Snapper123
Well-known member
Hey guys, an interesting topic I brought up once is that it seems like it would be a efficient thing to do if we set up an official distance chart based on the distances from subjects - such as brick walls, cloth furniture, stripe clothing, patterns, ect ect that you can be before moire happens on our DSLR's.
For instance, I just did a test using my 50mm and the cloth spine of one of my Harry Potter books, which is woven using tiny threads (like a miniature patterned couch) and the results were absolutely awful until I positioned the camera about 2 + 1/2 feet from the book, and then the video turned out absolutely lovely. Even beyond the complete absence of moire, the video quality and sharpness improved triplefold (where there's visible moire, there's invisible moire causing low quality looks all over.) I'm sure we've all done tests like this, but it's all the more reason to band together and create a chart generally based on the most common scenarios. The downside to moire that many people just don't recognize is again, moire hurts a video's sharpness and quality extremely wether you can see the moire going on or not, and it's not until the video's subject is large enough to override the moire that our videos will look sharp all over, not just the benefit of not having rainbows in our videos.
Meaning, if you were paid to film the spine of a Harry Potter book, you couldn't film it from more than 2 feet away with a 50mm lens. And no matter what lens you use, it could be a 75mm, or a 35mm, you have to make sure that the cloth pattern, each individual thread and string, is larger than what the moire lines would be. Making it even tougher, the LCD on the DSLR's don't give us any clue when this breaking point happens - yes, the LCD will start to moire somewhere down the line, but when you get close enough for it to stop on the LCD, you can garuntee you have to get at least a foot or too closer still. But this process probably variates so much depending on the focal length and such, but if website's like the-digital-photograph.com or whatever exist, I don't see why we can't create a chart equally as diverse!
But all this is just another reason to break down and invest in a moire filter, they also seem to work wonders and generally creates less worry - although, purchasing one of these seems complex as well, as it seems different strengths only work with different focal lengths and such, about as worrysome as a Panasonic hack!
Here's two videos (watch full screen and 1080P for best viewing + I've reviewed the Youtube uploads and they truly represent the original videos as they were recorded, it didn't seem that Youtube made the flaws better or worse):
-http://www.youtube.com/watch?v=XRQEteKlqkg
That video was filmed at about 4 feet away, roughly. Minimal aliasing.
-http://www.youtube.com/watch?v=KLeRfRfiPeQ
That one was filmed about 2 & 1/2 feet away, roughly, with absolutely no aliasing.
The negative to all of this is the limited size of the focal length and distance from the subject. As seen, with a 50mm, the camera has to be close enough to the book that effectively, I would have barely been able to fit the entire book's spine in the video - that's almost exactly as wide as someone's head! Imagine if there was a law that said you can't film a subject that is wider than a foot and a half, bigger than the human skull! But that's exactly what's going on with these cameras, they cannot record detail that is smaller than the aliasing lines, or else, aliasing happens, which also equals bad video quality, rainbow effects or not. When put this way, the full fram of the MKII becomes a bit appealing!
And don't count on the image alternatively improving the farther you stand back, believing that you're knocking the aliasing out with a far away distance, the effect there is that no matter how sharp you can get the surroundings to look in the LCD, everything will just blur like a bad moire filter or 640P camcorder. Also throwing detail slightly out of focus doesn't help, it just makes things look ugly and low quality as opposed to sharp and rainbowed, as seen here:
http://www.youtube.com/watch?v=7qXUfQ1R4Go
I say the distance chart is probably near as follows:
18mm = 1 foot.
35mm = 2 feet.
50mm = 3 feet.
75mm = 4 feet
100mm = 5 feet.
Overall, it's all exactly relative, say you're filming the sides of a book as I did, you have to make sure that book stays the same size in the frame, just, different lenses give you the availibility to stand farther back if there's a technical reason, or if you need to change the perception. But if you want a high quality, no moire, sharp video, that video will have to capture a subject no wider than 2 feet, most likely. Tough doo-doo.
P.S. Since most 1080P camcorders effectively have absolutely no moire whatsoever, that literally means you can film a subject that's about 20 feet across with virtually no quality loss or aliasing whatsoever, doesn't that sound appealing vs. being only able to film subjects that are 2 feet across? Quite literally, to have a sharp, high-def video on a DSLR, a hole in a woven cloth has to be bigger than a pixel on the video. That's really sucky stuff.
