Manual focus question

[nilangsu mahanty]

I shoot a lot with my Sigma 10-20/4-5.6 EX DG HSM lens. Almost all my

compositions involve a very high depth of field. Earlier, I had been using AF,

but I wasn't too happy with the results. I have now switched to manual

focusing, and my question pertains to that.

 

Question 1: When I auto-focus (with the central AF point) at an object in the

far distance, the indicator on the distance scale sometimes stays between 3 ft

and infinity after the focus freezes. That surprises me ! What is the problem?

 

Question 2: In manual focus, the focusing ring turns smoothly to a point

beyond the infinity mark ( See both the photos attached). At this point

(around 5 or 6 mm to the right of the infinity mark), the smooth action stops

but the ring can still be rotated , though the movement becomes very tight.

Where should I stop to attain perfect infinity focus, at the infinity mark or

the point where the smooth movement of the ring stops?

 

I shoot in poor light a lot of times, and with the combination of small

maximum aperture of this lens, the 350D's dim viewfinder and the smallness of

the distant objects, I have to fully depend on the distance scale.

 

This is the first time I am using a lens with a distance scale and a smooth

and wide focusing ring. So I am new to using MF.

 

Any help will be sincerely appreciated.

 

Regards,

Nilangsu<div></div>

[nilangsu mahanty]

Sorry, I've attached the photos in the wrong order. The caption on the first should have read," Smooth movement of ring stops here". This one shows the ring parked at the infinity mark. :)<div></div>

[bill_mussett1]

I have a 30D and Canon's 10-22mm lens and have also noticed when using AF with only the

center AF point activated that infinity will be beyond the vertical focus mark. According to the

Canon manual this happens due to variations in temperature:

"To compensate for shifting of the infinity focus point that results from changes in

temperature. The infinity position at normal temperature is the point at which the vertical line

of the L mark is aligned with the distance indicator on the distance scale."

I don't know why AF would focus to the left of the vertical mark for a far away subject. It

might be worthwhile to do a focus test on the lens. Try: focustestchart.com.

[alan_myers]

Hi,

 

I don't have that lens, but imagine it has quite a broad depth of field, especially at the 10mm end of its focal length range. Heck, stopped down to f8 or f11, it may put everything from 3 or 5 feet to infinity into reasonably sharp focus.

 

However, you may be using it at or near wide open aperture, if shooting in low light where AF is near impossible.

 

One thing that's lacking on most zoom lenses is a complete distance scale. It's simply impractical to try to imprint on the lens, since the focal length changes. If you were using a prime lens, you'd be more likely to see a more detailed focus scale imprinted on it (but even then, DOF scales are not on all lenses, since AF has become so ubiquitous).

 

What you are looking for is the hyperfocal distance setting. This is the ideal focus point, in feet or meters or whatever, where depth of field will cover all subjects you wish to be in focus, near and far.

 

It will be a setting somewhere in between the closest object and infinity, but DOF will extend enough to cover both. It's usually about 1/3 of the distance from the closest to the farthest subject to be kept in focus.

 

The coverage of DOF changes with lens focal length (which is variable in a zoom of course) and aperture size. It also varies a little depending upon camera format (yours must be a 1.6X crop sensor, since this lens is designed as "digital only). Smaller sensor cameras essentially give greater DOF (Technically not, in reality, but for all practical purposes.)

 

Look on the internet for hyperfocal distance charts. These will show you more than the simplistic distance scale on the lens, combining distance with DOF indications. I know here are some people who are far smarter than me out there who have done all the calculations and posted the info online. You'll need to find a chart that covers the focal lengths of your lens, as well as it's aperture range, plus is applicable to 1.5X or 1.6X crop sensor cameras. (One for full frame/35mm film cameras might do... You'll actually end up with a little leeway because of the 1.6X sensor size.)

 

In order to provide accurate calculations, any chart will need to show specific, individual focal lengths. Maybe something like 10mm, 12mm, 15mm and 20mm, for example. There may be separate charts for each focal length. If you find yourself using your lens in between those settings, you'll have to estimate DOF and hyperfocal distance "between the lines" on the chart(s).

 

Also, if you have a primary subject that's particularly close to you, DOF can be reduced and may stop well short of infinity. In other words, the further the hyperfocal distance is from you, the deeper DOF will be.

 

Now, as to your question about the slight overrun of focus at infinity, I really can't say. Some lenses are deliberately designed to focus beyond infinity slightly, but those are usually telephotos and that's done to compensate for expansion inside the lens in hot weather, that otherwise might prevent infinity focus.

