Canon 70-300 USM IS extremely unsharp

Discussion in 'Canon EOS' started by grzegorz_szczotka, Apr 18, 2010.

  1. No much information here so I'll take a stab in the dark. The subject is so small it is unlikely the camera (auto AF selection) had the smarts to know what you wanted in focus. You need to take control of focus and select what you want to be sharp (manually select the AF point and place it on the bird). If you did place the active AF point on the bird, then your lens is suffering from front focus and needs calibration at Canon Service. No biggie. They'll fix it in 10 days or so including shipping and you'll be in like friggen flint.
  2. Of course AF point was in dead center of the bird
  3. I also think you might have experienced a focus issue, and not a lens failure. I never had a sharpness issue with my old 70-300 lens. This shot was made with the DO version of the 70-300, but it has always had about the same sharpness as my previous 70-300, and the size of this bird serves to make the point about the camera having a sure target area to focus on. Barring that, you might need to have your camera and 70-300 tweaked by the folks at Canon service to nail correct focus spot on.
  4. So I should send them both?
    This focus is the camera problem?
    Another thing is that the service is in another city and it can take long time.
    And I would be without my lovely camera...
  5. Maybe it's just because the AF accuracy is not as great as advertised? I do a lot of critical focusing and I notice that the prosumer *0D dSLRs are much worse than the 5D, which in turn is probably not as good as the 1D cameras. Also keep in mind that the AF points in reality are three times as large as the AF selectors in the viewfinder, so if there is focus archieved anywhere in this area, the camera signals you that AF is spot-on.
    1. Compare AF to MF.
    2. If you have a camera with MA, check that as well.
    Happy shooting,
  6. As the old maxim goes," if your subject isn't big enough, you aren't close enough." Without knowing your shooting technique, it's unfair to blame the lens. That is one small pheasant in a large area. Try using a tripod and focusing on something larger that you can test the lens on. Try shooting some larger birds under good light - A frame-filling swan or a canada goose ought to have plenty of detail in sharp focus from 10- 20 feet away with such a lens.
  7. In your first image the focus point is clearly a couple of feet in front of the bird where the grass borders the dirt. The depth of field appears to be a foot or less.
    You are too far away for it to be very good even in sharp focus. Try to get closer or think about a longer lens if wildlife photography is of high interest to you.
  8. I agree with Mark O'Brien and most of what else is written here....
    What I see here is not a reason to suspect the lens and start shipping things off to Canon to be checked. I think some easy tests with frame-filling subjects on a tripod at 300mm would be the next logical step.
    While I agree (and Canon plainly states) that the auto-focus point coverage is actually larger than the black squares in your viewfinder, I'm not sure where Bueh got the information that "...the AF points in reality are three times as large as the AF selectors in the viewfinder". I don't think that's entirely correct...
    Just try some 300mm tripod testing on a subject that fills the whole frame and see what happens.
  9. I'm not sure where Bueh got the information that "...the AF points in reality are three times as large as the AF selectors in the viewfinder". I don't think that's entirely correct...​
    This is from a high-ranking Canon columnist... I can't find it right now... it's been a while. The actual focus sensor areas of the 20D(?) were shown, which are in fact three times the squares in the viewfinder -- or, in case of the center spot, 5x as big (it's a big cross).
    Found something: These are similar to the diagram I remember. You see the 50D's AF sensors are all five times as large as the indicated point. Now all these focus errors make more sense, eh?
  10. Greg,
    I had a look at the first picture and noticed that you are shooting @ 1250 iso. I am presuming that you are shooting when the light is still low. Focus of the camera may not be optimal at that light level and that maybe why it has focused just short of the bird. Again iso 1250. Noise suppression and subsequent loss of detail is starting to show at 100%viewing. And again 100% viewing. Looking too close. The birds body only takes up about 450*460 pixels in this image. Not much detail in the bird can be captured at that size. Which brings us to,, you are quite far away from the bird for any meaningful shot. You have to get much closer to get that detail that you are looking for.
    Personally I would not take the camera out for this distance with exception to just recording the scene, or taking a landscape shot with the bird(s) somewhere in it. Stick with the lens and learn how to use it to make better images. I am sure you will be happy with it once your get your technique down. I would leave canon service out if it till you have verified that the focus is truly off.

