XL1s vs XL2 Resolution comparison at DVinfo.net

A. J. deLange · January 8th, 2005, 12:31 PM

I finally have the information I have been looking for since the announcement of the XL2 - comparative resolution data for the XL1s and XL2 and the story is quite interesting (at least to me). At

http://www.pbase.com/agamid/image/38413496
http://www.pbase.com/agamid/image/38413504
http://www.pbase.com/agamid/image/38413534
http://www.pbase.com/agamid/image/38413625

I've posted 4 frame grabs of a Putora chart with the cameras set for 4:3 in both 30p and 60i modes. I used several lenses in getting the data and concluded that the CCD's/DV are limiting i.e. none of the lenses (16x manual, 20x stock, 17-35 Zoom Nikkor, 35mm Nikkor prime) was noticeably sharper than any of the others and so I theorized that the prime was actually the sharpest even if I couldn't readily detect that and so used that for the four frames I posted. The frames were grabbed using BTV Pro and no processing was done so the pixels are 720 x 480 which isn't 4:3 and the circles don't look quite round. As usual I caution people to think about what they are doing before resampling and note that I don't think the site I posted to has done any resampling because the images look the same from the site as they do in BTV.

Rather than give my own conclusions I think I'll just let you all have a look and draw your own! For those unfamiliar the circles in the Putora chart correspond to, respectively, 10, 13.3,18,23.5,31.3 and 42 lines per mm ON A 2/3" TARGET (so they are approximately double that on the XL cameras but remember that the active areas aren't the same size on the XL1s and XL2). The centre of the circle array contains another circle with parts at 55.5, 73.6 and 97 and you will note that some aliased response is seen in some of the frames in the parts of this circle that correspond to the lowest spatial frequency (55.5).

The horizontal bar across the top is in terms of lines (I assume per picture width but I don't know for sure). The block labeled 1 has 80, 2 ~ 160; 3~240; 4 ~ 320; 5 ~ 400 and 6 ~ 480.

Hope you find these as intersting as I did.

Jacob Ehrichs · January 8th, 2005, 09:53 PM

Very cool. You can definitely tell the difference from the XL1 to the XL2 and a bit more from the 2 XL2 shots. I like the XL2 30p shot best.

A. J. deLange · January 15th, 2005, 07:34 AM

I've added another image to the set. This is for the Canon 3X zoom. It's at http://www.pbase.com/image/38413625

Josh Dahlberg · January 16th, 2005, 10:32 PM

Thanks for the superb images...

Interesting to see very little discernable difference between the 3x and the 20x. The 20x was noticeably softer at f11 - it would be interesting to see if the same loss of sharpness occurs with the iris wide open.

I noticed on another thread that you confirm the 3x has focusing issues, as others have mentioned. Provided that you set the lens at a fixed focal length and focus there (ie: rather than going full telephoto to focus before pulling back), is it possible to achieve a sharp focus at any focal length? In your experience, how much can you open up the lens (in terms of f-stop) without it going all mushy?

Great work, love to see more!

Richard Hunter · January 17th, 2005, 01:24 AM

<<<-- Originally posted by Josh Dahlberg : is it possible to achieve a sharp focus at any focal length? -->>>

Hi Josh. If you have a decent-sized monitor you should be able to set the focus OK. If you have to use the XL2 viewfinder, you will be struggling. That's why people normally try to zoom in to set the focus, and then frame the shot after that. If the focus drifts off during the framing, it's not very useful, especially if you don't find out about it until after the shoot.

Richard

Chris Mills · January 17th, 2005, 01:29 AM

Nice example images. Thank you for sharing those.

Josh Dahlberg · January 17th, 2005, 02:38 AM

Thanks Richard,

That's exactly what I wanted to know. So if one shoots in a controlled/staged environment (dramatic work) as I do, and does not alter the focal length during a take, it is possible to work around the problem. I hope I understand that correctly.

But it's still a pain in the neck... esp. as people seem to be having problems with the 0.6x and 0.7x adaptors as well

A. J. deLange · January 17th, 2005, 07:13 AM

Yes, if you focus each time you set the focal length you can get a sharp picture but, as has been noted by others, this is difficult with the viewfinder and is certainly a PITA. I guess the fact that this lens has a reputation for being soft speaks to the extent of this problem. So the bottom line is focus every shot (or let auto do this) and don't zoom within a shot or between shots unless you refocus (but then they say that zooms mark the amateur).

I haven't checked the 3X against aperture.

