Are these as good as I can expect?

DanP

CR Pro
Dec 8, 2014
126
391
I tried taking some BIF pictures with a 5DIV, 600mm V2 & 1.4 Ext V3 on a tripod mounted gimbal. I used AI Servo AF with Zone AF (9 point with center) with Case 2 (I tried other cases, but 2 seemed the best). I found the AF had trouble locking on to the birds, particularly when they were low with a background of long grass or trees. It worked better when the background was sky, but it did still lose focus fairly easily and I'm not convinced the AF could keep up with the movement. I think it was locked on for the pictures I have attached, but it may not have been at the instance of the exposure. I estimate the owls were about 150 meters away. The pictures have all been cropped to 2400px X 3600px @ 300dpi, with file sizes ranging from 1.25MB to 1.81MB, but I don't know if this site changes that on the upload. The first picture was taken at f5.6, 1/1600 sec & ISO 250 with +.67 EC. The last 2 were at f8, 1/2000 sec & ISO 640 with +.67EC. None of them look particularly sharp to me. What I'd like to know from others that use this combination of equipment is whether it is possible to get significantly better sharpness or is this about the best I can expect given the equipment and distance involved. Perhaps there just aren't enough pixels on the owls. From the top of the heads to the end of the claws is about 470 pixels.054A2554 Adj Crop 8X12.jpg054A2685 Adj Crop 8X12.jpg054A2736 Adj Crop 8X12.jpg
 
Jan 29, 2011
10,675
6,121
Don’t forget case modes are a starting point for AF settings not the be all and end all of the system. Different people in the same situation will have different optimal settings as their ability to track and acquire can help or hinder the camera.

Take a look at this series of videos on the 5D IV AF by a regular poster here Grant Atkinson, he really knows his stuff and after watching his series I forgot the case modes completely and have the AF metrics on a custom menu tab. But watch the videos, he explains it all in great detail.


I’m sure people like AlanF, who is very familiar with cropping for birds, can give you a detailed breakdown of the expected sharpness from that number of pixels on a bird, along with the best software to use in post processing to get the very finest detail out of your files.
 
Last edited:
Upvote 0

AlanF

Desperately seeking birds
CR Pro
Aug 16, 2012
12,348
22,524
I tried taking some BIF pictures with a 5DIV, 600mm V2 & 1.4 Ext V3 on a tripod mounted gimbal. I used AI Servo AF with Zone AF (9 point with center) with Case 2 (I tried other cases, but 2 seemed the best). I found the AF had trouble locking on to the birds, particularly when they were low with a background of long grass or trees. It worked better when the background was sky, but it did still lose focus fairly easily and I'm not convinced the AF could keep up with the movement. I think it was locked on for the pictures I have attached, but it may not have been at the instance of the exposure. I estimate the owls were about 150 meters away. The pictures have all been cropped to 2400px X 3600px @ 300dpi, with file sizes ranging from 1.25MB to 1.81MB, but I don't know if this site changes that on the upload. The first picture was taken at f5.6, 1/1600 sec & ISO 250 with +.67 EC. The last 2 were at f8, 1/2000 sec & ISO 640 with +.67EC. None of them look particularly sharp to me. What I'd like to know from others that use this combination of equipment is whether it is possible to get significantly better sharpness or is this about the best I can expect given the equipment and distance involved. Perhaps there just aren't enough pixels on the owls. From the top of the heads to the end of the claws is about 470 pixels.View attachment 194372View attachment 194373View attachment 194374
First, do you get sharp shots of perched birds at that distance with your set up?
 
Upvote 0

DanP

CR Pro
Dec 8, 2014
126
391
Don’t forget case modes are a starting point for AF settings not the be all and end all of the system. Different people in the same situation will have different optimal settings as their ability to track and acquire can help or hinder the camera.

Take a look at this series of videos on the 5D IV AF by a regular poster here Grant Atkinson, he really knows his stuff and after watching his series I forgot the case modes completely and have the AF metrics on a custom menu tab. But watch the videos, he explains it all in great detail.


I’m sure people like Alan F, who I is very familiar with cropping for birds, can give you a detailed breakdown of the expected sharpness from that number of pixels on a bird Along with the best software to use in post processing to get the very finest detail out of your files.
Thanks PBD. I watched the videos and will try experimenting with my own parameters. I was aware of what Grant covered, but it was a good refresher.
 
Upvote 0

Aussie shooter

https://brettguyphotography.picfair.com/
Dec 6, 2016
1,183
1,817
brettguyphotography.picfair.com
At 150m. Yes. That is the best you can expect. THB I wouldnt even pull the trigger at 150m. I would rarely pull the trigger at more than 50m with a bird. And then only for a 'record' shot. Even with a 600 you want to be as close as you can possibly get. For a small bird I would want to be within 15-20m. For a large bird, within 20-40m.
 
  • Like
Reactions: 1 users
Upvote 0

docsmith

CR Pro
Sep 17, 2010
1,223
1,109
From the top of the heads to the end of the claws is about 470 pixels.

For fine detail, I tend to use a minimum of 150 pixels per inch (ppi) to resolve a subject, 300 ppi is better, some go as low as 75 ppi but that is about as low of resolution as I tend to go. Given that your owl is likely taller than 3 inches, I am thinking is this a classic resolution issue. I've created a spreadsheet for this, which if correct: 840 mm on the 5DIV, you are resolving a subject at 75 ppi/150 ppi/300 ppi at 52 m/26m/13 m distance, respectively. So, Aussie shooter is resolving the small birds with 150-300 ppi and larger birds with ~100 ppi.

At 150 m, if my math is correct, you are about 26 ppi.
 
Last edited:
  • Like
Reactions: 1 user
Upvote 0

StoicalEtcher

CR Pro
Jan 3, 2018
417
360
Yorkshire
At 150m. Yes. That is the best you can expect. THB I wouldnt even pull the trigger at 150m. I would rarely pull the trigger at more than 50m with a bird. And then only for a 'record' shot. Even with a 600 you want to be as close as you can possibly get. For a small bird I would want to be within 15-20m. For a large bird, within 20-40m.
And on top of everything else, you also need to factor in atmospheric aberration, the impact of which increases with distance and magnification.

Nice looking encounter anyway though DanP
 
  • Like
Reactions: 1 users
Upvote 0

DanP

CR Pro
Dec 8, 2014
126
391
First, do you get sharp shots of perched birds at that distance with your set up?
I didn't take any pictures of perched birds at that distance as they would be too small to use, and I can generally get closer than that before spooking them. However, that's a good thought and next time I'll try to find a similarly distanced bird to photograph and compare. With the flying ones that are fighting, you have to take what you can get.
 
Upvote 0

DanP

CR Pro
Dec 8, 2014
126
391
At 150m. Yes. That is the best you can expect. THB I wouldnt even pull the trigger at 150m. I would rarely pull the trigger at more than 50m with a bird. And then only for a 'record' shot. Even with a 600 you want to be as close as you can possibly get. For a small bird I would want to be within 15-20m. For a large bird, within 20-40m.
I agree that 150 m is probably too far for good results, but they wouldn't cooperate and fight closer to me, so better to get something than nothing. For a single bird at that distance, I wouldn't bother taking a shot.
 
  • Like
Reactions: 1 user
Upvote 0

DanP

CR Pro
Dec 8, 2014
126
391
For fine detail, I tend to use a minimum of 150 pixels per inch (ppi) to resolve a subject, 300 ppi is better, some go as low as 75 ppi but that is about as low of resolution as I tend to go. Given that your owl is likely taller than 3 inches, I am thinking is this a classic resolution issue. I've created a spreadsheet for this, which if correct: 840 mm on the 5DIV, you are resolving a subject at 75 ppi/150 ppi/300 ppi at 52 m/26m/13 m distance, respectively. So, Aussie shooter is resolving the small birds with 150-300 ppi and larger birds with ~100 ppi.

At 150 m, if my math is correct, you are about 26 ppi.
I didn't think to look at it in those terms. i went back and looked at the original frame. I originally calculated the estimated distance of 150m based on the assumption that the body of the upright owl in the middle picture was 40cm (within the range of sizes for short-eared owls) as it made the math easier given the body of the owl (without including the extended legs) is about 400px. I then used the angle of view of the 600 & 1.4 ext. to calculate the distance using trigonometry. Extending the math, 40cm (15.75") at 400 px is 25.4 ppi, so pretty much bang on with your calculation.
 
  • Like
Reactions: 1 users
Upvote 0

docsmith

CR Pro
Sep 17, 2010
1,223
1,109
I didn't think to look at it in those terms. i went back and looked at the original frame. I originally calculated the estimated distance of 150m based on the assumption that the body of the upright owl in the middle picture was 40cm (within the range of sizes for short-eared owls) as it made the math easier given the body of the owl (without including the extended legs) is about 400px. I then used the angle of view of the 600 & 1.4 ext. to calculate the distance using trigonometry. Extending the math, 40cm (15.75") at 400 px is 25.4 ppi, so pretty much bang on with your calculation.
Ha! Love it when the math and assumptions work.

Call it pixels on target, resolving power or what have you, it is necessary to get good images.
 
  • Like
Reactions: 1 user
Upvote 0

Aussie shooter

https://brettguyphotography.picfair.com/
Dec 6, 2016
1,183
1,817
brettguyphotography.picfair.com
I agree that 150 m is probably too far for good results, but they wouldn't cooperate and fight closer to me, so better to get something than nothing. For a single bird at that distance, I wouldn't bother taking a shot.
Fair enough. As I said i will often take 'record' shots of something even if I know the IQ will be crap. For example I have a polar bear at well over a 1000m because it was the first one i ever saw. TBH though I reckon that those shots are pretty damn good at that distance. Quite the advertisement for the camera really.
 
