Bird Photography Critique/Tips

[AlanF]

Private
Let's stop splitting hairs and get back to advice that we do in fact agree 100% on. It's not even fair comparing lenses on a 7D to those on a 5D III because the FF has better image quality. To get a good photo of a bird, you have to be close enough for the camera and lens you have in your hands. The better the camera and the better and longer the lens, the further away you can be and get a keeper.

The first tip is to get close enough, the closer the better, to resolve the plumage.
The second as pointed out is to get a good head angle.
The third is to get a decent composition in terms of background, foreground and placement.
Best of all is to have the bird doing something interesting instead of just sitting still.
Then clean it up as done by privatebydesign.
Plus lots more tips.

 

[jrista]

Thanks for that Alan.

Here are three images that I hope illustrate my point, that this is basically a distance/cropping issue and has nothing to do with the fact that the 100-400 is not "as good" as several other lenses. I agree that it isn't, but it is more than good enough to take some superb images of birds, assuming you have AFMA'd and your lens is not faulty and you get close enough.

First is a size corrected picture in a picture. It demonstrates how much bigger Alan's bird is than chasinglight's bird. By my calculations Alan's is around 3.5 times bigger, by area. I laid chasinglight's full image on top of Alan's and corrected for sensor size, so these two images are what they both saw through the viewfinder. The red bounding box in the 7D image is the 100% crop I posted earlier and represents less then 12% of the sensor area, around 5% of the 5D MkIII sensor area, the blue box.

Second is a crop of the same sensor area, for chasinglight I used the 100% crop from earlier, from Alan I cropped the same sensor area, so same magnification, of his bird's body. I have rotated Alan's bird body to better fit the comparison.

Third is a same magnification comparison of detail size. Yet again, I agree that the 100-400 is not the best lens available, but, it is clearly resolving detail comparable in size to the plumage in Alan's image.

Just to point this out...I don't think your relative size comparison is fair. You have a "vertical" frame scaled to the vertical height of a horizontal frame. Technically speaking, would it not be correct to rotate the vertical frame such that it is in the same orientation as the horizontal frame? At which point, I believe the birds are the same size, if not even giving the edge to the RWB.

If the OP's photo was indeed taken with the camera in a vertical orientation, then I really do think his lens needs to be AFMAed. With proper tuning, the 100-400 can produce some pretty sharp results. Within a reasonably acceptable distance, the 7D and 100-400 can produce acceptably "razor sharp" results, even...although that distance is indeed fairly close.

I would also point out that (just for reference), with some extensive testing, it appears that the AF confirm dot technique DOES NOT seem to work with the 7D. I've tried it about 50 times now with every lens I own, and it gets different, often significantly different results from FoCal or simple manual AFMA. Manual and FoCal tend to be more similar to each other, and I often choose a setting in between the two (if there IS an in between). If the lens was AFMAed with the AF confirmation dot technique, I'd either reset to zero and manually zero in on a more ideal AFMA setting, or use FoCal, which will do a pretty darn good job these days if you use good light.

 

[privatebydesign]

Thanks for that Alan.

Here are three images that I hope illustrate my point, that this is basically a distance/cropping issue and has nothing to do with the fact that the 100-400 is not "as good" as several other lenses. I agree that it isn't, but it is more than good enough to take some superb images of birds, assuming you have AFMA'd and your lens is not faulty and you get close enough.

First is a size corrected picture in a picture. It demonstrates how much bigger Alan's bird is than chasinglight's bird. By my calculations Alan's is around 3.5 times bigger, by area. I laid chasinglight's full image on top of Alan's and corrected for sensor size, so these two images are what they both saw through the viewfinder. The red bounding box in the 7D image is the 100% crop I posted earlier and represents less then 12% of the sensor area, around 5% of the 5D MkIII sensor area, the blue box.

Second is a crop of the same sensor area, for chasinglight I used the 100% crop from earlier, from Alan I cropped the same sensor area, so same magnification, of his bird's body. I have rotated Alan's bird body to better fit the comparison.

Third is a same magnification comparison of detail size. Yet again, I agree that the 100-400 is not the best lens available, but, it is clearly resolving detail comparable in size to the plumage in Alan's image.

Just to point this out...I don't think your relative size comparison is fair. You have a "vertical" frame scaled to the vertical height of a horizontal frame. Technically speaking, would it not be correct to rotate the vertical frame such that it is in the same orientation as the horizontal frame? At which point, I believe the birds are the same size, if not even giving the edge to the RWB.

If the OP's photo was indeed taken with the camera in a vertical orientation, then I really do think his lens needs to be AFMAed. With proper tuning, the 100-400 can produce some pretty sharp results. Within a reasonably acceptable distance, the 7D and 100-400 can produce acceptably "razor sharp" results, even...although that distance is indeed fairly close.

I would also point out that (just for reference), with some extensive testing, it appears that the AF confirm dot technique DOES NOT seem to work with the 7D. I've tried it about 50 times now with every lens I own, and it gets different, often significantly different results from FoCal or simple manual AFMA. Manual and FoCal tend to be more similar to each other, and I often choose a setting in between the two (if there IS an in between). If the lens was AFMAed with the AF confirmation dot technique, I'd either reset to zero and manually zero in on a more ideal AFMA setting, or use FoCal, which will do a pretty darn good job these days if you use good light.

The 7D sensor is 329mm², the 5D MkIII is 864mm², that makes a ff sensor 2.63 times bigger by area, a crop sensor easily fits inside a ff sensor vertically, a crop sensor is 22mm long, a ff sensor is 24mm high. My graphics are correct.

As I have demonstrated before, in a focal length limited situation (i.e. cropped) when using the same lens, the 7D will slightly out resolve a cropped 21mp ff sensor, not too sure about the 24mp 5D MkIII, but any differences will be absolutely minimal.

This is not a lens issue, it is just a distance/cropping issue.

 

[jrista]

One of the most important things in bird photography, more so than composition IMO, is head angle. You want the bird to engage the viewer...do a HA where the bill is 3-4 degrees inward towards the viewer tends to be best. Parallel to the sensor is ok, too. More than 5-7 degrees, and your getting into more specialized territory...it can work, but often not as well as a slighter angle. Any angle outward, away from the viewer, and the photo quickly loses its appeal, with a few exceptions (i.e a parent feeding a chick, where the parent may be facing slightly away.)

Once you get HA down, then worry about composition. BTW, I should note that the HA in your posted photo is a good example of GOOD HA...so keep striving for that. I think as long as HA is good, explicitly following the "rules" of composition is less important (especially since they are guidelines, not rules, in the first place.)

Wow that is the first time I have read that, but it makes sense. I will try to make use of that. Thanks for the advice!

Here are some examples of various head angles. Take note that a lot of the differences are very subtle, but there are clear issues with having a poor head angle that will pop out at you once you know what to look for.

First, a good HA. This is only a couple degrees forward. The lighting here is not ideal...I'd have preferred the side of the bird towards me be fully lit, but I did not really have that much control over it. The body pose and HA are within the ideal range, however:

This is an example of bad HA. It is a few degrees back. You should be able to see how the bill softens towards the end, as it moves out of the depth of field. You should also notice that the bird isn't quite looking into the frame, but ever so slightly ahead of it. Not readily visible to most viewers, but they usually still "sense" that the photo isn't quite as engaging. (Composition, in terms of scene contents, isn't great...but it demonstrates MUCH better lighting.)

Here is a head angle that is pretty much entirely parallel. This is usually quite acceptable, but not quite as engaging as a HA a few degrees towards the camera.

Finally, here is a HA turned much farther toward the camera than is usually ideal. I like the more cute pose here, and if I had increased my DOF a bit, I think it would have still been acceptable. I think this is a good example of why such a strong angle is less than ideal, however...again, you should see that towards the tip of the bill, it starts to soften a bit as it exits the depth of field. (I am still a relative beginner when it comes to bird photography myself...I have about a hear and a half under my belt. Keeping an eye on my DOF, and expanding it when necessary to include my whole subject, is an area I am still working on...as I do like the birds pose in this shot!)

Anyway, hopefully these Spotted Sandpiper photos will help demonstrate the subtle importance of head angle in bird photography. Aim for parallel or a few degrees forward (towards the camera), and you should be good.

Sure there are many "better" lenses, but the one he has is capable of much higher quality output. Why not move forwards with what we have, the 100-400 and PS, than spend other peoples money when there are basic techniques we can suggest to improve our skills and output that would need work even if he had a 600 f4 IS MkII.

