Canon 50D focus/sharpness testing
As a preface, I had been noticing that my keeper rate with the Sigma 30mm had gone down a bit since the firmware update. Yesterday, I had some time to myself and decided to redo the microAF adjustment for my lenses. Went through variations of the procedure using Scott's tree suggestion and the Northern Lights LCD one. Even though I was able to decide on setting for each lens, I was left feeling that the resolution/focus wasn't as great as I expected. All shots that day were done on a tripod with mirror lock.
Today, I decided to do some real-world testing. I taped my business card at the height that I normally shoot my kids and shot a sequence with the Canon 50D and my old Canon 350D using the Sigma 30mm initially and then the 50mm/1.8 mkI. Those are my fastest lenses thus prompting the choices. Below is a link to a gallery with 100% crops and captions detailing the conditions (camera, lens, aperture and AF point used). Each camera seemed to have problems resolving the details and I am just wondering if the crops are 'normal' for this kind of testing. Today's tests were done handheld but at shutter speeds of >1/500 (again I was aiming for a real world test).
As a side note - the Sigma has previously been adjusted at the Sigma repair center and has 0 microAF adjustment. The Canon is dialed in at +5. I focused at infinity between each shot.
Greatly appreciate your thoughts!
E
The setup:
The link:
http://eoren1.smugmug.com/gallery/8551245_Y4Afp/3/563244777_dhjag#563214965_UDiJj
Today, I decided to do some real-world testing. I taped my business card at the height that I normally shoot my kids and shot a sequence with the Canon 50D and my old Canon 350D using the Sigma 30mm initially and then the 50mm/1.8 mkI. Those are my fastest lenses thus prompting the choices. Below is a link to a gallery with 100% crops and captions detailing the conditions (camera, lens, aperture and AF point used). Each camera seemed to have problems resolving the details and I am just wondering if the crops are 'normal' for this kind of testing. Today's tests were done handheld but at shutter speeds of >1/500 (again I was aiming for a real world test).
As a side note - the Sigma has previously been adjusted at the Sigma repair center and has 0 microAF adjustment. The Canon is dialed in at +5. I focused at infinity between each shot.
Greatly appreciate your thoughts!
E
The setup:
The link:
http://eoren1.smugmug.com/gallery/8551245_Y4Afp/3/563244777_dhjag#563214965_UDiJj
Eyal
My site | Non-MHD Landscapes |Google+ | Twitter | Facebook | Smugmug photos
My site | Non-MHD Landscapes |Google+ | Twitter | Facebook | Smugmug photos
0
Comments
On the 50/1.8, the 50D did pretty well wide open but hit only 1 of 3 at f/3.2. The 350D hit good focus on 2 of 3 at f/3.2.
E
My site | Non-MHD Landscapes |Google+ | Twitter | Facebook | Smugmug photos
Alternatively, if you are using a focus point other than the center, moving the camera to get that focus point over the business card, initiating focus, then re-composing .... there again, it stands to reason that the business card won't be in focus.
To test the accuracy of the focus points and/or the lens and camera combination, put the camera on a tripod, select the focus point of interest, use that focus point, and then supply a 100% crop of the area of the image under that focus point as that area is the only part of the image that has a chance of being in focus (unless the entire scene in the viewfinder is flat and parallel with respect to the plane of the imager).
So, did I miss something in what you were trying to say about your testing methods - that's as likely as not
My Photos
Thoughts on photographing a wedding, How to post a picture, AF Microadjustments?, Light Scoop
Equipment List - Check my profile
Thanks for taking the time to look at the post - sorry that I left you confused
The focal point was always the business card. I shot it with either the center focus point, a peripheral focus point (such as one I would use to focus on the eyes of a 3 year old) or focused using the center AF point and recomposed so that the business card lined back up under the peripher AF point used previously. I then opened the images in ACR, zeroed out the sharpness settings and took a 100% crop around the business card that should have been in focus. Tried wide open and stopped down a bit with the Sigma 30 and original Canon 50/1.8. I had used a tripod and mirror lockup the previous day in doing micro adjustments. Today's exercise was mostly about 'real world' shooting to assess the focusing abilities. I'll note that I was seeing similar focus 'errors' yesterday in more ideal conditions (shot in brighter light - iso 100 f/1.4 1/3200 speeds).
The fundamental question I have is - am I expecting too much of these lenses/cameras? Should a business card that fills up the frame seen in the 'setup' shot be in sharp enough focus to be legible? Also, are the errors in focus seen on the 100% crops 'normal' or 'within tolerance' for this camera?
Thanks again and I hope I was able to clear that up.
