Go Back   Steve's Digicams Forums > Digital SLR and Interchangeable Lens Cameras > Canon EOS dSLR

Reply
 
Thread Tools Search this Thread
Old Oct 30, 2003, 1:10 PM   #11
Senior Member
 
Join Date: Dec 2002
Posts: 5,803
Default

Quote:
Originally Posted by Alex Campuzano
I'm no pro, but I do know that increasing the zoom DECREASES your aperature. Larger aperatures generate sharper images. Similar to comparing a 35mm shot (small aperature) and a 4x5 chrome (large aperature). There is no comparison in sharpness.
If I am misunderstanding you, then please correct me.
Your logic is flawed. Breaking down your argument you said:
4x5 chrome has a larger aperture than 35mm (I can believe that.)
4x5 chrome is sharper than 35mm (could be true, I don't know.)

therefor:
a larger aperture generates sharper images.

That logic is false. You can't make that conclusion because 4x5 chrome uses different lenses than 35mm, not to mention different image capturing materials.

Quality glass stopped down (i.e. using smaller than max aperture) generates the sharpest picture (without CA and other problems) and then there is a point where you stop down too much and get distortion in the light. Technically, this isn't true for cameras with a really small sensor and physical really small lenses (consumer grade digicam) but in this forum we'll stick to DSLRs.

Take a look at the lens MTF charts at www.canoneos.com. For example, this very, very good 35mm lens, the 300mm f2.8L:
http://www.usa.canon.com/eflenses/le...300_28mtf.html

The light blue lines are wide open, the black lines are at f8. It resolves more lines across more of the lens at f8 than f2.8. Resolving more detail means (potentially) sharper picture. When you stop down, you use less of the edges of the lens. The center of the lens is the highest quality portion of the lens for the vast majority of 35mm lenses (Canon's Macro prime lenses break this general rule.)

Eric
eric s is offline   Reply With Quote
Old Oct 30, 2003, 1:32 PM   #12
Member
 
Alex Campuzano's Avatar
 
Join Date: Jun 2003
Posts: 42
Default More About Sharpness

Eric:

After reading my post again, I confused myself. I was using a metaphor with 35mm film representing a smaller aperature and the 4x5 the opposite. 35 will usually be grainier than a larger chunk of film.

The same image taken with a 35mm vs. 4x5 chrome using "similar" lenses will always show more detail in the chrome. It's just a matter of physical size. There's just more "material" to work with. Print an 8x10 of each.

Similar, my 1.2 megapixel Olympus point and shoot takes fantanstic images, but the same shot with my 10D will always have more to offer. More pixels is like more "material." Assuming a fair test, you're starting out with more information out of the shoot will get smoother details.

You are right about f8 though. I figured that out by accident and a pro buddy of mine verified it. I took a room shot at f22/AV, which came out fantastic, and assumed it best for a product shot. Wrong. Reshot the product at f8 and it jumped off the screen.

I appreciate you correcting/educating me in the process.

Alex.
Alex Campuzano is offline   Reply With Quote
Old Oct 30, 2003, 3:41 PM   #13
NHL
Senior Member
 
NHL's Avatar
 
Join Date: Jun 2002
Location: 39.18776, -77.311353333333
Posts: 11,547
Default

... there's also the DOF, ie more DOF at f/22 than at f/8

... similarly to the picture that was posted... we don't know the aperture value, and it looks like it was shot wide open as well. Hence some area look to be focused (ie the left wall) and some areas were out of focus (or/and moving)!
NHL is offline   Reply With Quote
Old Nov 1, 2003, 12:32 PM   #14
Junior Member
 
Join Date: Oct 2003
Posts: 11
Default Here's An Example of Soft Focus for EF 24-85mm

Looks to me like

1. slow shutter speed (leg moved)
2. Open aperture (Low DOF) due to lighting
3. Focussed forward of boy, wood to left and front of telephone and shoes are more in focus than boys face.


