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 :)

Red One and Shooting Mesh Objects

Brook Willard said:
It's called Moiré.

All digital cameras have this issue, including 3CCD cameras. I've got the footage to prove it. :)

In front of every sensor is an optical low pass filter designed to combat moiré. While moiré is not a huge issue with the RED in general, RED is redesigning their OLPF [presumably for other reasons]. So the moiré may get better with the new filter... we'll find out soon enough.

In the mean time, you should either blur it slightly or just 86 the offending wardrobe. There's a reason you don't see many stripes in a wardrobe department.
Click to expand...

I Think there is one example here for miore
But is clear that is very seldom to happen .

Concrete said:
Code:
I've seen the tests, Gregorios. There is a significant rez improvement in the last circle. Sadly, some aliasing remains. It is interesting to see theat the camera blurs the right side of the circle. Does that have an explanation?

I've cropped the image to show that.

1434_1197533710.jpg
Click to expand...
 
Yes, you can provoke some chroma moire if you try hard enough. But we do have a pretty darn good OLPF in there that will avoid aliasing (which is what moire is) on anything "real world". If you do see anything on a real world shoot, however, I need to know, and would need a R3D from you to troubleshoot.

As for moire removal - once aliasing gets into a sampled system (and this is any camera, of any kind) it's very tough to remove without taking real detail with it. I'll watch the Hassleblad video and comment afterwards.

Graeme
 
I've asked someone on a photography forum to get me some images from Phocus, especially with regards to Moire. The problem is that moire is mathematically undetectable from real image data, and they were showing chroma moire removal and not luma moire removal :-)

Graeme
 
Ever heard the phrase "perfect is worse than good"? Sometimes in trying to kill an artifact completely, you lower the quality of the overall process. There are trade-offs that take place to get the right balance of image quality. Just like why you can't have a high-saturation film stock with wide exposure latitude.
 
Very interesting Graeme. But then again, if it works visibly, wouldn't that be good enough since all we really care about is how the issue is perceived by a viewer? Then again, I suppose this could still be an issue if its a motion picture since you might still see the pattern as a kind of texture in motion instead.

Any thoughts on that??
 
Yes, it works visibly and appears to do what it says. However, I'm an inquisitive guy and want to know exactly what it's up to. Only by learning about things can we know when they work and when they don't. If the tool only works for chroma moire, then that's an important consideration.

Graeme
 
You can deal with chroma moire separately from luma moire. Chroma moire happens twice as easily as luma moire or so, and is more obviously noticible, but luma moire can get bad, and is especially bad on moving images.

Graeme
 
Cardmaverick said:
Ah, I know what your talking about, thats actually Andromeda... shot by me!

http://www.chrisnuzzaco.com/special_uploads/pencilcomparison.mov
http://www.chrisnuzzaco.com/special_uploads/movingpencils1.mov

I think there is a sleight instance of it though for like a quick second, but yeah, not terrible looking.

Crewpix, good point, I agree about that. I can't imagine this problem coming up on film if it never steps its foot into a digital/electronic realm. Certainly an
important detail worth mentioning.
Click to expand...
No I wasn't thinking of that, what I was refering to was definitely shot with a very early red camera prototype.

I did a bit of searching and I believe it was part of the tests David Stump made.

At any rate you can download the high res shots from cinematography.net and see for yourself:

http://www.cinematography.net/Red/comp-matrix.html

Paul
 
Its interesting if its on chroma moire only. As we are less sensitive on chroma than luma, and moire reduction usually cuts high frequency and thus detail, it wouldn't be too bad if it'd only affect chroma!
 
Well, any sampled system can alias - just that film uses random sampling which turns artifacts into noise, making them hard to see :-)

Graeme
 
A sort of "spread spectrum" sample interval is a technique used in some instrumentation applications. Very interesting stuff.

Kevin
 
Had a fairly interesting pattern shooting a LCD screen yesterday (with Red camera).

Framed on the full laptop screen was fine

Morie became more apparent closer (about 50% closer), but by slightly defocussing the image the pattern could be almost eliminated, while keeping the image apparently sharp.

Even closer and it became a lot stronger.

ECU (frame was about 3-4 inches edge to edge), and you could actually pull focus through the screen, past the morie pattern and onto the image behind it, eliminating the pattern.

Lens was a Ultra 85mm with +1 & +2 diopters about 10 inches from the screen at the closest. We pulled some 4K stills from the footage and the effect was similar to what was being seen on the 720P monitor.
 
You must log in or register to reply here.
Back
Top