Go Back   Steve's Digicams Forums >

LinkBack Thread Tools Search this Thread
Old Jul 18, 2006, 4:17 PM   #1
Senior Member
nelmr's Avatar
Join Date: May 2005
Posts: 493

With my Rebel XT and Tamron 17-50 f2.8 lens i am wondering what is the best way to set the hyperfocal distance?

I know how to get the hyperfocal. I can even download a program to run on my Pocket PC to give me the hyperfocal distance and such.

How though, is it best to focus at the given hyperfocal distance? For example, let's say that I try to shoot at 17mm and f11. The hyperfocal (with COC being 0.019mm) would be 4.47 feet. But how do I know where 4.47 feet away would be? If I misjudge by just a slight amount, lets say I focus at 4.2 feet instead, I would have a DOF behind the focus point of only 65 feet instead of infinity. This would ruin the photograph, esecially if printed large.

This leads me to another question, if I used autofocus how do I prevent the lens/camera from making a error in focusing?

If I switch to manual focus, how can I tell IF I have focused at exactly 4.47 feet for example. To me, the AF seems to do a better job than what I can do via the viewfinder since the viewfinder is so small that it is hard to see what is the sharpest focus.

Anyone that has had experince with doing this please let me know. I'm use to using a Fuji S5100 (1/2.7" sensor) so DOF was huge before and now with the XT I need to be careful.
nelmr is offline   Reply With Quote
Sponsored Links
Old Jul 18, 2006, 10:11 PM   #2
Senior Member
rey's Avatar
Join Date: Oct 2005
Posts: 949

This is also something that I've been interested on perfecting, but it looks like without hyperfocal distance on lenses, it's close to impossible to do.

My technique is just as you described. I use autofocus and eyeball the distance. If I want to lock focus, like when shooting multiple shots for pano, I focus using AF and then switch the lens to manual to "lock" the focus at that distance. I also lock exposure, obviously.

I'm also interested to hear from our experts on this.

rey is offline   Reply With Quote
Old Jul 19, 2006, 4:44 AM   #3
Nagasaki's Avatar
Join Date: Jan 2004
Posts: 897

I think the Rebel XT has a depth of field preview button. This stops down the lens so the image is fairly dark at small apertures but it does give some idea of what will be in foocus on the shot.

Nagasaki is offline   Reply With Quote
Old Jul 19, 2006, 7:31 AM   #4
Senior Member
nelmr's Avatar
Join Date: May 2005
Posts: 493

Supposing thata hyperfocal distance is 7 feet or less (my example was) could I use the focus distance marking on my lens:

How accurate are these? 7 feet, 3 feet, 2 feet, 1.5 feet, 0.89 feet? These markings seem to be on a logrithmic or exponential scale. I guess I could even use the 3 meter (about 10 feet) as my biggest measurement. IF these are accuarate couldn't I just set my lens to 10 feet, 7, 3, 2, 1.5, or .89 feet?

10 feet (3 meters)would allow me to have everthing from 5 feet to infinity to be in focus at 17mm f5.6 (actual hyperfocal would be 2.3m but 3 should be enough "safety" if the lens isn't percise with these markings, this is for viewing an 8x12 at arms length). Does anyone know if these markings are accurate?

Still, I'm faced with the same problem. What if I want to focus at 50mm? f8 would be about the highest I'd want to go since diffraction will start to occur if I went smalllerif I were to have a print at 16x24 inchs of a landscape viewed at arms length. This would need a hyperfocal of 27.3 m (90 feet). Were is this setting on my lens? How do I know how far away 90 feet is?

Can anyone help me with this delimia? How do I set the hyperfocal distance with my lens and camera?

DOF button may help but again, the viewfinder is tiny and it isn't always easy to see what is in focus and is not.
nelmr is offline   Reply With Quote
Old Jul 19, 2006, 3:13 PM   #5
Senior Member
nelmr's Avatar
Join Date: May 2005
Posts: 493

Did some more research and have some more questions:

Okay, I've predetermined that based partically on PPI that the largest I'd want to print a XT image is about 16x24 (144 PPI). But I want to make sure the rest of my absumptions are correct (also I realize that print size also has to do with subject matter, but in this case I am refering to landscapes).

For me to figure the smallest apperture possible (to use with hyperfocal calulations) I need to calculate the circle of confusion needed.

Fisrt, convert the 144 PPI to millimeters = 5.66 pixels per mm. Maximum lp/mm would be 2.8. So my goal for the 16x24 inch print would be roughly 3 lp/mm.

A 16x24 inch print is a 26.9 times inlargement for the canon APS-C sensor. This would mean that for a 3 lp/mm 16x24 inch print ( I would need a sensor resolution of about 80 lp/mm. This correlates to a CoC of 12.5 microns (1/80mm).

In this case the point at which an ariy disc's diameter meets the 12.5 micron CoC is just after f9.3 (for 17mm lens).

Thus if I wanted a hyperfocal landscape with a resolution of about 3 lp/mm when printed 16x24 inches then I should not go any larger than f9.3 or diffraction would destroy my image. Is this a correct assumption?

Also at http://www.normankoren.com/Tutorials/MTF6.html , norman says that hyperfocal is a MYTH for best focus. Here is what he has to say:

"The focus point in the above example is called the hyperfocal distance for f/16. When you focus at this distance, everything between the front DOF mark (about 2.7 meters in the example) and infinity is supposed to be "in focus." Well, sort of. Some authors, for example, photofocus.com, recommend focusing at the hyperfocal distance if you want a large range of focus out to infinity. I don't. Neither does Harold M. Merklinger in his page, Depth of Field Revisited. Nor does Zeiss.

If the part of the scene at infinity is at all important in the image— it's often visually dominant— you'll be disappointed with the sharpness, which is only 40% that of a high quality lens in focus; about one third what the eye can distinguish. Merklinger recommends focusing at infinity— you lose very little forward depth of field. I feel safe setting infinity focus opposite the far DOF mark corresponding to 2 stops larger than the actual f-stop setting (half the number). For example, if you are using f/8, it's safe to put the far f/4 DOF mark opposite infinity. It's a judgment call. When you make it, think about what parts of the image will be dominant. There is no rule to blindly follow. "

Thus I need to do some experiments with hyperfocal, infinity focus, and infinity focus at 2 stops lower.

However, my Tamron 17-50 f2.8 lens only has the distance markings and not the depth of field. Thus, how do I know when my lens is focused at isuch that at f9.3 the lens is focused at infinity for 2 stops lower? Maybe I don't understand what normen was meaning in this situation.

Any help on how to get the best landscape potos would be appreciated. Thanks.
nelmr is offline   Reply With Quote
Old Jul 19, 2006, 3:19 PM   #6
Senior Member
Sintares's Avatar
Join Date: Jul 2005
Posts: 647



Cheap laser ranger finder ($?) , even cheaper pocket range finder found on ebay ? (Got a Hugo Meyer for around $20)

Sintares is offline   Reply With Quote
Old Jul 19, 2006, 3:39 PM   #7
Senior Member
nelmr's Avatar
Join Date: May 2005
Posts: 493


This article has me thinking that hyperfocal may not be the best way to focus. And is a quite interesting read too regarding how to get best focus.

But when it talks about diffraction and its "D/1600d" formula is this refering to diffraction caused in the air or the diffraction of the lens (and thus it's effect on the sensor)? Any help in this would be really useful to me. Thanks.

nelmr is offline   Reply With Quote

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 On
Pingbacks are On
Refbacks are On

All times are GMT -5. The time now is 3:18 AM.