Justin Jenkins
November 18th, 2010, 07:57 AM
I was hoping some of the technical guys here could confirm or deny the calculations I did to determine the actual size of the image sensor in the xh a1. I'm not reffering to the the 1/3 inch CCD meaurement in this case. I'm reffering to the actual usuable image sensor size that determines the camera's angle of view and perspective.
I did some trigonometry based on the 35mm equivalent field of view listed in the manuel. In the manuel it says '16:9 still images' but since that is the same aspect ratio as video I'm hoping that measurement applies to the video as well. Assuming that it does and the engineers at canon were accurate and the field of view of a 32.5mm lens on a 35 mm camera is the same as the field of view on the xh a1 at a focal length of 4.5mm, I came up with the following image sensor size of the xh a1:
4.98mm x 2.8 mm
If anyone can confirm or deny the accuracy of this measurement it would be great. I'm doing some matchmoving with footage shot using the xh a1 so this measurement is pretty important to me. No one on the Canon customer service line could give me this measurement. They said the only people with this info are the engineers in Japan.
Thanks so much!
-Justin
Les Wilson
November 18th, 2010, 12:52 PM
Here's an informative article albeit not one that answers your question other than to say that the sensor is not actually a rectangle that measures 1/3". There's some links in the article.
ProVideo Coalition.com: SONY Livewire | Vendor Chanel (http://provideocoalition.com/index.php/sony/story/sony_publishes_guide_to_sensor_sizes/)
Justin Jenkins
November 18th, 2010, 01:36 PM
Thanks a lot for that link. I'll definitely read up on that and check it out. I'm definitely aware that the 1/3" 'measurement' isn't actually the size of the image sensor the camera actually uses to create the final image. Hence the need to do a bunch of trigonometry and annoying stuff to try and figure it out.
I've been impressed with how knowledgeable and technical the community is on this particular forum. You guys know way more about the how these cameras work than myself. I was hoping one of you had figured this out already or could help verify or deny the validity of the result of the calculations I did to figure it out.
Thanks again!
-Justin
Don Palomaki
November 19th, 2010, 05:08 AM
You may recall that ~7x factor in equivalent field of view/focal length when mounting a 35mm film lens on an XL1 series camcorder. That implies the active sensor dimensions are about 1/7 the size of a 35mm still frame. This is confirmed by the Sony table at the above link.
Justin Jenkins
November 19th, 2010, 07:45 AM
That implies the active sensor dimensions are about 1/7 the size of a 35mm still frame. This is confirmed by the Sony table at the above link.
Thanks so much for that information! How accurate of a measurement would you say that is? So if I punch that into a calculator and come up with an exact size in millimeters, what would you say the margin of error would be? Is it so small as to not really make a difference, or is there a possibility it could be off a decent amount to the actual size of the sensor in my particular camera.
I came up with a different measurement for the sensor based on another set of caluculations. This set of calculations is much less accurate than my first, however, because it involves me actually mearsuing things in the real world with a tape measure and unfortunately my measurements probably aren't accurate enough do to a number of reasons. What I did was set my lens at a 4.5 mm focal length and aim the camera level at a wall. I measured the disance from thte nodal point of the camera to the wall, and then marked two lines on the wall that corresponded to the horizontal edges of the frame from the camera's view point. I used the LCD to judge this.
I then did this:
image area / distance from camera to wall
atan (result of previous calculation)
convert result from radians to degrees
**EDIT**
Apologies to all the trig gurus out there. I typed this equation out in a rush this morning and forgot to add that I divide the image area by 2 and multiply the final angle by 2. This is to make all the right triangle math work correctly.
**
All of this works mathmetically to give me my angle in question, which is the angle of view of the camera. But I doubt my ability to measure things accurately enough with a tape measure from Lowes. :) If this were accurate I could then determine the sensor size using the focal length of the lens and the field of view of the camera.
thanks again for all the help!
-Justin
Justin Jenkins
November 19th, 2010, 02:54 PM
I thought I would outline the calculations I made and the measurements I took to determine the image sensor size based on my particular camera in case anyone can spot any errors in what I've done.
I placed a chart on the wall so I could center the camera on the chart. I measured the distance from the floor to the center of the chart. I put the xh a1 on a tripod and set the camera to the exact same height as the center of the chart.
After that I centered the camera on the chart. To do this I got my friend to place a mark on the wall at the very edge of the frame. I looked through the LCD and used the level display to make sure that the mark was placed on the same line as the level display to be as accurate as possible. It was neccesary to eyeball the marks so there is some room for error here but we tried to be as exact as possible. I measured the distance from the left mark to the center of the chart and made sure that the measurement from the center of the chart to the mark on the right side was within 1/16th of an inch of the previous measurment. This ensures a centered camera. After that I used the center mark display to make sure the camera was pointed directly at the center of the chart vertically. After all of this the camera is as centered on the wall and chart as we could possibly make it with the tools we had at hand. Finally, we measured the distance between the mark on the left side of the wall to the mark on the right side and recorded that as the horizontal image plane.
From here we measured the distance from the wall to the nodal point of the camera. We recorded this distance as the focal distance.
Final measurements:
Image plane: 8 feet 4 5/16 inches (2.5479 m)
Focal distance: 7 feet 3/4 inches (2.15265 m)
Focal length: 4.5 mm
Calculations to determine Angle of View:
2.5479 / 2 = 1.2739
atan (1.2739 / 2.15265) = 30.61764
30.61764 * 2 = 61.235294
Angle of View @ 4.5mm focal length = 61.235294
I used this angle of view along with the focal length of the lens to arrive at these values for the image sensor size:
5.32 mm x 2.996 mm
It's slightly different than my previous calculations based on the Canon 35 mm equivalent focal lenght but it's still kind of close...kind of. :)
Let me know what you think or if you spot any errors! (Hopefully I haven't made any typos).
Thanks,
Justin
Don Palomaki
November 20th, 2010, 07:01 AM
The Canon documentation for the EF Adapter XL lists 7.2x as the effective focal length multiplier. Keep in mind that it was based on a SD image format (4x3) and used the image diagonal as the field of view.
I expect the figures in the Sony table are about right.
Justin Jenkins
November 24th, 2010, 09:22 AM
Man, that's so cool that Sony measurement and the measurement I came up with using trig are extremely close. I meant to post about this earlier but forgot. That leads me to believe that this is, in fact, the correct image sensor size for this camera within a very, very small margin of error.
Sony measurement for 1/3 inch sensor size 16:9 digital motion picture camera:
5.23 x 2.94
My size for the canon xh a1 sensor based on trig and some measurements taken with a tape measure:
5.32 mm x 2.996 mm
If you would have told me 5 years ago math could actually be useful to my life I would have laughed. I guess the math teachers from high school get the last laugh now. :)
-Justin