For the record I think this is great advice. I want to learn to get everything out of the gear I have before I plunk down 4-12k. I would hate to upgrade and still get not stellar results solely because I don't have the proper knowledge or skills.

I'd say that is a good practice. If you are anything like me, you will know when your gear is holding you back. I also have the 100-400. I think the 7D is a fine camera, produces great IQ in most circumstances (which for my bird photography is usually in good to evening light, ISO 200 - 1600), and has great features that support bird photography. The 100-400, when properly tuned with AFMA, produces acceptably sharp images most of the time. It should be noted that at 400mm, f/7.1 tends to be the sharpest, while f/5.6 will be visibly soft. Before getting my new lens, I shot at f/7.1 almost exclusively, sometimes stopping down to f/8 and rarely opening up to f/6.3.

I would tune your lens for your copy of the 7D, and start shooting at f/7.1. You should see individual barbs of each feather (a feather is a central shaft, on either side of which is a vane of barbes, which are interconnected via barbules off each barb...you will RARELY see barbules in a photo, but in an acceptably sharp photo, you should see barbs.) There are three things that will soften the barbs of a birds feathers...distance too great, missfocus, bird motion or camera shake. Distance is usually the biggest problem early on. Depending on the type of bird, either learning the right behavior to exhibit that gets you close, or camouflaging yourself to hide in plain side, are was of solving that problem.

Missfocus is, sadly, just a fact of life with older Canon cameras (i.e. pre 61pt AF). The 7D does not have the most reliable AF system. When it nails it, it usually NAILS it. When it doesn't, its off just slightly enough that you can't really tell in the viewfinder, but definitely can in post. The best tactic here is to always shoot a burst. I try to get at least three shots in every time, if not five. That usually results in one of the frames being acceptably sharp. Sadly, the other two-four are usually soft, reducing barbs to mush. I use rear-button focus as well (I reconfigure the camera to unlink AF from the shutter button, and link it to the * button on the back of the camera...this direct control can be very useful in forcing AF to occur when and where you need it.) If you can't seem to grab focus, move the current AF point off the subject, press the * button to immediately force AF on the background, then return the AF point to the subject and press the * button to immediately force AF again. This will usually get you sharper results for a short time. It ain't perfect, but bursting definitely helps. You either just have to either buy a lot of disk space, or become a culling nutcase who rejects anything that isn't razor sharp.

Finally, bird motion, and sometimes camera shake, are the last culprit of soft barbs. You usually need pretty high shutter speeds to stop bird motion, especially songbirds and smaller shorebirds. A bird like a chickadee will often require over 1/1000s shutter speeds, with 1/2000s usually being more ideal. A bird like your RWB is usually less jittery than the ever-moving chickadee, so you might be able to get away with a shutter speed as low as 1/800th before you experience detrimental barb softening. As I'm sure you already know, high shutter speed means high ISO. I used to shoot at ISO settings as high as 2500, but I've come to the conclusion that the 7D is good up to 1600, and beyond that it is usually unacceptable. So long as you keep your hands steady with IS on, or use a tripod (with IS off!!), you should get sharp shots.

(NOTE: Regarding ISO...if you have really good light...such as light a couple hours after sunrise or a couple hours before sunset...bright, but at a good angle to shade well, with the sun behind and slightly over your shoulder, then if necessary you might be able to get away with ISO 2000, 2500, and maybe 3200. I suspect the diminished dynamic range will be a problem, but if you REALLY need the shutter speed, then go for it. Otherwise, the extra noise at those settings just isn't worth it (after ISO 1600, a secondary downstream amplifier kicks in, which is why noise performance falls off a cliff on the 7D at that point). I highly recommend Topaz DeNoise 5 as the best denoise tool for the 7D. Its default RAW settings are pretty good, it does a decent job automatically identifying and masking off detail areas, so it cleans up smooth backgrounds REALLY well. It is well worth the money, and it definitely cheaper than moving to the 5D III.)

Well, that is the best series of tips I can offer you right now. If you see sharp barbs, your doing good. If not, then one of those three things will most likely be the culprit. Keep an eye on HA, and composition will eventually work itself out as the other things become second nature and you have time to focus your attention on it.

 

[jrista]

Thanks for that Alan.

Here are three images that I hope illustrate my point, that this is basically a distance/cropping issue and has nothing to do with the fact that the 100-400 is not "as good" as several other lenses. I agree that it isn't, but it is more than good enough to take some superb images of birds, assuming you have AFMA'd and your lens is not faulty and you get close enough.

First is a size corrected picture in a picture. It demonstrates how much bigger Alan's bird is than chasinglight's bird. By my calculations Alan's is around 3.5 times bigger, by area. I laid chasinglight's full image on top of Alan's and corrected for sensor size, so these two images are what they both saw through the viewfinder. The red bounding box in the 7D image is the 100% crop I posted earlier and represents less then 12% of the sensor area, around 5% of the 5D MkIII sensor area, the blue box.

Second is a crop of the same sensor area, for chasinglight I used the 100% crop from earlier, from Alan I cropped the same sensor area, so same magnification, of his bird's body. I have rotated Alan's bird body to better fit the comparison.

Third is a same magnification comparison of detail size. Yet again, I agree that the 100-400 is not the best lens available, but, it is clearly resolving detail comparable in size to the plumage in Alan's image.

Just to point this out...I don't think your relative size comparison is fair. You have a "vertical" frame scaled to the vertical height of a horizontal frame. Technically speaking, would it not be correct to rotate the vertical frame such that it is in the same orientation as the horizontal frame? At which point, I believe the birds are the same size, if not even giving the edge to the RWB.

If the OP's photo was indeed taken with the camera in a vertical orientation, then I really do think his lens needs to be AFMAed. With proper tuning, the 100-400 can produce some pretty sharp results. Within a reasonably acceptable distance, the 7D and 100-400 can produce acceptably "razor sharp" results, even...although that distance is indeed fairly close.

I would also point out that (just for reference), with some extensive testing, it appears that the AF confirm dot technique DOES NOT seem to work with the 7D. I've tried it about 50 times now with every lens I own, and it gets different, often significantly different results from FoCal or simple manual AFMA. Manual and FoCal tend to be more similar to each other, and I often choose a setting in between the two (if there IS an in between). If the lens was AFMAed with the AF confirmation dot technique, I'd either reset to zero and manually zero in on a more ideal AFMA setting, or use FoCal, which will do a pretty darn good job these days if you use good light.

The 7D sensor is 329mm², the 5D MkIII is 864mm², that makes a ff sensor 2.63 times bigger by area, a crop sensor easily fits inside a ff sensor vertically, a crop sensor is 22mm long, a ff sensor is 24mm high. My graphics are correct.

As I have demonstrated before, in a focal length limited situation (i.e. cropped) when using the same lens, the 7D will slightly out resolve a cropped 21mp ff sensor, not too sure about the 24mp 5D MkIII, but any differences will be absolutely minimal.

This is not a lens issue, it is just a distance/cropping issue.

I think you are underestimating the pixel density of the 7D. If we stick to purely mathematical terms (ignoring noise for the moment, as it does not seem to be a problem in either photo), the 7D pixels are 1.45x smaller than the 5D III pixels. That means you can fit almost one and a half 7D pixels horizontally, and one and a half 7D pixels vertically, into each 5D III pixel. That is an increase in resolution, in two dimensions, of 111% (or 2.11x more spatial resolution than the 5D III):

Fig: A comparison of 7D pixels to 5D III pixels, actually to scale (blue squares are 625 pixels a side, green squares are 430 pixels a side.) You can see that you can pack in 100% of one 7D pixel, and another 111% of three other pixels in varying ratios, into the area of a single 5D III pixel. You can fit nearly nine 7D pixels into the area of four 5D III pixels.

Even if we cut that spatial resolution benefit in HALF, the 7D is still capturing 50% more detail than the 5D III. That isn't "absolutely minimal". You are correct about the dimensions (I missed mention of the 5D III in your post, and thought you were comparing two 7D photos), however if you rotate it by 90°, then increase its size by 50%, that would produce a much more accurate representation of the relative sizes of the birds as far as output image dimensions go.

 

[privatebydesign]

I think you are underestimating the pixel density of the 7D. If we stick to purely mathematical terms (ignoring noise for the moment, as it does not seem to be a problem in either photo), the 7D pixels are 1.45x smaller than the 5D III pixels. That means you can fit almost one and a half 7D pixels horizontally, and one and a half 7D pixels vertically, into each 5D III pixel. That is an increase in resolution, in two dimensions, of 111%.

Even if we cut that spatial resolution benefit in HALF, the 7D is still capturing 50% more detail than the 5D III. That isn't "absolutely minimal". You are correct about the dimensions (I missed mention of the 5D III in your post, and thought you were comparing two 7D photos), however if you rotate it by 90°, then increase its size by 50%, that would produce a much more accurate representation of the relative sizes of the birds as far as output image dimensions go.

I absolutely am not underestimating the pixel density of the 7D, I have done extensive tests with the 7D, most people vastly overestimate the value of pixel density, all pixels are not equal.

Besides, the demonstration was not about pixel density, it was about reproduction ratios, or, making a same sized print or screen image from both Alan's 5D MkIII and chasinglights 7D cameras, that is the only relevant metric. I was pointing out that Alan's bird took up a much bigger area of sensor real estate than chasinglight's did, and it does, so any comparisons of detail, or criticisms of the 7D's lack of detail, was fundamentally flawed, and it is.

 

[jrista]

I think you are underestimating the pixel density of the 7D. If we stick to purely mathematical terms (ignoring noise for the moment, as it does not seem to be a problem in either photo), the 7D pixels are 1.45x smaller than the 5D III pixels. That means you can fit almost one and a half 7D pixels horizontally, and one and a half 7D pixels vertically, into each 5D III pixel. That is an increase in resolution, in two dimensions, of 111%.

Even if we cut that spatial resolution benefit in HALF, the 7D is still capturing 50% more detail than the 5D III. That isn't "absolutely minimal". You are correct about the dimensions (I missed mention of the 5D III in your post, and thought you were comparing two 7D photos), however if you rotate it by 90°, then increase its size by 50%, that would produce a much more accurate representation of the relative sizes of the birds as far as output image dimensions go.

I absolutely am not underestimating the pixel density of the 7D, I have done extensive tests with the 7D, most people vastly overestimate the value of pixel density, all pixels are not equal.

Besides, the demonstration was not about pixel density, it was about reproduction ratios, or, making a same sized print or screen image from both Alan's 5D MkIII and chasinglights 7D cameras, that is the only relevant metric. I was pointing out that Alan's bird took up a much bigger area of sensor real estate than chasinglight's did, and it does, so any comparisons of detail, or criticisms of the 7D's lack of detail, was fundamentally flawed, and it is.

I'm not sure where print came into play. I think, when evaluating sharpness, the only thing that really matters is actual image dimensions. The 7D's physical APS-C frame is indeed smaller than the 5D III's by the ratio you mention, but the output image is MUCH closer to the 5D III's output image dimensions.

The 5D III full RAW image size is 5760x3840. The 7D full RAW image size is 5184x3456. In relative terms, the image sizes are red and blue below, where as your skewed sizing, based on physical sensor dimensions, greatly puts the 7D at an undue and unrealistic disadvantage (green):

I know you dislike the 7D, but I think you are making a very unfair comparison that isn't benefiting the underlying point at all. The 7D it IS capable of being razor sharp, even with the 100-400...and with that particular lens, the 7D is not capable of really reaching its full potential.

I don't think chasinglight's problem is that he is using a 7D. Far more likely than that, I think there is probably some lens/body tuning that could be done, and after that, it is largely a matter of technique. If you use a tool properly, you can maximize it's potential, and the 7D has a LOT of potential. That doesn't mean it will produce better results if you have the ability to fill the frame with both the 7D and the 5D III...the simple fact that the 5D III has more total pixels means it will produce a better image. All it means is that chasinglight can work on a few things that cost him nothing, yet allow him to produce better results.

 

[privatebydesign]

I'm not sure where print came into play. I think, when evaluating sharpness, the only thing that really matters is actual image dimensions. The 7D's physical APS-C frame is indeed smaller than the 5D III's by the ratio you mention, but the output image is MUCH closer to the 5D III's output image dimensions.

The 5D III full RAW image size is 5760x3840. The 7D full RAW image size is 5184x3456. In relative terms, the image sizes are red and blue below, where as your skewed sizing, based on physical sensor dimensions, greatly puts the 7D at an undue and unrealistic disadvantage (green):

I know you dislike the 7D, but I think you are making a very unfair comparison that isn't benefiting the underlying point at all. The 7D it IS capable of being razor sharp, even with the 100-400...and with that particular lens, the 7D is not capable of really reaching its full potential.

I don't think chasinglight's problem is that he is using a 7D. Far more likely than that, I think there is probably some lens/body tuning that could be done, and after that, it is largely a matter of technique. If you use a tool properly, you can maximize it's potential, and the 7D has a LOT of potential. That doesn't mean it will produce better results if you have the ability to fill the frame with both the 7D and the 5D III...the simple fact that the 5D III has more total pixels means it will produce a better image. All it means is that chasinglight can work on a few things that cost him nothing, yet allow him to produce better results.

I don't quite know where to start addressing this post.

I think the best thing would be for you to go and reread the thread. I am the one who has constantly said the 7D is a more than capable camera, how you have managed to infer from that "I know you dislike the 7D" , I just don't fathom and am in complete disbelief at. Further "The 7D it IS capable of being razor sharp, even with the 100-400.." I KNOW, and I posted 11 links of razor sharp images shot with a 7D and 100-400, of birds, at 400, and some wide open!!!!! DID YOU NOT SEE THEM? Or this "I don't think chasinglight's problem is that he is using a 7D" I have repeatedly said his main problem is subject distance (magnification) NOT THE 7D. Funnily enough that is why I used MAGNIFICATION as the comparison between his and Alan's images. Alan suggested an $8,000 investment, I said no, GET CLOSER. I know in focal length limited situations the 7D will outperform (by a very small margin) the cropped 21MP 5D MkII/1Ds MkIII, and while I have not tested it against the 5D MkIII I would expect the results to be similar. To summarize that last sentence, I know in focal length limited situations the 7D will perform better than a cropped FF image, though not by much and certainly not as many as the pixel numbers would have you believe. I also KNOW, ALL PIXELS ARE NOT EQUAL, to compare one 7D pixel to one 5D MkIII pixel is moronic, it just doesn't work like that.

In focal length limited situations, as this one is, per pixel comparisons are worse than useless, subject magnification is THE ONLY FAIR comparison, it is the only relevant metric for comparison of output. It maintains equivalency, you are comparing like with like. How do you adjust for different DOF figures if you compare on a per pixel basis? How do you adjust a sharpness figure when you magnify one image more than the other? You can't! Compare same magnification and you get a true comparison, you know what will look better in print, on a screen, or any other end use output.

 

[jrista]

I'm not sure where print came into play. I think, when evaluating sharpness, the only thing that really matters is actual image dimensions. The 7D's physical APS-C frame is indeed smaller than the 5D III's by the ratio you mention, but the output image is MUCH closer to the 5D III's output image dimensions.

The 5D III full RAW image size is 5760x3840. The 7D full RAW image size is 5184x3456. In relative terms, the image sizes are red and blue below, where as your skewed sizing, based on physical sensor dimensions, greatly puts the 7D at an undue and unrealistic disadvantage (green):

I know you dislike the 7D, but I think you are making a very unfair comparison that isn't benefiting the underlying point at all. The 7D it IS capable of being razor sharp, even with the 100-400...and with that particular lens, the 7D is not capable of really reaching its full potential.

I don't think chasinglight's problem is that he is using a 7D. Far more likely than that, I think there is probably some lens/body tuning that could be done, and after that, it is largely a matter of technique. If you use a tool properly, you can maximize it's potential, and the 7D has a LOT of potential. That doesn't mean it will produce better results if you have the ability to fill the frame with both the 7D and the 5D III...the simple fact that the 5D III has more total pixels means it will produce a better image. All it means is that chasinglight can work on a few things that cost him nothing, yet allow him to produce better results.

I don't quite know where to start addressing this post.

I think the best thing would be for you to go and reread the thread.

I'm not debating the entire thread...only "Reply #38", where I believe you to make an incorrect comparison between a 7D image and a 5D III image by scaling the 7D incorrectly, which results in a significant and further reduction in detail of the 7D image relative to the 5D III image. When it comes to what was actually resolved, magnification relative to the final output is not the sole factor.

I am the one who has constantly said the 7D is a more than capable camera, how you have managed to infer from that "I know you dislike the 7D" , I just don't fathom and am in complete disbelief at.

You've come off at me fairly strong in a number of threads over the last couple of months where you seem to put the 7D in exceptionally bad light relative to other cameras. You have repeatedly made a point of calling out how the 7D's resolution advantage is effectively meaningless. Given the repetition, I can only conclude you dislike the 7D. I also still believe you are comparing things incorrectly, and putting the 7D in unfair light. Even detailed articles about the 7D's resolution advantage from a well respected Ph.D., Roger Clark, seem insufficient to get you to even meet at some middle ground.