  • Like
Reactions: 1 users
Upvote 0

AlanF

Desperately seeking birds
CR Pro
Aug 16, 2012
12,348
22,524
I didn't think to look at it in those terms. i went back and looked at the original frame. I originally calculated the estimated distance of 150m based on the assumption that the body of the upright owl in the middle picture was 40cm (within the range of sizes for short-eared owls) as it made the math easier given the body of the owl (without including the extended legs) is about 400px. I then used the angle of view of the 600 & 1.4 ext. to calculate the distance using trigonometry. Extending the math, 40cm (15.75") at 400 px is 25.4 ppi, so pretty much bang on with your calculation.
I get a longer distance using the simple equation (object size)/(image size) = (object distance)/(image distance), without invoking trig. 400px on the sensor is a length of 1.75mm for a sensor of pixel pitch 4.38 microns. The image distance is close to 0.84m for a distant object and a 600mm lens with a 1.4xTC. So, the distance to the 400mm long owl is: (400mm/1.75mm)*0.84m. That is, 192m. (The actual image distance is 0.844m and distance to owl 193m if you don't use the approximation image distance ~ f). Either I have done something silly or you have used the wrong angle for the lens?
 
Upvote 0

DanP

CR Pro
Dec 8, 2014
126
391
I get a longer distance using the simple equation (object size)/(image size) = (object distance)/(image distance), without invoking trig. 400px on the sensor is a length of 1.75mm for a sensor of pixel pitch 4.38 microns. The image distance is close to 0.84m for a distant object and a 600mm lens with a 1.4xTC. So, the distance to the 400mm long owl is: (400mm/1.75mm)*0.84m. That is, 192m. (The actual image distance is 0.844m and distance to owl 193m if you don't use the approximation image distance ~ f). Either I have done something silly or you have used the wrong angle for the lens?
Your method certainly seems simpler than mine. I was not aware of this method. I used an angle of view of 1.63714 degrees based on the formula that angle of view (in radians) equals sensor height / focal length (24/840). Converting radians to degrees (@57.3) gives the 1.63714. Since then I have found my manual for the lens and it states the angle of view for the 600 + 1.4 ext is 1degree, 40 minutes, which equals 1.66667 degrees, so close enough (Canon has probably rounded the minutes). My detailed trig calculation gave a distance of 156.7m, which i sloppily rounded down to 150m as an approximation for purposes of my post. If I use your calculation, but use 5.357 effective microns (24mm/4480px), then I get 156.8m. I'm guessing the pixel pitch you quoted is for the actual photosite, and doesn't take into account gaps between the sites. I'm curious about the .84 factor as I'm not familiar with it. Where does it come from?
 
Upvote 0
Mar 25, 2011
16,848
1,835
There are two factors that might affect the sharpness at this distance.

The first, of course is the number of pixels on the subject. Its pretty low in those images.

The second is the air itself. At nearly 200m, there is substantial distortion caused by the air and it alone can keep you from seeing fine details. So, even with a really long focal length, you still have to deal with the air. Cold days seem to be better than warm to reduce the impact of air currents.

I think its the sum of these two issues that makes the image less sharp.
 
  • Like
Reactions: 1 users
Upvote 0

AlanF

Desperately seeking birds
CR Pro
Aug 16, 2012
12,348
22,524
Your method certainly seems simpler than mine. I was not aware of this method. I used an angle of view of 1.63714 degrees based on the formula that angle of view (in radians) equals sensor height / focal length (24/840). Converting radians to degrees (@57.3) gives the 1.63714. Since then I have found my manual for the lens and it states the angle of view for the 600 + 1.4 ext is 1degree, 40 minutes, which equals 1.66667 degrees, so close enough (Canon has probably rounded the minutes). My detailed trig calculation gave a distance of 156.7m, which i sloppily rounded down to 150m as an approximation for purposes of my post. If I use your calculation, but use 5.357 effective microns (24mm/4480px), then I get 156.8m. I'm guessing the pixel pitch you quoted is for the actual photosite, and doesn't take into account gaps between the sites. I'm curious about the .84 factor as I'm not familiar with it. Where does it come from?
My mistake - I somehow thought you were using an R5, and I used the pixel pitch for that. Using the correct one for the 5DIV does give the same answer as you got of course! 600mm*1.4 = 0.84m. At least I showed you a simple method of doing the calculation.
 
Upvote 0

DanP

CR Pro
Dec 8, 2014
126
391
My mistake - I somehow thought you were using an R5, and I used the pixel pitch for that. Using the correct one for the 5DIV does give the same answer as you got of course! 600mm*1.4 = 0.84m. At least I showed you a simple method of doing the calculation.
Yes you did, and now I need to buy an R5 :)
 
  • Like
  • Haha
Reactions: 2 users
Upvote 0

DanP

CR Pro
Dec 8, 2014
126
391
There are two factors that might affect the sharpness at this distance.

The first, of course is the number of pixels on the subject. Its pretty low in those images.

The second is the air itself. At nearly 200m, there is substantial distortion caused by the air and it alone can keep you from seeing fine details. So, even with a really long focal length, you still have to deal with the air. Cold days seem to be better than warm to reduce the impact of air currents.

I think its the sum of these two issues that makes the image less sharp.
I'm glad I can blame it on something other than myself!
 
Upvote 0