For instance, I just did a test using my 50mm and the cloth spine of one of my Harry Potter books, which is woven using tiny threads (like a miniature patterned couch) and the results were absolutely awful until I positioned the camera about 2 + 1/2 feet from the book, and then the video turned out absolutely lovely. Even beyond the complete absence of moire, the video quality and sharpness improved triplefold (where there's visible moire, there's invisible moire causing low quality looks all over.) I'm sure we've all done tests like this, but it's all the more reason to band together and create a chart generally based on the most common scenarios. The downside to moire that many people just don't recognize is again, moire hurts a video's sharpness and quality extremely wether you can see the moire going on or not, and it's not until the video's subject is large enough to override the moire that our videos will look sharp all over, not just the benefit of not having rainbows in our videos.
Meaning, if you were paid to film the spine of a Harry Potter book, you couldn't film it from more than 2 feet away with a 50mm lens. And no matter what lens you use, it could be a 75mm, or a 35mm, you have to make sure that the cloth pattern, each individual thread and string, is larger than what the moire lines would be. Making it even tougher, the LCD on the DSLR's don't give us any clue when this breaking point happens - yes, the LCD will start to moire somewhere down the line, but when you get close enough for it to stop on the LCD, you can garuntee you have to get at least a foot or too closer still. But this process probably variates so much depending on the focal length and such, but if website's like the-digital-photograph.com or whatever exist, I don't see why we can't create a chart equally as diverse!
But all this is just another reason to break down and invest in a moire filter, they also seem to work wonders and generally creates less worry - although, purchasing one of these seems complex as well, as it seems different strengths only work with different focal lengths and such, about as worrysome as a Panasonic hack!
Here's two videos (watch full screen and 1080P for best viewing + I've reviewed the Youtube uploads and they truly represent the original videos as they were recorded, it didn't seem that Youtube made the flaws better or worse):
-http://www.youtube.com/watch?v=XRQEteKlqkg
That video was filmed at about 4 feet away, roughly. Minimal aliasing.
-http://www.youtube.com/watch?v=KLeRfRfiPeQ
That one was filmed about 2 & 1/2 feet away, roughly, with absolutely no aliasing.
The negative to all of this is the limited size of the focal length and distance from the subject. As seen, with a 50mm, the camera has to be close enough to the book that effectively, I would have barely been able to fit the entire book's spine in the video - that's almost exactly as wide as someone's head! Imagine if there was a law that said you can't film a subject that is wider than a foot and a half, bigger than the human skull! But that's exactly what's going on with these cameras, they cannot record detail that is smaller than the aliasing lines, or else, aliasing happens, which also equals bad video quality, rainbow effects or not. When put this way, the full fram of the MKII becomes a bit appealing!
And don't count on the image alternatively improving the farther you stand back, believing that you're knocking the aliasing out with a far away distance, the effect there is that no matter how sharp you can get the surroundings to look in the LCD, everything will just blur like a bad moire filter or 640P camcorder. Also throwing detail slightly out of focus doesn't help, it just makes things look ugly and low quality as opposed to sharp and rainbowed, as seen here:
http://www.youtube.com/watch?v=7qXUfQ1R4Go
I say the distance chart is probably near as follows:
18mm = 1 foot.
35mm = 2 feet.
50mm = 3 feet.
75mm = 4 feet
100mm = 5 feet.
Overall, it's all exactly relative, say you're filming the sides of a book as I did, you have to make sure that book stays the same size in the frame, just, different lenses give you the availibility to stand farther back if there's a technical reason, or if you need to change the perception. But if you want a high quality, no moire, sharp video, that video will have to capture a subject no wider than 2 feet, most likely. Tough doo-doo.
P.S. Since most 1080P camcorders effectively have absolutely no moire whatsoever, that literally means you can film a subject that's about 20 feet across with virtually no quality loss or aliasing whatsoever, doesn't that sound appealing vs. being only able to film subjects that are 2 feet across? Quite literally, to have a sharp, high-def video on a DSLR, a hole in a woven cloth has to be bigger than a pixel on the video. That's really sucky stuff.
Last edited: