How robust is the signal of the XL2? at DVinfo.net

Ronald Lee · December 15th, 2004, 08:54 PM

Hi there

I am considering using the XL2 to shoot a dramatic TV series, it's my first choice btw this and the Sony Z1U. However, an HD expert here told me that the signal on the XL2 is easy to break i.e. if you shoot aimed at the end of a long runway, or if you shoot horizontal lines at an angle (his exact test was to shoot venetian blinds with the camera rotated to an angle).

Have you guys experienced anything like this, where you get the 'jaggies' or artifacting under less than idea shooting conditions or even non-conventional shooting?

Thanks.

Chris Hurd · December 15th, 2004, 09:22 PM

Ronald, its signal is as "robust" as any other standard definition DV camcorder. How would it be any different?

Ronald Lee · December 15th, 2004, 11:07 PM

What he meant was that if you tilter the camera, or shot straight lines at weird angles, or converging lines, that you can 'break' the signal, meaning get artifacts or jiggies or whatever.

I've seen similar stuff happen on my PDX10, I'm just hoping on he XL2 it's not a major problem, but rather a negligiable one.

A. J. deLange · December 16th, 2004, 08:07 AM

Yes, you will get "jaggies" with the XL2 or any other video camera unless the manufacturer has incorporated a blur (antializasing) filter which reduces the vertical resolution sufficiently. As "sufficiently" would result in an appreciable reduction of vertical resolution and as the situations where jaggies are annoying are relatively infrequent most manufacturers do not incorporate a strictly band limiting filter though good cameras apparently do have some. I do not know whether Canon has an optical antialiasing filter nor, if they do, where it cuts but it is apparent it does not attenutae high frequencies enough to completely suppress aliasing from near horizontal lines.

At http://homepage.mac.com/ajdel/FileSharing5.html you will find a file "XL2 Samples" which is DV straight from the camera. In about the middle of the file there are images of some Zapotec pyramids the one of which closest to the viewer shows aliasing between courses of stone at the edge of the lowest platform. It's quite obvious and you should have no trouble spotting it if you are looking for it. At nominal viewing distance I doubt you would even notice it but you will have to judge for yourself whether it is major or negligable. Be sure to look at the clip on a studio monitor or, if on a computer, that the pixel size of the display is 720 x 480 or some integer multiple of this unless you are confident that you graphics card does interpolation properly. This is to insure that you see the aliasing of the XL2 - not the display system!

Kevin Kocak · December 16th, 2004, 08:59 AM

I have had extreme difficulty shooting outside. The camera seems to have a hard time with grass, leaves, trees, etc. While it is true that most cameras do have these problems they seem to be more pronounced with the XL2. However I have found some workarounds for these issues although it is a pain.

Ronald Lee · December 16th, 2004, 07:23 PM

<<<-- Originally posted by Kevin Kocak : I have had extreme difficulty shooting outside. The camera seems to have a hard time with grass, leaves, trees, etc. While it is true that most cameras do have these problems they seem to be more pronounced with the XL2. However I have found some workarounds for these issues although it is a pain. -->>>

Do you think this camera could be controlled so that the above problems don't happen, with careful shooting? Is it safe for broadcast and professional use (meaning these problems can be controlled?)?

Lauri Kettunen · December 17th, 2004, 12:33 PM

<<<-- Originally posted by Kevin Kocak : I have had extreme difficulty shooting outside. -->>>

Kevin, could you be more specific. I shoot all the time outside and have not observed any problems. The only problem which comes into mind is that the viewfinder is not able to show "grass, leaves, trees" etc. properly, but the signal recorded on tape is just perfect. Could it be that you talk of what you see on the viewfinder?

Barry Green · December 17th, 2004, 01:12 PM

There are limits to the abilities of DV compression, well-documented, especially as related to high-frequency detail (such as grass and leaves). There's not much the camera can do about it, as it's the compression that's unable to cope, not the camera itself.

That said, many people don't notice a problem. Those that do will see blocking and mosquito noise artifacts, things like that around areas of very high-frequency detail.

Kevin Kocak · December 17th, 2004, 01:34 PM

I use a NTSC 8" and you can see artifacting on it. I have done some experimenting and found it can be reduced but not eliminated. Don't get me wrong, I love my XL2. There are going to be different problems with different cameras. You have to know the limitations of the medium you are working with and how to make it work for you... That is what makes a good shooter.
To answer your question Ronald NG... You can definitely achieve broadcast quality. You can't possibly expect a camera to produce the perfect image for you all the time. You have to use the camera and its settings to capture the image you want. Also just watch tv for a while, specifically reality tv and you can always find a moire or artifacting.