Sorry if I've misread you, but I've been going by both the intensity of your responses to me when I debate the issue with you, and the frequency at which you bring up how nearly worthless the 7D's resolution advantage supposedly is.

Further "The 7D it IS capable of being razor sharp, even with the 100-400.." I KNOW, and I posted 11 links of razor sharp images shot with a 7D and 100-400, of birds, at 400, and some wide open!!!!! DID YOU NOT SEE THEM?

No, I came in on page three, I guess...saw a post by you that put the 7D in unfairly poor light, and responded. Again, I wasn't responding to the entire thread, and having read through the rest, I'm a bit confused by what seems to be conflicting information in different posts.

Or this "I don't think chasinglight's problem is that he is using a 7D" I have repeatedly said his main problem is subject distance (magnification) NOT THE 7D. Funnily enough that is why I used MAGNIFICATION as the comparison between his and Alan's images.

Using magnification alone, and ignoring pixel sizes completely, is the issue I was debating. I think that is oversimplifying the issue. There are effectively an unlimited number of potential output sizes, both smaller and larger than the native image size for each camera. I think it is only fair to compare them at their native sizes. Assuming the same bird is photographed at the same distance with both cameras, there is both a magnification difference as well as a pixels on subject difference. Since the issue is sharpness, pixels on subject matter. This is effectively a pixel peeping situation, so considerably reducing the size of the 7D's native output is skewing the basis for comparison.

I get the feeling we are just going to agree to disagree on this point...so I'm happy to end the debate here.

Alan suggested an $8,000 investment, I said no, GET CLOSER. I know in focal length limited situations the 7D will outperform (by a very small margin) the cropped 21MP 5D MkII/1Ds MkIII, and while I have not tested it against the 5D MkIII I would expect the results to be similar. To summarize that last sentence, I know in focal length limited situations the 7D will perform better than a cropped FF image, though not by much and certainly not as many as the pixel numbers would have you believe. I also KNOW, ALL PIXELS ARE NOT EQUAL, to compare one 7D pixel to one 5D MkIII pixel is moronic, it just doesn't work like that.

I agree, one needs to get closer. But in practice, there are limitations in doing that. Birds infrequently just sit still and let you get as close as you want. There IS an argument for buying better gear. It may not be the best advice for someone in chasinglights position, but it is not an invalid argument either. There are also consequences of getting closer. There is a zone within which one is closer to the bird, but also affecting it's behavior. If all you have is a 400mm lens, then either you deal with the behavioral changes, wait long enough for those behavioral changes to subside (which may never occur), slap the lens on a sensor with denser pixels, or get a longer lens.

We disagree about the "by a very small margin" part. All sensor pixels may not be equal, but that doesn't really matter. One only has to look at the two photos at full size to compare sharpness. If one reduces the resolution of one image from its original size, THAT makes for an unfair comparison. You are converting your bias against the 7D, the notion that its pixels only give it a "very small margin" benefit, directly into actual image comparisons. If you are going to scale for comparison, scale them both to the same size, don't scale one and leave the other unscaled.

In focal length limited situations, as this one is, per pixel comparisons are worse than useless, subject magnification is THE ONLY FAIR comparison, it is the only relevant metric for comparison of output. It maintains equivalency, you are comparing like with like. How do you adjust for different DOF figures if you compare on a per pixel basis? How do you adjust a sharpness figure when you magnify one image more than the other? You can't! Compare same magnification and you get a true comparison, you know what will look better in print, on a screen, or any other end use output.

Again, I disagree. As a matter of detail resolved, for example feather barbs, pixel peeping WILL show a meaningful difference. That difference may or may not matter for the artists final choice of presentation. If the image is shrunk and uploaded online, and that is the only thing ever done with it, then sure...per-pixel comparisons are effectively meaningless. On the other hand, if you scale up a heavily cropped 5D III image against a lightly cropped 7D image, and print at 16x24, then the 7D's resolution benefit IS meaningful. Per-pixel comparisons...or rather more realistically comparing both cropped images at full size, is an entirely valid way to evaluate IQ.

I am not even considering anything like a scientific test here, either...it doesn't take much to see improved sharpness with a little bit of eyeballing. I'm not talking noise here, either...just sharpness. The 7D image may indeed be considerably noisier in smooth OOF backgrounds, however even a noisy 7D image can still be razor sharp in the detail areas where it matters. Noise is also considerably harder to see in a print, so the greater resolving power of the 7D will have a meaningful impact on the sharpness of the final print. The 5D III noise would likely clean up considerably better, but the improvement is going to be less visible in a print than the sharpness factor.

As for DOF, from a technical standpoint, you are correct. From a practical standpoint, it usually doesn't matter. Unless you have a DOF issue such that a critical part of your subject is affected...for example the end of a birds bill ends up out of focus...that matters. If the important parts of your subject are in focus, at least in bird photography, DOF usually isn't a make or break issue. More of the butt end of a bird might be more blurry with one camera over the other, but if the eyes, bill, and a sufficient part of the birds head or body are sharp and clear...that's what really matters.

Anyway, I don't want to antagonize you further. You and I will likely never see eye to eye on the 7D/resolution issue (or, for that matter, the pixel density/resolution issue for any set of cameras, now or in the future). We disagree. I guess we just agree to disagree.

 

[chasinglight]

I'd say that is a good practice. If you are anything like me, you will know when your gear is holding you back. I also have the 100-400. I think the 7D is a fine camera, produces great IQ in most circumstances (which for my bird photography is usually in good to evening light, ISO 200 - 1600), and has great features that support bird photography. The 100-400, when properly tuned with AFMA, produces acceptably sharp images most of the time. It should be noted that at 400mm, f/7.1 tends to be the sharpest, while f/5.6 will be visibly soft. Before getting my new lens, I shot at f/7.1 almost exclusively, sometimes stopping down to f/8 and rarely opening up to f/6.3.

I would tune your lens for your copy of the 7D, and start shooting at f/7.1. You should see individual barbs of each feather (a feather is a central shaft, on either side of which is a vane of barbes, which are interconnected via barbules off each barb...you will RARELY see barbules in a photo, but in an acceptably sharp photo, you should see barbs.) There are three things that will soften the barbs of a birds feathers...distance too great, missfocus, bird motion or camera shake. Distance is usually the biggest problem early on. Depending on the type of bird, either learning the right behavior to exhibit that gets you close, or camouflaging yourself to hide in plain side, are was of solving that problem.

Thanks for the advice (all of it, not just what I quoted). I have seen my copy of the 100-400 produce much sharper images than this (such as the one below; AFMA 0). I think actually took this shot of the eastern screech owl at f/7.1. I think I actually started to notice more inconsistency after I used Focal to AFMA the lens a few months ago; this produced a -3. I performed Focal a few times today getting -3, 3, and 1. So I decided to go back to 0 and see how that works out.... could be that I was just trying to over sharpen the knife and instead made it dull...

 

[Rick Massie]

100-400mm is really irritating on the 7d too... (sarc., implied )
1/320 sec.
f/5.6
ISO 100

I love my 400mm 5.6 but I will never get a shot as sharp as this at 1/320sec, probably not even with a monopod. That's my one dislike of the lens.

Back on topic for the OP- your edits sounds about right. Getting closer will definitely help. The other option that sometimes works for me is bumping up the "clarity" a little in lightroom. It's very easy to overdo it, but it can help accentuate sharpening if you're careful.

 

[privatebydesign]

My replies are in red.

I'm not sure where print came into play. I think, when evaluating sharpness, the only thing that really matters is actual image dimensions. The 7D's physical APS-C frame is indeed smaller than the 5D III's by the ratio you mention, but the output image is MUCH closer to the 5D III's output image dimensions.

The 5D III full RAW image size is 5760x3840. The 7D full RAW image size is 5184x3456. In relative terms, the image sizes are red and blue below, where as your skewed sizing, based on physical sensor dimensions, greatly puts the 7D at an undue and unrealistic disadvantage (green):

I know you dislike the 7D, but I think you are making a very unfair comparison that isn't benefiting the underlying point at all. The 7D it IS capable of being razor sharp, even with the 100-400...and with that particular lens, the 7D is not capable of really reaching its full potential.

I don't think chasinglight's problem is that he is using a 7D. Far more likely than that, I think there is probably some lens/body tuning that could be done, and after that, it is largely a matter of technique. If you use a tool properly, you can maximize it's potential, and the 7D has a LOT of potential. That doesn't mean it will produce better results if you have the ability to fill the frame with both the 7D and the 5D III...the simple fact that the 5D III has more total pixels means it will produce a better image. All it means is that chasinglight can work on a few things that cost him nothing, yet allow him to produce better results.

