Welcome to our community

Be a part of something great, join today!

  • 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 :)

CMOS - set in stone?

There are tradeoffs in going with a rolling shutter.

There are tradeoffs in going with a global shutter.

I think the rolling shutter in a Red One is probably the most acceptable place for it, if you're going to have a rolling shutter (and, in general, I hate the rolling shutter). At least with a Red one, you're not out shooting under uncontrolled circumstances, you can plan and account for rolling shutter circumstances.

I think the rolling shutter in an ENG/"video" camera is the worst place for it, precisely because you're going to be in uncontrolled circumstances.

I am certain Scarlet will be a rolling shutter because they couldn't do the frame rates, nor get the needed sensitivity, nor meet the price point, with a global shutter.

As technology improves, I'd love to see the rolling shutter disappear and be replaced with a global shutter so that these artifacts/issues go away. Until then, for today, there are things that the rolling shutter enables that just couldn't be done otherwise. And there are tradeoffs that come with that (look at the HMI thread for an example). If you don't educate yourself as to what the rolling shutter does and how it can affect your footage, you're asking for a screwed-up shoot sooner or later. If you educate yourself, you'll be forearmed and able to prevent problems before they occur.
 
I am certain Scarlet will be a rolling shutter because they couldn't do the frame rates, nor get the needed sensitivity, nor meet the price point, with a global shutter.
More than likely you're right, but the scheduled release date is about a year away. Red has been fairly reticent about discussing the Mysterium X sensor. The "40% improvement" statement (mentioned elsewhere in there forums) regarding rolling-shutter artifacts is an odd comment in some ways -- how do you quantify the reduction of such visual artifacts so precisely? Or does it refer to the data readout rate (which is easier to measure)?

What I'm trying to say is that Red is being tight-lipped about the new sensor, possibly because it may be a global-shutter design or, more likely, a rolling-shutter design whose artifacts they are feverishly trying to reduce to an imperceptible, trivial level. They know that Scarlet is a big gamble: a design that looks like a near-miracle on paper, but if it's not adequately implemented it could become a marketing fiasco. You'd think that releasing a more expensive camera (Red One) with glitches would have been a problem for them, but no, they were anticipated and the target market (film professionals) also knew to expect, work around, and patiently allow Red to resolve them. A large part of the potential market for Scarlet is a little different: people who might not be so understanding, and this would only be worsened by selling Scarlet through the likes of Best Buy and Circuit City.

Red knows that at $3,000 Scarlet is a steal; at the same time, they know that the sensor has to produce far less image wobble than the HV20/30. And it undoubtedly will. I believe that by this time next year, we'll consider the issue largely moot.
 
The 40% number was referring to the Epic, I believe, since it will have a maximum full sensor frame rate of 100fps vs. 60fps on the Red One. A 40% faster full frame read out. Scarlet is a 2/3" chip with a 180fps maximum in burst mode. So if the chip is reading out the data as fast as it's able regardless of frame rate, then that's a 50% improvement over the 120fps maximum of the Red one in 2K mode. These are just guesses based on my limited understanding of all of this.

Also, the artifacts that are being cited by the original post don't really have anything to do with the rolling shutter. The half frame strobes, which can be seen in that footage, but weren't mentioned, and skew on panning or objects moving quickly in the frame, which I didn't notice much of in that footage are caused by a rolling CMOS shutter. The blurring and overexposure mentioned as the undesirable elements can happen with any camera CCD or CMOS. Blurring is caused by longer shutter speeds and the overexposure is because those strobes are really bright. That footage looks bad mostly because it wasn't shot well, not because of the camera. Just my observation and opinion.
 
The 40% number was referring to the Epic, I believe, since it will have a maximum full sensor frame rate of 100fps vs. 60fps on the Red One. A 40% faster full frame read out. Scarlet is a 2/3" chip with a 180fps maximum in burst mode. So if the chip is reading out the data as fast as it's able regardless of frame rate, then that's a 50% improvement over the 120fps maximum of the Red one in 2K mode. These are just guesses based on my limited understanding of all of this.

Also, the artifacts that are being cited by the original post don't really have anything to do with the rolling shutter. The half frame strobes, which can be seen in that footage, but weren't mentioned, and skew on panning or objects moving quickly in the frame, which I didn't notice much of in that footage are caused by a rolling CMOS shutter. The blurring and overexposure mentioned as the undesirable elements can happen with any camera CCD or CMOS. Blurring is caused by longer shutter speeds and the overexposure is because those strobes are really bright. That footage looks bad mostly because it wasn't shot well, not because of the camera. Just my observation and opinion.

You do understand it more than You think...

The rolling-shutter issue relates directly to the "speed" of the sensor readout combined with the shutter speed (which are "welded" in their limitation). And as correctly pointed out - Scarlet is capable of up to 180 fps (in this case limited by speed of the storage device - hence it is only available as "burst"). That could result in up to 66% improvement of the "skew" effect.

In reality however You would have to pan way to fast to make it visible and if You do - Your footage is going to be so blurred that You will not see it anyway...

Here comes to play the expertise of "traditional" cameramen. Fast pans can be very disturbing - one should "zoom out" follow the subject and then gently "zoom back in" - which results in much more "viewable" footage. I personally dislike the "shaky-cam" style of late. Good solid shots with reasonable amount of camera movement is more like what we do with our heads. Just look around Your place where You are right now and notice how fast are You moving Your head (unless You are watching F1 live and up close, or a ping-pong game that is :biggrin: )...
 
Panning is only one of many circumstances where the rolling shutter causes issues with the footage. Flashes, strobes, magnetic-ballast HMI and fluorescent light sources, and vibration are all affected by a rolling shutter. The panning thing is (subjectively) one of the least objectionable of the rolling shutter artifacts, there's lots more to consider.
 
Panning is only one of many circumstances where the rolling shutter causes issues with the footage. Flashes, strobes, magnetic-ballast HMI and fluorescent light sources, and vibration are all affected by a rolling shutter. The panning thing is (subjectively) one of the least objectionable of the rolling shutter artifacts, there's lots more to consider.

Analog film had (has) its own restrictions too and people learned to work around them.

"Flashes, strobes, magnetic-ballast HMI and fluorescent light sources" related issues could be addressed with the new LED technology.

http://www.reduser.net/forum/showthread.php?t=6879

Vibration should be avoided at all costs and when needed for "artistic expression" it can be added in post...

http://www.reduser.net/forum/showthread.php?t=12581

What ever the problem - a smart filmmaker will find a way...

I believe that rolling shutter is (will be) least of our problems - especially once the new Mysterium-X sensors are out...
 
Analog film had (has) its own restrictions too and people learned to work around them.
That's the point -- people need to LEARN about them, and actively WORK AROUND them. Ignoring these challenges will just lead to ruined shoots. There are real issues, and they need to be actively worked around.
 
Back
Top