    My two cents…
  11. It's my understanding that the red box is only an approximation of the focus point. A hair off one way or another and you don't get what you thought you were going to get. That's a good reason to work with a smaller aperture for a bit of DOF. I know the 70-300 IS can get sharp pictures, at least my previous version did--but I also got within 12 feet of the bird.
  12. Yes, you need to get much closer. The 300 f4L is sharper than the 70-300 IS so will allow more cropping and take an extender too. But mostly you should think about how to get closer. In this case it was just a matter of feeding them till they were close enough.
  13. Big thanks for all advice! It's very helpful.
    Two last photos are taken with tripod and the objects are not so much far away (about 30-40 feet).
    I cannot understand this lack of details. I understand this lens has it limits. But please look again at the two last photos, is it really the maximums sharpness for this lens?
  14. It's hard to compare the last 2 files with the first file. The ISO is way lower than the 1250 you were using with the bird. I too think the high ISO use could have been the culprit. As a min the lens does front focus a tad, nothing to get upset about. Do some testing on the tripos with the IS OFF at different apertures. Make sure your not too close to the target. I bet at ISO setting 100 at F-11 on a tripod you will get sharp pictures. The alternative is to move to a pro lens like the 70-200 2.8! v/r Buffdr
  15. No need to compare.
    The two last photos are taken with tripod aperture 8 or 11 and low iso. Pictures are NOT sharp
  16. OPK


    I've made similar statement these days with 50D body. no matter what lens is on camera - there's a big amount of misfocused pictures. only way to get in focus is switch each lens to manual and ignore AF points confirmation just rely on your eye....
  17. Martin, if it's the case I would have to replace it for something that works!
  18. Looking at the shot of the grass doesn’t seem to show any problem or at least very hard to find. The curtain shot is a bit better to analyze but I have a question. did you use mirror lockup and 2 second delay or a cable release. I ask because the shutter speed is .8 of a second. a candidate for vibration. maye a shot at a higher speed and use flash?
  19. yes, I used both
  20. You're shooting with a 50D, you should be able to use AF Microadjustment to fix apparent front focussing.
    Enable in C.FnIII:Autofocus/Drive / AF Microadjustment. Choose 2:Adjust by lens. Press info button with lens on to modify.
  21. Should I move it to left or to right?
  22. My understanding is that your lens is currently front focussing, so I would try shifting this lens back so to the right some. I've never had to do this myself so have no first hand experience. Take a picture on tripod with your current setup, change the AF Microadjustment to the right and take another picture to see if your making it better or worst.
    You can register up to 20 lens with this AF Microadjustment on the 50D. Let me know how you fair.
  23. It does makes a difference!
    I'm uploading all photos.
    It appears the photo is sharpest at from 0 to 5.
    My eye tells me it's from 1 to 3 but it's really hard to tell the difference.
  25. Hi Grzegorz,
    The focus chart shot is easier to help in determining focus accuracy. The sample shot you uploaded with no AF adjustment shows that the bottom 2mm and 4mm lines are in better focus then the center line and that therefore your lens is front focusing. The curtain shots are not as easy to judge good focus.
    Also, try to get more light on the chart so that other factors such as low shutter speed, high iso or noise does not effect outcome in order to better judge focusing accuracy and making better adjustment.
    IS to off on sturdy tripod should also help.
    Taken with better light, iso 400, sturdy tripod, IS off
  27. And even better:
  28. The best one ;)
    Daylight (1/1250, iso 200)
  29. Grzegorz
    I hope you get acceptable results. I was extremely dissapointed with my 70-300 IS and eventually sold it. It seems sharp when you shoot something closeby in good light (i.e. newspaper on a wall), but when the subject is far away and light conditions are not ideal, it is impossible to get a photo that is up to standard.
    My lens was not back- or frontfocusing so I could not correct it with microadjustment. I tested the lens against my 100-400, my 70-200 f2.8 and 70-200 f4 (the last two with 1.4X converter) under the specific conditions and the 70-300 was not close to the other lenses.
  30. I can't see the photos you posted -- the links don't work due to too much traffic.
    That said, did you turn IS *OFF* before taking the shot? Since you're on a tripod, IS will actually make the pictures blurry. I learned that the hard way.... :)
  31. Photos are here:
    Yes, the IS is off
  32. Raw files aren't supposed to be sharp, but you already know that, right?
    I downloaded raw file 8845, it looks about like any other raw file that comes from my 40d or 5d or 1d2 prior to post processing/sharpening.
    It's like the others have said about the focus points, it's not rocket science. I've missed some shots using AF with the above cameras that I was sure I nailed, but didn't. If it's really critical use MF. I use it about half the time with my 300 f4 is lens. And set the CF to focus with the thumb.
    Maybe you're expecting too much out of your equipment - the raw file i downloaded was shot with a continuous light, the shawl with the plant. I never would use autofocus for a still life like that, but bottom line is it looks to be in focus, and needs sharpening applied for intended purpose.
  33. W T, a quick introduction to sharpening raws using DPP? Just using sliders "sharpness" ?
    I am beginner ;) Thanks
  34. The basic RAW page on the floating DPP toolbar allows you to adjust sharpness from the standard setting selected for each picture style. I would suggest that you look at a 100% view of an area in the file while you tweak sharpness. It's difficult to really judge the changes at less than 80% or so. Remember, you don't modify the RAW image at all, so you can't kill it if you go overboard with tweaks. Just back them off, and start again.
  35. Grzegorz, search here or in the "digital darkroom" forum. post processing & sharpening are like opinions, everyone has one and they're all different.
    Canon website is a good source of info, and there are video training links for dpp online.
  36. So, to sum it up. Does this lens have front focus or not?

Share This Page