Chris Hurd · December 17th, 2004, 01:37 PM

"Jaggies" and stair-stepped lines are more of a limitation of the DV format than the camcorder.

Andre De Clercq · December 17th, 2004, 01:48 PM

Indeed DCT based compression like DV has its limits w.r.t artifacts (not in subjective resolution), and most people don't "see" them. What is recognized more often is what A.J.deLange pointed out : optical aliasing on the CCD level. Not only in the vertical direction but also horizontally. These problems are often masked by the CCD motion blurr on sharp moving edges. On the contrary these effects become worse with high shutterspeeds and high detail settings. "Jaggies" have a simular origine but are seen in non moving pics. In near horizontal structures there is additional interline flicker produced with interlace pics. Consumer cams sum up (weighted) individual pics (and lines) for their ''optical'' filtering, pro cams use real (birefringence based)sharp optical cut-off filters.(Li-niobate)

Lauri Kettunen · December 18th, 2004, 04:32 AM

<<<-- Originally posted by Andre De Clercq : ... w.r.t artifacts (not in subjective resolution), and most people don't "see" them. -->>>

as well, many like me may consider the artifacts as a fact of life.

Besides, the question is not only of the DV format. For example, when they broadcast the formula one races in live, especially from the helicopter view, the sides of the track (the border between the black track and the red-white side region) are quite badly stair-stepped. With a fairly limited number of horizontal lines such things just appear.

December 15th, 2004, 08:54 PM	#1
Ronald Lee Major Player Join Date: Jan 2004 Location: Vancouver, Canada Posts: 435	How robust is the signal of the XL2? Hi there I am considering using the XL2 to shoot a dramatic TV series, it's my first choice btw this and the Sony Z1U. However, an HD expert here told me that the signal on the XL2 is easy to break i.e. if you shoot aimed at the end of a long runway, or if you shoot horizontal lines at an angle (his exact test was to shoot venetian blinds with the camera rotated to an angle). Have you guys experienced anything like this, where you get the 'jaggies' or artifacting under less than idea shooting conditions or even non-conventional shooting? Thanks.

December 15th, 2004, 09:22 PM	#2
Chris Hurd Obstreperous Rex Join Date: Jan 2001 Location: San Marcos, TX Posts: 27,368 Images: 513	Ronald, its signal is as "robust" as any other standard definition DV camcorder. How would it be any different? __________________ CH Search DV Info Net \| 20 years of DVi \| ...Tuesday is Soylent Green Day!

December 17th, 2004, 01:37 PM	#10
Chris Hurd Obstreperous Rex Join Date: Jan 2001 Location: San Marcos, TX Posts: 27,368 Images: 513	"Jaggies" and stair-stepped lines are more of a limitation of the DV format than the camcorder. __________________ CH Search DV Info Net \| 20 years of DVi \| ...Tuesday is Soylent Green Day!

December 15th, 2004, 11:07 PM	#3
Ronald Lee Major Player Join Date: Jan 2004 Location: Vancouver, Canada Posts: 435	What he meant was that if you tilter the camera, or shot straight lines at weird angles, or converging lines, that you can 'break' the signal, meaning get artifacts or jiggies or whatever. I've seen similar stuff happen on my PDX10, I'm just hoping on he XL2 it's not a major problem, but rather a negligiable one.

December 16th, 2004, 08:07 AM	#4
A. J. deLange Major Player Join Date: Jun 2004 Location: McLean, VA United States Posts: 749	Yes, you will get "jaggies" with the XL2 or any other video camera unless the manufacturer has incorporated a blur (antializasing) filter which reduces the vertical resolution sufficiently. As "sufficiently" would result in an appreciable reduction of vertical resolution and as the situations where jaggies are annoying are relatively infrequent most manufacturers do not incorporate a strictly band limiting filter though good cameras apparently do have some. I do not know whether Canon has an optical antialiasing filter nor, if they do, where it cuts but it is apparent it does not attenutae high frequencies enough to completely suppress aliasing from near horizontal lines. At http://homepage.mac.com/ajdel/FileSharing5.html you will find a file "XL2 Samples" which is DV straight from the camera. In about the middle of the file there are images of some Zapotec pyramids the one of which closest to the viewer shows aliasing between courses of stone at the edge of the lowest platform. It's quite obvious and you should have no trouble spotting it if you are looking for it. At nominal viewing distance I doubt you would even notice it but you will have to judge for yourself whether it is major or negligable. Be sure to look at the clip on a studio monitor or, if on a computer, that the pixel size of the display is 720 x 480 or some integer multiple of this unless you are confident that you graphics card does interpolation properly. This is to insure that you see the aliasing of the XL2 - not the display system!