I don't quite know where to start addressing this post.

I think the best thing would be for you to go and reread the thread.

I'm not debating the entire thread...only "Reply #38",It does amuse me, however, that you claim to limit your involvement now to a specific post in this thread yet are obviously carrying baggage from other threads that have no relevance to this one. where I believe you to make an incorrect comparison between a 7D image and a 5D III image by scaling the 7D incorrectly, which results in a significant and further reduction in detail of the 7D image relative to the 5D III image. Err, that would be completely contrary to the effects of enlargement, enlarging more makes things less sharp, my comparison is set up to be fair and favours the 7D, When it comes to what was actually resolved, magnification relative to the final output is not the sole factor. Output is the sole factor that needs or should be considered, be that an 8"x10" print, a 20"x30" print, a 30"hi res monitor or an iPad, you can only make a fair comparison if both outputs are the same size, theory and speculation is not worth anything, output is king. What is the point of comparing different sized outputs? None, it is an exercise in futility.

I am the one who has constantly said the 7D is a more than capable camera, how you have managed to infer from that "I know you dislike the 7D" , I just don't fathom and am in complete disbelief at.

You've come off at me fairly strong in a number of threads over the last couple of months where you seem to put the 7D in exceptionally bad light relative to other cameras. No, you have taken umbrage at my position, supported with my own images and hands on experience that I don't believe that it has a "tele advantage" in real world shooting over a cropped same generation ff sensor, I have never berated the 7D and always qualified that opinion that there are still very good reasons to buy one, just not because it gives you a "free" TC. You have repeatedly made a point of calling out how the 7D's resolution advantage is effectively meaningless. From my personal hands on real world testing that was the case. Given the repetition, I can only conclude you dislike the 7D. I'd suggest you reread the actual comments I made, they were always restricted to the output resolution and peoples false opinions of it based on theory, never the cameras capabilities. I also still believe you are comparing things incorrectly. That is your perogative, but just as your objections to my posts in the DOF/compression thread, you are wrong, you just don't see it yet, and putting the 7D in unfair light. If you understood what I was doing you would realise that reducing the size of the 7D output, to compare like with like, actually helps the 7D. Making things smaller makes them sharper. Even detailed articles about the 7D's resolution advantage from a well respected Ph.D., Roger Clark, seem insufficient to get you to even meet at some middle ground. What middle ground is there to meet? This isn't a negotiation! I have used and tested both formats against each other, I set up specific tests to determine the actual realisable resolution differences, I have posted my results of a test set up to maximise the advantage the 7D has and even in perfect test bench situations the differences are small, in real world shooting AF makes more difference. You have displayed no experience with the ff cameras, have run no tests, have posted no comparisons and yet continue to argue your theoretical opinion, based on... theory. Photography is a visual medium, theory is all well and good, but show me the photographs. I tested them, I made them, I printed them, my experience is not based on theory,it is based on images.

Sorry if I've misread you, but I've been going by both the intensity of your responses to me when I debate the issue with you, and the frequency at which you bring up how nearly worthless the 7D's resolution advantage supposedly is. In my experience it is, but that is a far cry and very different from calling the 7D worthless. Just because I don't respect your theory, when I have prints that prove it incorrect, doesn't mean I don't like the camera, I was debating your opinion, not the cameras worth. As for the frequency, well the more often I see incorrect theory the more often I will try to educate, DOF and equivelance, and perspective are more of my pets, I have been linking to the Equivelance page for 18 months now and there is a definite turn of common knowledge that now gets the concept, I don't claim many converts, but a few.

Further "The 7D it IS capable of being razor sharp, even with the 100-400.." I KNOW, and I posted 11 links of razor sharp images shot with a 7D and 100-400, of birds, at 400, and some wide open!!!!! DID YOU NOT SEE THEM?

No, I came in on page three, Well there's the thing, take something out of context and you don't know what people are talking about. I guess...saw a post by you that put the 7D in unfairly poor light, and responded. Again, I wasn't responding to the entire thread, and having read through the rest, I'm a bit confused by what seems to be conflicting information in different posts. I suspect that is your preconceived discrimination of what you expected my opinion to be, not any inconsistencies in my posting.

Or this "I don't think chasinglight's problem is that he is using a 7D" I have repeatedly said his main problem is subject distance (magnification) NOT THE 7D. Funnily enough that is why I used MAGNIFICATION as the comparison between his and Alan's images.

Using magnification alone, and ignoring pixel sizes completely, is the issue I was debating. I think that is oversimplifying the issue. There are effectively an unlimited number of potential output sizes, Yes, and the only fair comparison is to equalise them, if I want an 8" x12" I want an 8" x12", I don't want an 8" x 12" from the 5D MkIII and a 12" x 18" from the 7D, which is what you are saying is a fair comparison. both smaller and larger than the native image size for each camera. Digital cameras don't have a native image size! They have a native number of pixels, but that is very different. I think it is only fair to compare them at their native sizes. If you were to standardise ppi as a comparison you would be going against any common sense idea of fair, or comparison, you would also put the 7D at considerable disadvantage as you would be enlargeing it around 50% more. Assuming the same bird is photographed at the same distance with both cameras, there is both a magnification difference Not in focal length limited situations, like this one, the magnification onto the sensor is the same. as well as a pixels on subject difference. Since the issue is sharpness, pixels on subject matter. This is effectively a pixel peeping situation, so considerably reducing the size of the 7D's native output is skewing the basis for comparison. No it isn't, it is favouring it, I don't understand why you struggle with this concept. More pixels equals sharper, less enlargement equals sharper, my fair comparison uses more pixels from the 7D and enlarges it less than yours, that puts it at an ADVANTAGE.

I get the feeling we are just going to agree to disagree on this point...so I'm happy to end the debate here. And yet you didn't.......

Alan suggested an $8,000 investment, I said no, GET CLOSER. I know in focal length limited situations the 7D will outperform (by a very small margin) the cropped 21MP 5D MkII/1Ds MkIII, and while I have not tested it against the 5D MkIII I would expect the results to be similar. To summarize that last sentence, I know in focal length limited situations the 7D will perform better than a cropped FF image, though not by much and certainly not as many as the pixel numbers would have you believe. I also KNOW, ALL PIXELS ARE NOT EQUAL, to compare one 7D pixel to one 5D MkIII pixel is moronic, it just doesn't work like that.

I agree, one needs to get closer. But in practice, there are limitations in doing that. Birds infrequently just sit still and let you get as close as you want. There IS an argument for buying better gear. It may not be the best advice for someone in chasinglights position, but it is not an invalid argument either. There are also consequences of getting closer. There is a zone within which one is closer to the bird, but also affecting it's behavior. If all you have is a 400mm lens, then either you deal with the behavioral changes, wait long enough for those behavioral changes to subside (which may never occur), slap the lens on a sensor with denser pixels there you go again with your theory, or get a longer lens. Again, if you had taken a few moments to acquaint yourself with the thread you would have realised we were talking about a very specific situation here. Alan has a FF body and a 300 lens with a 2XTC, chasinglight has a 1.6 crop body with a 100-400 lens, they are both focal length limited, they both need to get closer, but obviously, not to the detriment of the bird or the image. Having said that chasinglights setup actually gives him a narrower fov than Alan (though less magnification), hence my (everybody's) consistent suggestion that that was where the main improvement would come.

We disagree about the "by a very small margin" part. That is fine, when I see your comparison images I might respect your opinion more, having said that I am not the only person who has come to the conclusion I have, several other prominent posters, have said exactly the same thing, but then they actually had experience with the cameras, they were not relying on theory. All sensor pixels may not be equal, but that doesn't really matter. In my fair comparison, not it doesn't matter, in your ridiculous and contrived comparison it is supremely important, in fact that is all you are comparing. One only has to look at the two photos at full size to compare sharpness. No, one has to look at two photos THE SAME SIZE to compare sharpness. I fear you are so anti me you will refuse to see that, just like the "compression" you were convinced was true, it wasn't, and your comparison isn't If one reduces the resolution of one image from its original size, THAT makes for an unfair comparison.You don't sound like you print much, though you claim you do, there is no reduction in resolution, there is an increase in resolution for the denser sensor and an increase in ppi figures to take that difference into account, simple really, nothing is lost. And, again, the denser sensor is at an advantage. You are converting your bias against the 7D, I don't have one, I have an opinion backed up with images and hands on experience that is agreed with by other owners of the two that contradicts common wisdom, your theory, that the much higher pixel density of the 7D gives it a meaningful resolution advantage in real world shooting over a same generation ff sensor cropped to the same sensor area. the notion that its pixels only give it a "very small margin" benefit, directly into actual image comparisons. If you are going to scale for comparison, scale them both to the same size YES, THAT IS THE POINT, THEY MUST BE SCALED TO THE SAME SIZE, not the same %, don't scale one and leave the other unscaled.

