Steve'sDigicams

dafiryde · Jun 26, 2009, 11:13 PM

Quote:

Originally Posted by jelpee

I have owned this lens for over a year and absolutely love it. It is my everyday walk about lens. The one complaint I may have is that it is a big lens (77 mm opening). But with the SDM focusing, it is silent and fast.
Here's a link to some pictures I took a couple of weekends ago. Jefferson Memorial

When I purchase a new lens, I use the following focus test chart to confirm some of the issues that Mtgal referred to (especially back/front focusing, unusual distortion, etc). I also recommend buying from a seller like Amazon.com which has a generous 30 day return policy (plenty of time to test out a lens).
http://focustestchart.com/focus21.pdf

hope you can help
i too have this lens and love it but,i never realised i had a back focus problem until i tried this focus test chart. i have not been able to make any head way with this chart as for some reason i get completely different results if i use different zoom settings and also different chart to lens distances. its never consistant.
is it that you only have a problem if the focus line is not clear?
do you adjust focus to a balance ?
do you adjust at an average of the different zoom settings?
do you adjust to an average to different distance settings?
do you adjust only for the 16mm setting?
my lens takes great pics but yet the focus chart says i have a very bad back focus problem that would require a +10 setting at 16mm, but only a +5 at 30
even my 50-135 lens takes stellar crisp photos, but yet the focus chart also inticates that that lens has a very bad back focus problem

What to do, what am i doing wrong. is this chart really for prime lenses and not zooms

Dave

JimC · Jun 27, 2009, 7:59 AM

Dave:

It's not uncommon to see slightly different focus characteristics at different focal lengths and focus distances. You'll probably find that lighting temperature can impact Autofocus to some degree, too.

Most wider zooms starting out at 16mm do not have a flat focus field. IOW, as you get further away from the center, you may find that subjects closer to the camera are also in focus, because of curvature of field.

So, trying to test focus accuracy using the 16mm end of a zoom like yours is "tricky", especially if you're trying to shoot at an angle (the way you see the instructions for this type of chart). Personally, I wouldn't try to adjust it at 16mm. I'd zoom in some so that I'd be closer to the middle of the zoom range first.

Also, you tend to be using a focus distance where most of the image is in focus at the wider end of a zoom. The Hyperfocal Distance (the focus distance where you'll have the most Depth of Field for a given aperture setting) is only about 7 1/2 Feet at f/5.6 (a typical aperture you'd use) on the 16mm end of your lens, allowing everything from less than 4 feet to Infinity to be acceptably sharp at most viewing sizes using those camera settings (Aperture, Focus Distance, Focal Length). Smaller aperture openings (represented by higher f/stop numbers) will give you more depth of field for a given focus distance and focal length

See this Depth of Field Calculator for more information. Select your camera from the drop down list (as sensor or film size will impact the circle of confusion needed), plug in the desired aperture and focus distance and press calculate to see depth of field (including the hyperfocal distance). You can select metric distances if desired (for example, meters).

http://www.dofmaster.com/dofjs.html

Focus accuracy is going to be more critical if you're filling the frame more (as you would normally do by zooming in more with a lens like yours) for a given aperture setting.

Personally, I would not rely on that type of focus chart using the wide end of a zoom starting out at 16mm (and I'd zoom in some from it's widest setting anyway, in order to help with distortion and curvature of field issues when checking focus). I prefer to use something like books on a table, staggered at different focus distances from the camera to check for focus issues, focusing on the middle book, as you can see in the attached image, shooting at a wider aperture (smaller f/stop number).

That way, I can look for my focus point being sharpest, with text on the book covers in front or in back of my focus point getting softer as expected. I'd make sure to check the lens in bright enough light (as AF accuracy can suffer in dimmer lighting). I'd make sure to use a tripod, too.

JimC · Jun 27, 2009, 9:18 AM

P.S.

It would probably be a good idea to set stabilization to off when checking it (since you can sometimes see an issue with it when using a tripod). That would also take any sensor shift out of the equation to insure more consistent results when testing.

Note that with narrower apertures (as in f/5.6 when zoomed in much with many zoom lenses), you'll need to stagger the books so they're further apart (closer or further away from the camera) to see differences in text sharpness at typical test distances. I prefer that approach to using a test chart, since you're using more realistic focus distances with most lenses (instead of very close distances to a test chart trying to shoot at an angle).

Note that at most shooting distances where depth of field is an issue, you'll have more distance that's acceptably sharp behind the subject than you will in front of the subject you're focusing on. So, take that into consideration when checking for AF accuracy. See the depth of field calculator I mentioned in my last post for more information.

dafiryde · Jun 27, 2009, 11:14 AM

thanks Jim
one quick question
as is, i get a 1/3 - 2/3 split on the focus chart. 2/3 being on the backfocus end.
i was trying to achieve a 50 - 50 split, which is where i was getting conflicting results.
should i leave the 1/3 - 2/3 split. or should i go for 50 - 50 split

Dave

JimC · Jun 27, 2009, 11:21 AM

It should be closer to a 1/3 - 2/3 split at most focus distances where you'd be concerned about Depth of Field. That's why I mentioned that "you'll have more distance that's acceptably sharp behind the subject than you will in front of the subject you're focusing on" in my last post.