E
PS - I am not normally a 'pixel peeper'. I just started seeing this in doing the micro adjustment and began to wonder if it had real-world implications as my shots have not been as sharp or at least the keeper ratio seemed to have dropped after that last firmware update.
My site | Non-MHD Landscapes |Google+ | Twitter | Facebook | Smugmug photos
The Sigma 30 f/1.4 is not a super performer at the edges though some copies are better than others, especially when fairly wide open.
It might be of more benefit for you to use a given focal point at progressively smaller apertures and see if the sharpness improves as you stop down the lens. In this manner, you have a better chance of seeing if the lens will meet you needs/expectations.
Finally, get some that are sharp in the center and get them printed - look at the edges of the prints. This is the real test - are the prints good enough for you?
My Photos
Thoughts on photographing a wedding, How to post a picture, AF Microadjustments?, Light Scoop
Equipment List - Check my profile
I did try a bit of stopping down to see if there was a difference.
Here's the 30mm on the 50D at f/1.4 using the center AF point:
And the same setup at f/2.0
Also tried it with the Canon 50mm f/1.8 mkI
And stopped down to f/3.2
Interestingly, here is the 50mm using a peripheral AF point
I find it strange that the image is sharper using a peripheral AF point. Also not sure if I should be seeing a sharper photo at f/3.2. Here is the setup for the 50mm shots
Thanks for the help/advice
E
My site | Non-MHD Landscapes |Google+ | Twitter | Facebook | Smugmug photos
My Photos
Thoughts on photographing a wedding, How to post a picture, AF Microadjustments?, Light Scoop
Equipment List - Check my profile
I wanted to provide a quick update to this thread. On Saturday, I went on a Canon-sponsored photo safari where we got to try out bodies and lenses. I really wanted to try the 17-55 IS but they didn't have it. Instead, I got the 70-200 f/4 IS which is arguably one of Canon's best focusing and sharpest lenses. At the end of the walk, I planted a tripod with my 50D body and one from Canon and the same 70-200 lens. Aimed at a tree with my business card taped on. Shot at f/4 with AF, with live-view AF on both bodies. Took the shots back home, reduced sharpness to 0 in ACR and zoomed in at 100%. The results from both cameras were identical and amazing. In fact, my shots from that day are pretty damn nice as well...
Looking back at the lenses that I own, the Canon 50/1.8, 70-210/3.5-4.5 and Canon 18-55 IS are all basically <$100 lenses. The Sigma was more but is a non-Canon brand thus has to make its AF compatible in other means. Then 10-22 has been my best lens and held up to the micro adjustment well (at zero adjustment). So the moral of the story is that lenses DO make a big difference. I've resolved to save up for the 17-55/2.8 IS and 70-200 f/4 IS next.
E
My site | Non-MHD Landscapes |Google+ | Twitter | Facebook | Smugmug photos
I'm glad you got it sorted out. The 2 lenses you mention plus your Canon EF-S 10-22mm, f3.5-f4.5 USM should make a "killer" combination.
I can also recommend the Canon EF 1.4x II teleconverter with the EF 70-200mm, f4L IS USM. Not the fastest combination in terms of effective aperture, it is still viable in good light and gives you (IMO) better results at the long end than most of the 70-300mm zooms, including the fairly expensive 70-300mm DO.
Moderator of the Cameras and Accessories forums
If by 'killer' you mean that the wife will kill me when she hears that I need $2k in gear (after spending $1600 on the 50D and 10-22) then, yes, you are absolutely correct
My site | Non-MHD Landscapes |Google+ | Twitter | Facebook | Smugmug photos
I'm sorry, what I really meant is that you need the Canon EF 800mm, f5.6L IS USM. Show your wife the sticker on that first and the other 2 lenses won't seem sooo bad.
Just tell your wife it's all my fault.
Moderator of the Cameras and Accessories forums
So after a few more attempts at focus adjustment, I gave up and sent the 50D to Canon in Irvine. Just got it back today (very quick turnaround - 6 days!) and saw this note:
Service Details:
We have examined the product according to your request and, upon close inspection the exact cause could not be identified but it was found that the focus did not operate properly. Electrical adjustments were carried out on the Main pcb assembly. Other inspection and cleaning and parts replacements were carried out.
Haven't had a chance to test it out formally but am looking forward to seeing if focus has become more consistent now. Glad I sent it in
E
My site | Non-MHD Landscapes |Google+ | Twitter | Facebook | Smugmug photos
I just had my 40D serviced in Irvine also. They did a good job and the turnaround was ok. (Just over 10 days if I remember right)
BUT...
I also found their explanation of failure and repair a little lacking in detail. they also used the prase "Other inspection and cleaning and parts replacements were carried out." in my explanaion of failure.
.