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 :)

Monitor image "fluttering"

A

Alexis Hanawalt

Well-known member
Joined
Mar 13, 2007
Messages
713
Reaction score
0
Points
0
Location
Los Angeles, CA
Website
www.bolstercamera.com
Having a bit of a problem with all monitor sources (the RED LCD and an external HD-SDI monitor) showing a sort of fluttering in lighter parts of the image - not sure if I'm describing this in a way that makes sense, but it's like little ripples in the image, rolling up... very faint. Not evident in the recorded footage, but it's been alarming the director and DP... anyone know what this is?
 
My LCD sometimes does some odd flickering/fluttering. It's hard to say if this is the same thing you see without a screenshot or something. I notice it often when I switch frame-rates or pan across a bright scene. It has never shown up in footage however. A re-boot or black calibration might help.
 
If you are talking about what i think you are then it's something that was introduced with the latest build. A friend of mine thinks that it starts after you zoom in on the LCD. Most of the times a reboot will fix it.
 
yup.. just a build bug, nothing wrong with your LCD.
 
Will there be another build 15 which addresses this bug?
Or do we have to wait for build 16?
 
This might be different - we're only shooting 2K, sounds like the problem people have described only crops up in 4K. I did briefly see some severe garbled image/flicker in the lower part of the screen a few days back while in 4K... this is different.

It's very very subtle, like someone's holding a magnet a foot away... noticeable only in bright areas of the whole monitor, just little ripples constantly pulsing. I'm guessing it is a build issue, but please, anyone who's seen this in 2K confirm this for me.
 
Ok, I was wrong - it IS in the footage.

It might be some sort of HMI flicker... if it crops up again today, I'll have lighting cycle through the HMI's to check. In the footage, it seems to only come from certain light sources - parts of the image in the foreground that have the same luma characteristics as the background may not flutter when the background does.

Either way, some of the footage is unusable. I'll investigate and report on any new findings.
 
Flicker?

Flicker?

Whoa, I thought it was a monitor only issue. Now it's in the footage? Can you upload some frame grabs please, Cinewalt?
Click to expand...

Peace. Yes there is a bug in the LCD driver that can show up under 4K (and only 4K) If there is an artifact in the footage you can bet there is a good reason for that ... lets wait for the report.
 
I have a sample of this issue in some footage that I've uploaded to my FTP. If anyone wants to see it to understand what I'm describing, please PM me and I will send the link - I can't publicly post the footage from this shoot.

It's totally in the footage. Strangely, we were as sure as could be that it was a strange Red-version of HMI flicker, which we thought we'd confirmed at the end of today's shooting when the problem kicked in again for the first time as the 12K HMI was fired up... but then when we had them turn it off, the "rolling pulses" remained in our footage.
 
Okay, I got permission to post this. Here is footage that shows the "rolling pulses" - as many including ourselves point out, it really appears to be caused by a light source. What, unfortunately, I can't show - is the footage where these "rolling pulses" appeared in our monitor without the suspect HMI powered up.

Anyway, here it is: http://www.rent-a-red.com/WavesSampleH264.mov
 
This is what happens when you use a magnetic-ballast HMI with a rolling-shutter CMOS camera. I tested and verified this with my Red One and a magnetic HMI.

Your only defense is to try to change the shutter speed to a multiple of 1/120 (usually, you want to try 1/40th or 1/60th). Or switch to an electronic-ballast HMI.
 
Anyone at Red care to comment?

Jarred, you were quick to point out a software bug which this clearly is not. It is a rolling shutter issue. Shooting flicker free with sound can be problematic as the ballasts sometimes tend to hiss in flicker free mode (the flicker free mode was designed for MOS shooting!)

Yes there are fixes: 1/40 or 1/60 shutter angle, flicker free settings on the ballasts, etc. but a big warning to all users in the manual is in order.

Basically: DO NOT SHOOT AT THE NORMAL 180 DEGREE SHUTTER WHEN SHOOTING WITH ANY 60HZ BASED ELECTRONIC LIGHT SOURCE AT 23.98 fps OTHER THAN AN ELECTRONIC HMI SET SPECIFICALLY AT FLICKER FREE!

This problem will also apply to HID's, Fluorescents, Sodium Vapors and about 80% of practical light sources encountered in the real world. I hope this can be fixed with a software update, otherwise we should resign ourselves to shoot at either a 1/60 or 1/40 because the problem can be subtle enough on a smaller monitor where it can easily be overlooked. Anyone tried to shoot a TV monitor (other than an LCD) yet? Please report...

Luckily I knew about the problem on my last shoot with a Red and had all the ballasts set to flicker free as a precaution. Thanks to Art Adams for publishing the clip of a white card lit with a fluorescent that exhibited the pulsing problem when the contrast was cranked up significantly.
 
I don't know why this thread is being ignored, but I think it is imperative to conduct and publish a test confirming HMI safe speeds for the Red.

We need a table like this:

flickerfree.jpg


Someone with an HMI and a Red camera at hand please do a thorough test.

I suggest aiming the HMI at a grey card, flag of any stray light and crank the contrast way up in Redcine.

Anyone?
 
I did, and that list doesn't work. There's no such thing as a flicker-free (or, flutter-free) speed at any angle. The shutter speed is vital. With what I found, you have to stick to the shutter speeds of 1/120, 2/120, 3/120, 4/120 and 5/120 to have any hope of immunity. Going to odd frame rates doesn't help, it complicates issues.

23.976 or 24.000 or 30.000 or 60.000 all work at 1/40 and 1/60. Stepping outside there caused issues. I don't remember where my notes are, I pretty much decided to forget magnetic-ballast HMIs after experimenting with this, but if I dig up the notes I can try to show what I found.
 
florianstadler.com said:
We need a table like this:
Click to expand...

That table is designed for a film-like shutter which distributes intensity variation in a different manner than a rolling-shutter based film camera. As an easy fix just substitute the words "frame rate" with "shutter speed" in the top table in the chart, so that it reads "shutter speed at *any* frame rate" and take the reciprocal. So starting from the lower end of the table the numbers (in seconds) become 1/120, 1/60, 1/40, ....
 
I understand the mathematical correlation between shutter angle, frame rate and shutter speed thank you. Let's just keep this discussion to shutter speed then for simplicity's sake.

Since 1/48 shutter speed with global shutter is in a HMI safe window and apparently not with the Red's rolling shutter I would like to encourage an investigation in what IS safe. is 1/120th safe? 1/180th? 1/90th? 1/600th? (etc.)
 
1/40th and 1/60th were safe at 23.976 and 24fps. I don't remember if they were safe at other speeds.
 
You must log in or register to reply here.
Back
Top