In focal length limited situations, as this one is, per pixel comparisons are worse than useless, subject magnification is THE ONLY FAIR comparison, it is the only relevant metric for comparison of output. It maintains equivalency, you are comparing like with like. How do you adjust for different DOF figures if you compare on a per pixel basis? How do you adjust a sharpness figure when you magnify one image more than the other? You can't! Compare same magnification and you get a true comparison, you know what will look better in print, on a screen, or any other end use output.

Again, I disagree. As a matter of detail resolved, for example feather barbs, pixel peeping WILL show a meaningful difference. That difference may or may not matter for the artists final choice of presentation. If the image is shrunk and uploaded online, and that is the only thing ever done with it, then sure...per-pixel comparisons are effectively meaningless. On the other hand, if you scale up a heavily cropped 5D III image against a lightly cropped 7D image, and print at 16x24, then the 7D's resolution benefit IS meaningful. Oh, you almost got it then! A tightly cropped 5D MkIII image and a lightly cropped 7D image WILL REPRESENT THE SAME SENSOR AREA! We strongly disagree on how meaningful the denser pixel advantage is, but then my opinion is based on actual hands on experience and I draw my own conclusions from my own tests, but you nearly stumbled into the whole point there without looking. Per-pixel comparisons...or rather more realistically comparing both cropped images at full size, is an entirely valid way to evaluate IQ. But then you lost it again :-( comparing smaller to bigger is not a fair comparison.

I am not even considering anything like a scientific test here, either...it doesn't take much to see improved sharpness with a little bit of eyeballing. I'm not talking noise here, either...just sharpness. The 7D image may indeed be considerably noisier in smooth OOF backgrounds, however even a noisy 7D image can still be razor sharp in the detail areas where it matters. Noise is also considerably harder to see in a print, so the greater resolving power of the 7D will have a meaningful impact on the sharpness of the final print. The 5D III noise would likely clean up considerably better, but the improvement is going to be less visible in a print than the sharpness factor. All theory and conjecture, so irrelevant. However you have posted images before that show heavy noise reduction in oof areas, but none in detailed areas, you seems to think noisy detailed areas are sharper because they have noise, they are not, they just appear sharper, put some noise in a very slightly soft image and it appears sharper too.

As for DOF, from a technical standpoint, you are correct. From a practical standpoint, it usually doesn't matter. No, it doesn't matter in same magnification enlargements, but look at stuff 50% more enlarged, or at the pixel level, and your CoC figures go out the window. Unless you have a DOF issue such that a critical part of your subject is affected...for example the end of a birds bill ends up out of focus...that matters. If the important parts of your subject are in focus, At what magnification? My same sized magnification, or your 50% more for the 7D? DoF and sharpness changes dramatically with enlargement, my comparison equalises that, yours does not. at least in bird photography, DOF usually isn't a make or break issue. More of the butt end of a bird might be more blurry with one camera over the other,No, in focal length limited situations, like this one, the DoF is the same for both cameras IF you use the same output magnification (if the birds in print are the same size), again, my comparison allows for this, yours does not. but if the eyes, bill, and a sufficient part of the birds head or body are sharp and clear...that's what really matters.

Anyway, I don't want to antagonize you further. You and I will likely never see eye to eye on the 7D/resolution issue (or, for that matter, the pixel density/resolution issue for any set of cameras, now or in the future). We disagree. I guess we just agree to disagree. No antagonism, seriously, just disbelief in how incorrectly you took the meaning of my post. I am sure we will disagree, until, like AlanF, you actually own/use/test your theory and find it to be as false as your compression theory.

http://www.canonrumors.com/forum/index.php?topic=12718.msg228755#msg228755
http://www.canonrumors.com/forum/index.php?topic=10289.msg186746#msg186746
http://www.canonrumors.com/forum/index.php?topic=12730.msg227863#msg227863
http://www.canonrumors.com/forum/index.php?topic=12767.msg228235#msg228235

But then you have already had that conversation. Both before, and after his 5D MkIII purchase........

http://www.canonrumors.com/forum/index.php?topic=12718.msg228755#msg228755

 

[Hillsilly]

..... Getting closer will definitely help.....

Bird photography is very popular. And with the ever increasing quality of bird photographs, it is really hard to stand out. I think your photo is fine. It is a good photo. You've clearly got the basics down pat and the gear to do quality work. But where to from here? How are you going to make an awesome bird photo?

Getting in closer is one idea. But I also like the idea of getting further away (or in my case as a focal length challenged photographer, using a smaller focal length). That way, you can see the bird in a more environmental setting. Let's see where it lives. Is it pristine wilderness? Is its habitat under threat? What's that yellow highlight on the right hand side? Is it something interesting? Is there anything special about this bird? Maybe get in close with a wider angle lens and capture a great sunset in the back? Maybe the sun streaming through branches? Maybe some mist or fog. A successful photo is one that tugs at a some emotional string. It is small things like this that turn a technically good photo into one that makes people pay attention. (Would love to show you one of mine...but I've yet to make an extraordinary bird photo myself. It is very very challenging.)

Luckily, its not just me thinking this way. A few years ago, a photo like this - http://www.nhm.ac.uk/visit-us/whats-on/temporary-exhibitions/wpy/photo.do?photo=2831&category=48&group=1 probably wouldn't have got much attention. Look how small that bird is! And its taken with an obviously unsharp 100-400mm! On a Canon crop body! On an ancient 30D no less. What the...?? Given some of the comments above, I'm surprised this bloke hasn't died of shame from owning such clearly inadequate gear. What were those judges thinking?

Anyways, have a look at how painters portray birds. Painters are interesting because the artist can pose and frame birds in any way imaginable - So they are free to choose the most aesthetically pleasing options. It is really interesting to ponder "Why did they choose to do that?". As one suggestion, John Audubon's Birds of America (http://www.lib.umich.edu/audubon-room/pictureit-rare-book-reader) is good to look at.

 

[chasinglight]

Anyways, have a look at how painters portray birds. Painters are interesting because the artist can pose and frame birds in any way imaginable - So they are free to choose the most aesthetically pleasing options. It is really interesting to ponder "Why did they choose to do that?". As one suggestion, John Audubon's Birds of America (http://www.lib.umich.edu/audubon-room/pictureit-rare-book-reader) is good to look at.

Art inspires art! (Though I I don't think of myself as an artist..more of a chronographer)

Your other advice is great too. If you think about it, if your goal is just to fill the frame with bird then your aim must be ultimate sharpness, but as you said if you step back an think about the whole scene, then you can achieve some stunning results as well. I recently photographed a great egret flying into the sunset. Unfortunately the bird is a little soft, but it still yeilded an interesting photograph. Ill try to post it when I get home.

 

[Krob78]

My replies are in red.

I'm not sure where print came into play. I think, when evaluating sharpness, the only thing that really matters is actual image dimensions. The 7D's physical APS-C frame is indeed smaller than the 5D III's by the ratio you mention, but the output image is MUCH closer to the 5D III's output image dimensions.

The 5D III full RAW image size is 5760x3840. The 7D full RAW image size is 5184x3456. In relative terms, the image sizes are red and blue below, where as your skewed sizing, based on physical sensor dimensions, greatly puts the 7D at an undue and unrealistic disadvantage (green):

I know you dislike the 7D, but I think you are making a very unfair comparison that isn't benefiting the underlying point at all. The 7D it IS capable of being razor sharp, even with the 100-400...and with that particular lens, the 7D is not capable of really reaching its full potential.

I don't think chasinglight's problem is that he is using a 7D. Far more likely than that, I think there is probably some lens/body tuning that could be done, and after that, it is largely a matter of technique. If you use a tool properly, you can maximize it's potential, and the 7D has a LOT of potential. That doesn't mean it will produce better results if you have the ability to fill the frame with both the 7D and the 5D III...the simple fact that the 5D III has more total pixels means it will produce a better image. All it means is that chasinglight can work on a few things that cost him nothing, yet allow him to produce better results.

I don't quite know where to start addressing this post.

I think the best thing would be for you to go and reread the thread.