January 8th, 2005, 12:31 PM	#1
A. J. deLange Major Player Join Date: Jun 2004 Location: McLean, VA United States Posts: 749	XL1s vs XL2 Resolution comparison I finally have the information I have been looking for since the announcement of the XL2 - comparative resolution data for the XL1s and XL2 and the story is quite interesting (at least to me). At http://www.pbase.com/agamid/image/38413496 http://www.pbase.com/agamid/image/38413504 http://www.pbase.com/agamid/image/38413534 http://www.pbase.com/agamid/image/38413625 I've posted 4 frame grabs of a Putora chart with the cameras set for 4:3 in both 30p and 60i modes. I used several lenses in getting the data and concluded that the CCD's/DV are limiting i.e. none of the lenses (16x manual, 20x stock, 17-35 Zoom Nikkor, 35mm Nikkor prime) was noticeably sharper than any of the others and so I theorized that the prime was actually the sharpest even if I couldn't readily detect that and so used that for the four frames I posted. The frames were grabbed using BTV Pro and no processing was done so the pixels are 720 x 480 which isn't 4:3 and the circles don't look quite round. As usual I caution people to think about what they are doing before resampling and note that I don't think the site I posted to has done any resampling because the images look the same from the site as they do in BTV. Rather than give my own conclusions I think I'll just let you all have a look and draw your own! For those unfamiliar the circles in the Putora chart correspond to, respectively, 10, 13.3,18,23.5,31.3 and 42 lines per mm ON A 2/3" TARGET (so they are approximately double that on the XL cameras but remember that the active areas aren't the same size on the XL1s and XL2). The centre of the circle array contains another circle with parts at 55.5, 73.6 and 97 and you will note that some aliased response is seen in some of the frames in the parts of this circle that correspond to the lowest spatial frequency (55.5). The horizontal bar across the top is in terms of lines (I assume per picture width but I don't know for sure). The block labeled 1 has 80, 2 ~ 160; 3~240; 4 ~ 320; 5 ~ 400 and 6 ~ 480. Hope you find these as intersting as I did.

January 17th, 2005, 01:29 AM	#6
Chris Mills Regular Crew Join Date: Oct 2004 Location: Wellington, New Zealand Posts: 77	Nice example images. Thank you for sharing those. __________________ Regards, Chris http://www.lizardlounge.com

January 8th, 2005, 09:53 PM	#2
Jacob Ehrichs Regular Crew Join Date: Mar 2004 Location: Dayton, OH Posts: 169	Very cool. You can definitely tell the difference from the XL1 to the XL2 and a bit more from the 2 XL2 shots. I like the XL2 30p shot best.

January 15th, 2005, 07:34 AM	#3
A. J. deLange Major Player Join Date: Jun 2004 Location: McLean, VA United States Posts: 749	I've added another image to the set. This is for the Canon 3X zoom. It's at http://www.pbase.com/image/38413625

January 16th, 2005, 10:32 PM	#4
Josh Dahlberg Major Player Join Date: Jan 2005 Location: Wellington, New Zealand Posts: 513	Thanks for the superb images... Interesting to see very little discernable difference between the 3x and the 20x. The 20x was noticeably softer at f11 - it would be interesting to see if the same loss of sharpness occurs with the iris wide open. I noticed on another thread that you confirm the 3x has focusing issues, as others have mentioned. Provided that you set the lens at a fixed focal length and focus there (ie: rather than going full telephoto to focus before pulling back), is it possible to achieve a sharp focus at any focal length? In your experience, how much can you open up the lens (in terms of f-stop) without it going all mushy? Great work, love to see more!

January 17th, 2005, 01:24 AM	#5
Richard Hunter Trustee Join Date: Dec 2004 Location: Lipa City Batangas, Philippines Posts: 1,110	<<<-- Originally posted by Josh Dahlberg : is it possible to achieve a sharp focus at any focal length? -->>> Hi Josh. If you have a decent-sized monitor you should be able to set the focus OK. If you have to use the XL2 viewfinder, you will be struggling. That's why people normally try to zoom in to set the focus, and then frame the shot after that. If the focus drifts off during the framing, it's not very useful, especially if you don't find out about it until after the shoot. Richard

January 17th, 2005, 02:38 AM	#7
Josh Dahlberg Major Player Join Date: Jan 2005 Location: Wellington, New Zealand Posts: 513	Thanks Richard, That's exactly what I wanted to know. So if one shoots in a controlled/staged environment (dramatic work) as I do, and does not alter the focal length during a take, it is possible to work around the problem. I hope I understand that correctly. But it's still a pain in the neck... esp. as people seem to be having problems with the 0.6x and 0.7x adaptors as well

January 17th, 2005, 07:13 AM	#8
A. J. deLange Major Player Join Date: Jun 2004 Location: McLean, VA United States Posts: 749	Yes, if you focus each time you set the focal length you can get a sharp picture but, as has been noted by others, this is difficult with the viewfinder and is certainly a PITA. I guess the fact that this lens has a reputation for being soft speaks to the extent of this problem. So the bottom line is focus every shot (or let auto do this) and don't zoom within a shot or between shots unless you refocus (but then they say that zooms mark the amateur). I haven't checked the 3X against aperture.