As you get closer to infinity (focusing on something further away), the split will change. You want to set it to where it's sharpest on your focus point. Then, the split should take care of itself for the focus distance, focal length and aperture you're using.

dafiryde · Jun 27, 2009, 1:33 PM

thanks again

Dave

Jun 27, 2009, 7:59 AM	#2
JimC Administrator Join Date: Jun 2003 Location: Savannah, GA (USA) Posts: 15,159	Dave: It's not uncommon to see slightly different focus characteristics at different focal lengths and focus distances. You'll probably find that lighting temperature can impact Autofocus to some degree, too. Most wider zooms starting out at 16mm do not have a flat focus field. IOW, as you get further away from the center, you may find that subjects closer to the camera are also in focus, because of curvature of field. So, trying to test focus accuracy using the 16mm end of a zoom like yours is "tricky", especially if you're trying to shoot at an angle (the way you see the instructions for this type of chart). Personally, I wouldn't try to adjust it at 16mm. I'd zoom in some so that I'd be closer to the middle of the zoom range first. Also, you tend to be using a focus distance where most of the image is in focus at the wider end of a zoom. The Hyperfocal Distance (the focus distance where you'll have the most Depth of Field for a given aperture setting) is only about 7 1/2 Feet at f/5.6 (a typical aperture you'd use) on the 16mm end of your lens, allowing everything from less than 4 feet to Infinity to be acceptably sharp at most viewing sizes using those camera settings (Aperture, Focus Distance, Focal Length). Smaller aperture openings (represented by higher f/stop numbers) will give you more depth of field for a given focus distance and focal length See this Depth of Field Calculator for more information. Select your camera from the drop down list (as sensor or film size will impact the circle of confusion needed), plug in the desired aperture and focus distance and press calculate to see depth of field (including the hyperfocal distance). You can select metric distances if desired (for example, meters). http://www.dofmaster.com/dofjs.html Focus accuracy is going to be more critical if you're filling the frame more (as you would normally do by zooming in more with a lens like yours) for a given aperture setting. Personally, I would not rely on that type of focus chart using the wide end of a zoom starting out at 16mm (and I'd zoom in some from it's widest setting anyway, in order to help with distortion and curvature of field issues when checking focus). I prefer to use something like books on a table, staggered at different focus distances from the camera to check for focus issues, focusing on the middle book, as you can see in the attached image, shooting at a wider aperture (smaller f/stop number). That way, I can look for my focus point being sharpest, with text on the book covers in front or in back of my focus point getting softer as expected. I'd make sure to check the lens in bright enough light (as AF accuracy can suffer in dimmer lighting). I'd make sure to use a tripod, too. Attached Images

Steve'sDigicams

Recent Camera Reviews

Most Active Forums

Most Popular Deals

Latest Feature Articles

Jun 27, 2009, 9:18 AM	#3
JimC Administrator Join Date: Jun 2003 Location: Savannah, GA (USA) Posts: 15,159	P.S. It would probably be a good idea to set stabilization to off when checking it (since you can sometimes see an issue with it when using a tripod). That would also take any sensor shift out of the equation to insure more consistent results when testing. Note that with narrower apertures (as in f/5.6 when zoomed in much with many zoom lenses), you'll need to stagger the books so they're further apart (closer or further away from the camera) to see differences in text sharpness at typical test distances. I prefer that approach to using a test chart, since you're using more realistic focus distances with most lenses (instead of very close distances to a test chart trying to shoot at an angle). Note that at most shooting distances where depth of field is an issue, you'll have more distance that's acceptably sharp behind the subject than you will in front of the subject you're focusing on. So, take that into consideration when checking for AF accuracy. See the depth of field calculator I mentioned in my last post for more information.

Jun 27, 2009, 11:14 AM	#4
dafiryde Senior Member Join Date: Apr 2004 Posts: 507	thanks Jim one quick question as is, i get a 1/3 - 2/3 split on the focus chart. 2/3 being on the backfocus end. i was trying to achieve a 50 - 50 split, which is where i was getting conflicting results. should i leave the 1/3 - 2/3 split. or should i go for 50 - 50 split Dave

Jun 27, 2009, 11:21 AM	#5
JimC Administrator Join Date: Jun 2003 Location: Savannah, GA (USA) Posts: 15,159	It should be closer to a 1/3 - 2/3 split at most focus distances where you'd be concerned about Depth of Field. That's why I mentioned that "you'll have more distance that's acceptably sharp behind the subject than you will in front of the subject you're focusing on" in my last post. As you get closer to infinity (focusing on something further away), the split will change. You want to set it to where it's sharpest on your focus point. Then, the split should take care of itself for the focus distance, focal length and aperture you're using.

Jun 27, 2009, 1:33 PM	#6
dafiryde Senior Member Join Date: Apr 2004 Posts: 507	thanks again Dave