I'm not debating the entire thread...only "Reply #38",It does amuse me, however, that you claim to limit your involvement now to a specific post in this thread yet are obviously carrying baggage from other threads that have no relevance to this one. where I believe you to make an incorrect comparison between a 7D image and a 5D III image by scaling the 7D incorrectly, which results in a significant and further reduction in detail of the 7D image relative to the 5D III image. Err, that would be completely contrary to the effects of enlargement, enlarging more makes things less sharp, my comparison is set up to be fair and favours the 7D, When it comes to what was actually resolved, magnification relative to the final output is not the sole factor. Output is the sole factor that needs or should be considered, be that an 8"x10" print, a 20"x30" print, a 30"hi res monitor or an iPad, you can only make a fair comparison if both outputs are the same size, theory and speculation is not worth anything, output is king. What is the point of comparing different sized outputs? None, it is an exercise in futility.

I am the one who has constantly said the 7D is a more than capable camera, how you have managed to infer from that "I know you dislike the 7D" , I just don't fathom and am in complete disbelief at.

You've come off at me fairly strong in a number of threads over the last couple of months where you seem to put the 7D in exceptionally bad light relative to other cameras. No, you have taken umbrage at my position, supported with my own images and hands on experience that I don't believe that it has a "tele advantage" in real world shooting over a cropped same generation ff sensor, I have never berated the 7D and always qualified that opinion that there are still very good reasons to buy one, just not because it gives you a "free" TC. You have repeatedly made a point of calling out how the 7D's resolution advantage is effectively meaningless. From my personal hands on real world testing that was the case. Given the repetition, I can only conclude you dislike the 7D. I'd suggest you reread the actual comments I made, they were always restricted to the output resolution and peoples false opinions of it based on theory, never the cameras capabilities. I also still believe you are comparing things incorrectly. That is your perogative, but just as your objections to my posts in the DOF/compression thread, you are wrong, you just don't see it yet, and putting the 7D in unfair light. If you understood what I was doing you would realise that reducing the size of the 7D output, to compare like with like, actually helps the 7D. Making things smaller makes them sharper. Even detailed articles about the 7D's resolution advantage from a well respected Ph.D., Roger Clark, seem insufficient to get you to even meet at some middle ground. What middle ground is there to meet? This isn't a negotiation! I have used and tested both formats against each other, I set up specific tests to determine the actual realisable resolution differences, I have posted my results of a test set up to maximise the advantage the 7D has and even in perfect test bench situations the differences are small, in real world shooting AF makes more difference. You have displayed no experience with the ff cameras, have run no tests, have posted no comparisons and yet continue to argue your theoretical opinion, based on... theory. Photography is a visual medium, theory is all well and good, but show me the photographs. I tested them, I made them, I printed them, my experience is not based on theory,it is based on images.

Sorry if I've misread you, but I've been going by both the intensity of your responses to me when I debate the issue with you, and the frequency at which you bring up how nearly worthless the 7D's resolution advantage supposedly is. In my experience it is, but that is a far cry and very different from calling the 7D worthless. Just because I don't respect your theory, when I have prints that prove it incorrect, doesn't mean I don't like the camera, I was debating your opinion, not the cameras worth. As for the frequency, well the more often I see incorrect theory the more often I will try to educate, DOF and equivelance, and perspective are more of my pets, I have been linking to the Equivelance page for 18 months now and there is a definite turn of common knowledge that now gets the concept, I don't claim many converts, but a few.

Further "The 7D it IS capable of being razor sharp, even with the 100-400.." I KNOW, and I posted 11 links of razor sharp images shot with a 7D and 100-400, of birds, at 400, and some wide open!!!!! DID YOU NOT SEE THEM?

No, I came in on page three, Well there's the thing, take something out of context and you don't know what people are talking about. I guess...saw a post by you that put the 7D in unfairly poor light, and responded. Again, I wasn't responding to the entire thread, and having read through the rest, I'm a bit confused by what seems to be conflicting information in different posts. I suspect that is your preconceived discrimination of what you expected my opinion to be, not any inconsistencies in my posting.

Or this "I don't think chasinglight's problem is that he is using a 7D" I have repeatedly said his main problem is subject distance (magnification) NOT THE 7D. Funnily enough that is why I used MAGNIFICATION as the comparison between his and Alan's images.

Using magnification alone, and ignoring pixel sizes completely, is the issue I was debating. I think that is oversimplifying the issue. There are effectively an unlimited number of potential output sizes, Yes, and the only fair comparison is to equalise them, if I want an 8" x12" I want an 8" x12", I don't want an 8" x 12" from the 5D MkIII and a 12" x 18" from the 7D, which is what you are saying is a fair comparison. both smaller and larger than the native image size for each camera. Digital cameras don't have a native image size! They have a native number of pixels, but that is very different. I think it is only fair to compare them at their native sizes. If you were to standardise ppi as a comparison you would be going against any common sense idea of fair, or comparison, you would also put the 7D at considerable disadvantage as you would be enlargeing it around 50% more. Assuming the same bird is photographed at the same distance with both cameras, there is both a magnification difference Not in focal length limited situations, like this one, the magnification onto the sensor is the same. as well as a pixels on subject difference. Since the issue is sharpness, pixels on subject matter. This is effectively a pixel peeping situation, so considerably reducing the size of the 7D's native output is skewing the basis for comparison. No it isn't, it is favouring it, I don't understand why you struggle with this concept. More pixels equals sharper, less enlargement equals sharper, my fair comparison uses more pixels from the 7D and enlarges it less than yours, that puts it at an ADVANTAGE.

I get the feeling we are just going to agree to disagree on this point...so I'm happy to end the debate here. And yet you didn't.......

Alan suggested an $8,000 investment, I said no, GET CLOSER. I know in focal length limited situations the 7D will outperform (by a very small margin) the cropped 21MP 5D MkII/1Ds MkIII, and while I have not tested it against the 5D MkIII I would expect the results to be similar. To summarize that last sentence, I know in focal length limited situations the 7D will perform better than a cropped FF image, though not by much and certainly not as many as the pixel numbers would have you believe. I also KNOW, ALL PIXELS ARE NOT EQUAL, to compare one 7D pixel to one 5D MkIII pixel is moronic, it just doesn't work like that.

I agree, one needs to get closer. But in practice, there are limitations in doing that. Birds infrequently just sit still and let you get as close as you want. There IS an argument for buying better gear. It may not be the best advice for someone in chasinglights position, but it is not an invalid argument either. There are also consequences of getting closer. There is a zone within which one is closer to the bird, but also affecting it's behavior. If all you have is a 400mm lens, then either you deal with the behavioral changes, wait long enough for those behavioral changes to subside (which may never occur), slap the lens on a sensor with denser pixels there you go again with your theory, or get a longer lens. Again, if you had taken a few moments to acquaint yourself with the thread you would have realised we were talking about a very specific situation here. Alan has a FF body and a 300 lens with a 2XTC, chasinglight has a 1.6 crop body with a 100-400 lens, they are both focal length limited, they both need to get closer, but obviously, not to the detriment of the bird or the image. Having said that chasinglights setup actually gives him a narrower fov than Alan (though less magnification), hence my (everybody's) consistent suggestion that that was where the main improvement would come.

We disagree about the "by a very small margin" part. That is fine, when I see your comparison images I might respect your opinion more, having said that I am not the only person who has come to the conclusion I have, several other prominent posters, have said exactly the same thing, but then they actually had experience with the cameras, they were not relying on theory. All sensor pixels may not be equal, but that doesn't really matter. In my fair comparison, not it doesn't matter, in your ridiculous and contrived comparison it is supremely important, in fact that is all you are comparing. One only has to look at the two photos at full size to compare sharpness. No, one has to look at two photos THE SAME SIZE to compare sharpness. I fear you are so anti me you will refuse to see that, just like the "compression" you were convinced was true, it wasn't, and your comparison isn't If one reduces the resolution of one image from its original size, THAT makes for an unfair comparison.You don't sound like you print much, though you claim you do, there is no reduction in resolution, there is an increase in resolution for the denser sensor and an increase in ppi figures to take that difference into account, simple really, nothing is lost. And, again, the denser sensor is at an advantage. You are converting your bias against the 7D, I don't have one, I have an opinion backed up with images and hands on experience that is agreed with by other owners of the two that contradicts common wisdom, your theory, that the much higher pixel density of the 7D gives it a meaningful resolution advantage in real world shooting over a same generation ff sensor cropped to the same sensor area. the notion that its pixels only give it a "very small margin" benefit, directly into actual image comparisons. If you are going to scale for comparison, scale them both to the same size YES, THAT IS THE POINT, THEY MUST BE SCALED TO THE SAME SIZE, not the same %, don't scale one and leave the other unscaled.

