@JackBayer Care to post some sample footage? I'd be interested to examine it.
I have this strange feeling that somehow the different patches may affect the timing of the cam taking the original frames. I might be wrong, but jitter in this place would explain alot. Is this possible?
@jackbauer interesting. In your experience, which hack provides less studdering?
HI all,
as I sadly just had to realize the studdering varies alot between hacks. The latest Flowmotion is horrible to my eyes for example. Is there some way to post-fix it or bring it down at least?
Interesting, but I don’t see much difference in their test footage (aside from the rotating disk example). The judder is still there.
Also, it turns out the Arri Alexa Studio also uses a rotating shutter: http://arri.com/camera/digital_cameras/learn/alexa_faq.html
A rotary shutter is not eliminating the effect, it even induces it on analog film too, albeit to a smaller degree. It needs time to rotate, so it will open one side of the frame first, then gradually open the rest, finally gradually closing again. Main difference to electronics: the edge of the shutter is out of focus and smeared by motion, that makes a difference. There is a company trying to mimic this for digital cameras:
www.tessive.com
@Blackout1 I’m going to have to disagree with you. I don’t see how fading helps with smooth motion. As far as the film frame is concerned, you’re just exposing it for 1/48th of a second, the entire area of the frame. Digital sensor readout (CMOS), on the other hand is not instantaneous and even the fastest sensors are a far cry from being “on and off in a perfect fashion.”
I think the excessive amount to judder in digital cinema has more to do with its availability to the general public. The sheer number of amateur videos shot at a “magic” rate of 24 fps outweighs the number of films shot professionally. Hence all the judder/noise/which-setting-is-better discussions.
By the way, a few months ago I did some cursory reading on the subject of judder, and was surprised to see the same discussions on the RED forums.
You’re right about the panning speed rule being very old. It predates digital cinema by a few decades. The reason it became a rule (well, a guideline, rather) was because following it created smoother motion.
Someone posted a short clip recently (with one of the newer Driftwood settings, I believe) demonstrating a perfectly smooth pan. No judder whatsoever, an extremely filmic look.
Here’s another thought to ponder: given the size of Arri and RED cameras, it would be very easy to implement a small rotary shutter in them. Heck, you could even have it at the same 45º angle as in film cameras! Then you could synchronize sensor readout with the shutter’s rotation and presumably get the same look. And yet neither Arri nor RED has implemented such a feature. Why haven’t they done that?
EDIT: Well, I’ll be! It turns out that some manufacturers of digital cinema cameras do use a mechanical rotary shutter. The Sony F65 has one, for example. The purpose is to eliminate (or at least to alleviate) the rolling shutter effect.
There is something in works for this, a kind of LCD shutter.
@Blackout1, that's interesting info and makes a lot of sense. I think that a digital senor could simulate this by ramping the sensor gain at the start and end of each frame capture. I don't know if current sensors can adjust their sensitivity that quickly, but if not it will probably be possible in the future.
It is and has always been known that "digital" 24p will always appear to strobe and stutter because of the process in which it is captured. Regardless of the camera. Film shutter gates open gradually creating a slight fading for each frame. It fades as it opens and fades out as it closes. Digital cams are on and off in a perfect fashion. It is a rule of thumb not to have anything pass through the frame faster than 5 seconds to avoid this. This is a very old rule.
It looks like you're new here. If you want to get involved, click one of these buttons!