree setting 25fps 1/50s possibly creates some temporal regularity when the camera is in motion and especially, if it shows up in the same place of the monitor, the eye tends to spot this immediately. Once a small "disturbance" is introduced by changing the shutter speed to 1/60s, 1/120s or 1/125s regularity is broken and the possible artefacts less distracting.

Still, not saying that understand this properly yet and thus interested in all thoughtful comments.
 
Bruce Newton said:
I wonder what film would have done in this situation, I get the feeling that it would be more blurred or less jarring but I will never know.Can you fix it in Post? Does anybody know?
Click to expand...

Blur comes from exposure time and wouldn't be more or less on film or other cameras. However, how you develop the image with respect to contrast and how you view it (contrasty lcd vs projection vs film out) do impact how strobing is percieved. The recommended rates are in the ASC manuals for film and they apply to digital cameras also.
 
Gunleik Groven said:
If the DP was a stills photographer, it is possible that he went for a high shutter to get the image crisp (like you would maybe would like, if you shot a still...) That would accentuate the effect. Just a hunch here, but you could check the r3d's for shutterspeed to narrow it down...

Ja we shot at 180 degrees, 1/50 th shutter speed.

I wonder if interlacing the images in post will help.
Our shot is a close up of a child moving around in front of a white background.In fact there are some shots where the camera is still and the kid is moving and she appears to strobe as she moves.You actually get the same effect if you look at your hand against a white background and without moving your eyes move your hand around.It is what it is as far as I'm concerned but the client is unhappy.We should have shot at 50 fps.
Click to expand...
 
I think 50fps only would help if you play it back with 25fps,
but then it would be slowmotion. Do you want that?
 
Na I think it will be good, 50 played back at 50 will work and will be real time.Twice the no. of frames vs half the amount of camera or artist movement will result in less strobing.
 
Gerhard Riesenhuber said:
I think 50fps only would help if you play it back with 25fps,
but then it would be slowmotion. Do you want that?
Click to expand...

You'd oretty much get the same effect shooting with a 1/100th shutter, if 25 is your playback format...

What deanan is suggesting is to make more motion-blur, which sounds smart.
 
(insert sound of a light bulb going on, in my head - ding ding ding) oh, i get it now.

this is a helpful thread. I recently shot some dolly shots of stationary (not being played, sitting on a tabletop) violin strings -- with a high contrast background -- and this effect was horrible! the image was jumping all over the place. strobe-y and awful. changing shutter speed seemed to fix it - but I really had to crank the shutter speed.

I figured this out on my own and am only now reading this thread, but it begs the question - where do I obtain these ASC guidelines for panning? everyone else seems to be in the know but me.

without reduser, I would never be able to have this kind of confirmation - it is a little frustrating that this type of information is not more reasonably accessible - might make a good THOUGHTS AND TIPS addition. or CRUCIAL ORDNANCE.

whatever happened to these efforts to widely disseminate important info? they seemed to have died on the vine.
 
@gunleik: i think it would not be the same as 25fps with 1/100 sec Shutter,
because the difference between each frame would be to a lesser extent.
But if you then increase the speed of movement of the actors it wouldn't
change the amount of strobing.

Filming with 50 fps and playing back with 50 fps would help, but it would look
more like video because the motion is more fluid and each frame is sharper
(due to 1/100 sec shutter).

More motionblur is indeed a solution :)

Gerhard
 
You must log in or register to reply here.
Back
Top