In focal length limited situations, as this one is, per pixel comparisons are worse than useless, subject magnification is THE ONLY FAIR comparison, it is the only relevant metric for comparison of output. It maintains equivalency, you are comparing like with like. How do you adjust for different DOF figures if you compare on a per pixel basis? How do you adjust a sharpness figure when you magnify one image more than the other? You can't! Compare same magnification and you get a true comparison, you know what will look better in print, on a screen, or any other end use output.

Again, I disagree. As a matter of detail resolved, for example feather barbs, pixel peeping WILL show a meaningful difference. That difference may or may not matter for the artists final choice of presentation. If the image is shrunk and uploaded online, and that is the only thing ever done with it, then sure...per-pixel comparisons are effectively meaningless. On the other hand, if you scale up a heavily cropped 5D III image against a lightly cropped 7D image, and print at 16x24, then the 7D's resolution benefit IS meaningful. Oh, you almost got it then! A tightly cropped 5D MkIII image and a lightly cropped 7D image WILL REPRESENT THE SAME SENSOR AREA! We strongly disagree on how meaningful the denser pixel advantage is, but then my opinion is based on actual hands on experience and I draw my own conclusions from my own tests, but you nearly stumbled into the whole point there without looking. Per-pixel comparisons...or rather more realistically comparing both cropped images at full size, is an entirely valid way to evaluate IQ. But then you lost it again :-( comparing smaller to bigger is not a fair comparison.

I am not even considering anything like a scientific test here, either...it doesn't take much to see improved sharpness with a little bit of eyeballing. I'm not talking noise here, either...just sharpness. The 7D image may indeed be considerably noisier in smooth OOF backgrounds, however even a noisy 7D image can still be razor sharp in the detail areas where it matters. Noise is also considerably harder to see in a print, so the greater resolving power of the 7D will have a meaningful impact on the sharpness of the final print. The 5D III noise would likely clean up considerably better, but the improvement is going to be less visible in a print than the sharpness factor. All theory and conjecture, so irrelevant. However you have posted images before that show heavy noise reduction in oof areas, but none in detailed areas, you seems to think noisy detailed areas are sharper because they have noise, they are not, they just appear sharper, put some noise in a very slightly soft image and it appears sharper too.

As for DOF, from a technical standpoint, you are correct. From a practical standpoint, it usually doesn't matter. No, it doesn't matter in same magnification enlargements, but look at stuff 50% more enlarged, or at the pixel level, and your CoC figures go out the window. Unless you have a DOF issue such that a critical part of your subject is affected...for example the end of a birds bill ends up out of focus...that matters. If the important parts of your subject are in focus, At what magnification? My same sized magnification, or your 50% more for the 7D? DoF and sharpness changes dramatically with enlargement, my comparison equalises that, yours does not. at least in bird photography, DOF usually isn't a make or break issue. More of the butt end of a bird might be more blurry with one camera over the other,No, in focal length limited situations, like this one, the DoF is the same for both cameras IF you use the same output magnification (if the birds in print are the same size), again, my comparison allows for this, yours does not. but if the eyes, bill, and a sufficient part of the birds head or body are sharp and clear...that's what really matters.

Anyway, I don't want to antagonize you further. You and I will likely never see eye to eye on the 7D/resolution issue (or, for that matter, the pixel density/resolution issue for any set of cameras, now or in the future). We disagree. I guess we just agree to disagree. No antagonism, seriously, just disbelief in how incorrectly you took the meaning of my post. I am sure we will disagree, until, like AlanF, you actually own/use/test your theory and find it to be as false as your compression theory.

http://www.canonrumors.com/forum/index.php?topic=12718.msg228755#msg228755
http://www.canonrumors.com/forum/index.php?topic=10289.msg186746#msg186746
http://www.canonrumors.com/forum/index.php?topic=12730.msg227863#msg227863
http://www.canonrumors.com/forum/index.php?topic=12767.msg228235#msg228235

But then you have already had that conversation. Both before, and after his 5D MkIII purchase........

http://www.canonrumors.com/forum/index.php?topic=12718.msg228755#msg228755

Sigh... :

 

[Krob78]

100-400mm is really irritating on the 7d too... (sarc., implied )
1/320 sec.
f/5.6
ISO 100

I downloaded the photo to check the exif data. It is at f = 190mm, the sweet spot of the lens, not the full 400mm, the weakest length. The lens is very sharp at ~200mm and f/5.6-8. I have had some great shots under those conditions, as well as 400mm when I could fill much of the frame.

You must have been very close to the owl to get it to fill so much of the frame. And, that is the way to get the most from any lens. Here is the lens with 100% crop on 7D at 100mm of terns feeding (is0 400, f/5, 1/1600 s).

Sorry about that Alan, you're right! I apparently didn't check the focal length of the owl shot before uploading, I have plenty of 400mm examples that are of excellent sharpness to choose from as well. The image of the Cardinal above the Owl is at 400mm, with the same lens...

When I first got my 100-400mm several years ago, I wasn't totally thrilled with the sharpness. After spending hours tweaking the MFA for that lens, I was blown away with excellent sharpness at both ends of the lens as well as the sweet spot around 200mm. Since then, the first thing I do with a new lens is MFA... Kind of a pita as it were, but totally worth it every time!

Thanks for pointing out the focal length on that image, as we were discussing maximum focal length for the lens, not just overall quality and/or sharpness.. Additionally, I may point out that it was handheld as well. Yes, relatively close as the Owl flew and landed nearby me. Also, this is a cropped image. I was probably in the 75'- 85' range from the subject and eventually worked my way quite a bit closer...

All the best,
Ken

 

[jrista]

My replies are in red.

I'm sorry, but I can't read all of that. The use of dark red text just makes separating what you are trying to say from what I said difficult, and not worth the effort.

On cursory read, you seem to be taking things I've said particularly personally. They aren't meant to be a personal attack, simply a dispute with your approach. I've already stated my opinion, and I no longer wish to disrupt the thread.

 

[jrista]

I'd say that is a good practice. If you are anything like me, you will know when your gear is holding you back. I also have the 100-400. I think the 7D is a fine camera, produces great IQ in most circumstances (which for my bird photography is usually in good to evening light, ISO 200 - 1600), and has great features that support bird photography. The 100-400, when properly tuned with AFMA, produces acceptably sharp images most of the time. It should be noted that at 400mm, f/7.1 tends to be the sharpest, while f/5.6 will be visibly soft. Before getting my new lens, I shot at f/7.1 almost exclusively, sometimes stopping down to f/8 and rarely opening up to f/6.3.

I would tune your lens for your copy of the 7D, and start shooting at f/7.1. You should see individual barbs of each feather (a feather is a central shaft, on either side of which is a vane of barbes, which are interconnected via barbules off each barb...you will RARELY see barbules in a photo, but in an acceptably sharp photo, you should see barbs.) There are three things that will soften the barbs of a birds feathers...distance too great, missfocus, bird motion or camera shake. Distance is usually the biggest problem early on. Depending on the type of bird, either learning the right behavior to exhibit that gets you close, or camouflaging yourself to hide in plain side, are was of solving that problem.

Thanks for the advice (all of it, not just what I quoted). I have seen my copy of the 100-400 produce much sharper images than this (such as the one below; AFMA 0). I think actually took this shot of the eastern screech owl at f/7.1. I think I actually started to notice more inconsistency after I used Focal to AFMA the lens a few months ago; this produced a -3. I performed Focal a few times today getting -3, 3, and 1. So I decided to go back to 0 and see how that works out.... could be that I was just trying to over sharpen the knife and instead made it dull...

Are you using the older version of FoCal, or the newer version released a month ago (from the white, rather than black, site)? I had problems with the older FoCal...it definitely was inconsistent, which is why I tried the AF confirmation dot technique (which seemed to be fairly consistent, just wrong.) I've found that the newer FoCal, which supposedly has some rewritten core code, seems to be much more consistent. When I ran it on my 7D+600mm, it consistently gives me a +1 for a 30 foot distance, and 0 for a 60 foot distance. I've repeated the tests multiple times, and I get the same AFMA each time for those distances (which usually covers the range of distances my subjects tend to be at, with the exception of songbirds...but closer than 30 feet the lens always seems to resolve more than enough detail.)

If you haven't updated to the most recent version of FoCal, I would give it a whirl, see if you get more consistent results. Also, remember to use good light...test chart outside in direct sunlight is usually best.