Assuming you chose the focus point on boys face it seems similar to some results on my first 10D, problems focussing, trying another.
Dylan is offline   Reply With Quote
Old Nov 1, 2003, 10:53 PM   #15
NHL
Senior Member
 
NHL's Avatar
 
Join Date: Jun 2002
Location: 39.18776, -77.311353333333
Posts: 11,547
Default

Quote:
Assuming you chose the focus point on boys face it seems similar to some results on my first 10D, problems focussing, trying another.
...Could also be problem with some Canon zooms (especially the 28-135 IS USM): If you focus first and recompose by zooming in or out -> the focus point change and the lens manual cautioned against this practice
NHL is offline   Reply With Quote
Old Nov 2, 2003, 4:29 AM   #16
Junior Member
 
Join Date: Oct 2003
Posts: 11
Default

I would agree with not zooming after focussing. I did some tests using a 85mm f1.8, to reduce DOF, when I suspected the focussing and in several shots the camera appeared to focus forward. I also tried testing using a chart suggested in another group, using scale45.jpg, and that worked perfectly but only when I had modified to target line to assist the AF otherwise it focussed very slightly forward. Maybe the AF doesn't work too well in low contrast. Results at www.knighttrain.freeserve.co.uk

If AF doesn't work too well in some circumstances then perhaps they should allow an option to fit a split screen and/or microprism to allow more accurate manual focus, as on their old SLRs. :idea:
Dylan is offline   Reply With Quote
Old Nov 2, 2003, 8:14 AM   #17
NHL
Senior Member
 
NHL's Avatar
 
Join Date: Jun 2002
Location: 39.18776, -77.311353333333
Posts: 11,547
Default

That was an execellent idea to use a fixed focal lenght to check for focus...

However with your posted results I can't help but noticed the focusing area is always to the right of the AF points ie not necessary front focus but the AF points seem to be not aligned with the icons in the viewfinder! It'll be interesting to see similar test with the camera rotated 90 degree (or flipped entirely upside down)...

I did notice for example on my camera that the center focus point straddle the areas outside the rectangle icon... ie if I put the bird in your last example over any of the square side lines instead of centered inside the AF box then my camera is much more accurate in auto focusing!


-> edited to say right instead of left (must be waking up the wrong side of bed!) :lol:
NHL is offline   Reply With Quote
Old Nov 2, 2003, 8:17 AM   #18
Senior Member
 
Join Date: Dec 2002
Posts: 5,803
Default

Personally, I don't see any problems with focus in those pictures. Its a bit hard to tell because the pictures are of things and not test chars.... A few oc those pictures looks a bit soft, but they don't look missfocused to me.

If you really want to do this testing, have you looked at:
http://www.canondslr.com/articles/m-dream/

It was, I believe, read over by a Canon USA Tech support/doc person and they said that it's basically how Canon tests the cameras.

But on to the pictures. I assume the red box is where the focus point is supposed to be. AF works by detecting contrast, so subjects which don't have much contrast will cause the camera to not focus well. So the statement that "Maybe the AF doesn't work too well in low contrast" is definitely true. If you want to enhance the AF system, you can buy a module that goes on to your hotshoe which adds an AF assist light. Some flashes have them built in. I have one on my monster 550EX and it works very well. Focused in extreme dark.

Realize that only the middle AF point detects contrast in both diagonal directions, the rest don't. I'm sure its a cost savings decision. So they won't focus as well.

It is possible that you have a fron focusing problem. Some number of very loud yelling customers do (i.e. some people at dpreview went nuts over this... but the actual number compared to units sold is fairly small.)

Eric
eric s is offline   Reply With Quote
Old Nov 2, 2003, 8:23 AM   #19
NHL
Senior Member
 
NHL's Avatar
 
Join Date: Jun 2002
Location: 39.18776, -77.311353333333
Posts: 11,547
Default

Eric

The pictures are clearly focused on the right edge of every sensors!

The milk bottles image is of particular interest, it's hard to tell: the closer area (front focus) is on the left, but the right "seems" sharper to me (ie back focus)... It'll be interesting if the same 1st three pictures of the fence are shot from the opposite angle (with the front focus on the left side of the sensors)... :?
NHL is offline   Reply With Quote
Old Nov 2, 2003, 12:44 PM   #20
Junior Member
 
Join Date: Oct 2003
Posts: 11
Default

I find it hard to believe some people can't see the focusing isn't right on some of the pictures. Take the first picture, sharpen it (to overcome the slightly soft output from the camera) and see the focus is clearly between the active focus point (red) and the one to the right. Some of the pictures, milk bottles, steps, look OK and the AF worked, that's the frustrating part it's not consistent, if it was I could compensate, but I don't see why I should have to. I had more consistent good quality photos from my old AE1 and T90, and they were easier to focus manually.

I've tested it with a chart, maybe not to Canon's spec (I have read about it), and I am happy it works with them, maybe it's a pity we don't all go around photographing charts and not 'awkward' things in real life
Dylan is offline   Reply With Quote
 
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



All times are GMT -5. The time now is 9:49 PM.