December 16th, 2004, 08:59 AM	#5
Kevin Kocak Regular Crew Join Date: Dec 2004 Location: Tampa, Fl Posts: 149	I have had extreme difficulty shooting outside. The camera seems to have a hard time with grass, leaves, trees, etc. While it is true that most cameras do have these problems they seem to be more pronounced with the XL2. However I have found some workarounds for these issues although it is a pain.

December 16th, 2004, 07:23 PM	#6
Ronald Lee Major Player Join Date: Jan 2004 Location: Vancouver, Canada Posts: 435	<<<-- Originally posted by Kevin Kocak : I have had extreme difficulty shooting outside. The camera seems to have a hard time with grass, leaves, trees, etc. While it is true that most cameras do have these problems they seem to be more pronounced with the XL2. However I have found some workarounds for these issues although it is a pain. -->>> Do you think this camera could be controlled so that the above problems don't happen, with careful shooting? Is it safe for broadcast and professional use (meaning these problems can be controlled?)?

December 17th, 2004, 12:33 PM	#7
Lauri Kettunen Major Player Join Date: Aug 2004 Location: Kangasala, Finland Posts: 445	<<<-- Originally posted by Kevin Kocak : I have had extreme difficulty shooting outside. -->>> Kevin, could you be more specific. I shoot all the time outside and have not observed any problems. The only problem which comes into mind is that the viewfinder is not able to show "grass, leaves, trees" etc. properly, but the signal recorded on tape is just perfect. Could it be that you talk of what you see on the viewfinder?

December 17th, 2004, 01:12 PM	#8
Barry Green Barry Wan Kenobi Join Date: Jul 2003 Location: North Carolina Posts: 3,863	There are limits to the abilities of DV compression, well-documented, especially as related to high-frequency detail (such as grass and leaves). There's not much the camera can do about it, as it's the compression that's unable to cope, not the camera itself. That said, many people don't notice a problem. Those that do will see blocking and mosquito noise artifacts, things like that around areas of very high-frequency detail.

December 17th, 2004, 01:34 PM	#9
Kevin Kocak Regular Crew Join Date: Dec 2004 Location: Tampa, Fl Posts: 149	I use a NTSC 8" and you can see artifacting on it. I have done some experimenting and found it can be reduced but not eliminated. Don't get me wrong, I love my XL2. There are going to be different problems with different cameras. You have to know the limitations of the medium you are working with and how to make it work for you... That is what makes a good shooter. To answer your question Ronald NG... You can definitely achieve broadcast quality. You can't possibly expect a camera to produce the perfect image for you all the time. You have to use the camera and its settings to capture the image you want. Also just watch tv for a while, specifically reality tv and you can always find a moire or artifacting.

December 17th, 2004, 01:48 PM	#11
Andre De Clercq Major Player Join Date: Sep 2002 Location: Belgium Posts: 804	Indeed DCT based compression like DV has its limits w.r.t artifacts (not in subjective resolution), and most people don't "see" them. What is recognized more often is what A.J.deLange pointed out : optical aliasing on the CCD level. Not only in the vertical direction but also horizontally. These problems are often masked by the CCD motion blurr on sharp moving edges. On the contrary these effects become worse with high shutterspeeds and high detail settings. "Jaggies" have a simular origine but are seen in non moving pics. In near horizontal structures there is additional interline flicker produced with interlace pics. Consumer cams sum up (weighted) individual pics (and lines) for their ''optical'' filtering, pro cams use real (birefringence based)sharp optical cut-off filters.(Li-niobate)

December 18th, 2004, 04:32 AM	#12
Lauri Kettunen Major Player Join Date: Aug 2004 Location: Kangasala, Finland Posts: 445	<<<-- Originally posted by Andre De Clercq : ... w.r.t artifacts (not in subjective resolution), and most people don't "see" them. -->>> as well, many like me may consider the artifacts as a fact of life. Besides, the question is not only of the DV format. For example, when they broadcast the formula one races in live, especially from the helicopter view, the sides of the track (the border between the black track and the red-white side region) are quite badly stair-stepped. With a fairly limited number of horizontal lines such things just appear.