 

All I can suggest is that you take a close look at more examples of your lens in a friendly local camera shop, to see if they also tend to focus slightly beyond infinity. If so, I wouldn't worry about it. If not, perhaps there is some sort of bumper or stop at the end of the focus range, and that may be adjustable. It could be loose, damaged or just out of adjustment, allowing your lens to focus past infinity. Might be repairable, if needed.

 

However, if you begin using hyperfocal focusing methods, you will actually almost never be setting the lens all the way out to infinity, anyway. So it may not even matter. If you rack the focus all the way out to infinity, it will nearly always be well beyond the ideal hyperfocal distance, so you will actually be "lopping off" DOF closer to you and reducing DOF overall in the process.

 

One other thing, have you tried using "focus confirmation" with your camera? Most EOS have this, shown by a green LED that lights up in the viewfinder, and often with a confirming "beep" (that some people find annoying and turn off). Many models have a higher sensitivity to the center AF point, too, and it can be positioned over some object that's about 1/3 of the distance from you to infinity (or 1/3 of the way from your closest subject to your most distance).

 

Choose an object at that distance that happens to be better lit and has enough contrast for AF lock on. Press the shutter button halfway (or use Custom Function 4 to enable the * button on your camera to start AF separately from metering), but leave the lens' AF switch turned to "OFF" (USM lenses can be overridden manually without turning it off), and then adjust manual focus until the green LED lights up in the viewfinder and you get that reassuring "beep". Now with focus set, reframe the image you want to take and press the shutter release. This should get you pretty close to the hyperfocal distance. But, again, a lot depends upon the exact focal length being used, the actual distance being focused upon, the aperture setting, etc. For more accurate results, get a chart and refer to it.

 

 

[adam rosser]

I would advise against focussing at infinity with a wide angle lens. Doing so greatly reduces your depth of field for nearer objects. With a wide-angle lens like the 10-20, even at 20mm, the depth of field is so extreme as to not require focussing at infinity.

 

According to Bob Atkins dof calculator, at 20mm on a 1.6x DSLR, at f/5.6, focussing on an object at 4 metres will give you a depth of field from 2m to infinity. Focussing at infinity will mean the depth of field starts at nearly 4m and close objects (which are largest in the frame) will be at the extreme of depth of field (and may be a bit fuzzy).

 

http://www.bobatkins.com/photography/technical/dofcalc.html#calc

[samuel_parker]

I have had the same issue with wide angle lenses in the past. My working theory is that at those focal lengths, the depth of field is so great that minor differences in the actual infinity focus position won't really make any noticeable difference in the image. I'm not completely comfortable with that, but I haven't found a better solution.

 

Now, regarding depth of field and hyperfocal distance, it is my opinion that in many situations you are actually better off focusing at infinity as you are doing. The theory and mathematics of the hyperfocal distance are all correct, but the fact is that the assumptions about what is "acceptable sharpness" were made with early 20th century film and lens technology, and those assumptions don't hold up very well today. I recommend you read this article for more information on this subject: http://www.trenholm.org/hmmerk/DOFR.html

[William Michael]

Firstly, although I have not used this lens, I would not place too much accuracy on the distances engraved, to be fair, they are rough indicatives and only at two or perhaps three of the lens` apertures.

 

 

>>> Earlier, I had been using AF, but I wasn't too happy with the results. <<<

 

 

I am unclear in what way you were unhappy with the results. Does the AF not work accurately, (i.e. does it NOT focus of the focussed point?)

 

>>> I shoot in poor light a lot of times, and with the combination of small maximum aperture of this lens, the 350D's dim viewfinder and the smallness of the distant objects, I have to fully depend on the distance scale. <<<

 

Yes noted: but if the AF does indeed latch on OK, then I would simply use Av or Manual mode (merely to lock the aperture) and Focus and Recompose at pre determined distances, committed to memory.

 

 

The following numbers for memory:

 

 

At F5.6: 1 / 2 / 4

 

 

At F8: 1 / 2 / 3

 

 

These two sets of numbers are distances (in meters) at which your lens will be guaranteed BEYOND the Hyperfocal Distance at the two mentioned apertures for the lens at FL 10mm / 15mm / 20mm respectively.

 

 

There is latitude in my computations as I rounded the exact H.D. figure out further toward infinity.

 

 

Trimming them tighter would be easy to do. If you can judge distances in feet, with reasonable accuracy, memorizing whole numbers of a small dimension is better.

 

 

WW

[nilangsu mahanty]

>>> Bill Mussett - "I don't know why AF would focus to the left of the vertical mark for a far away subject. It might be worthwhile to do a focus test on the lens"

 

After asking these questions last night ( at India, we are around 10 hours ahead in time), I did some controlled MF tests ( tripod, MLU , self-timer) at 10 mm focal length today. The sharpness of objects in the far distance remains the same regardless of the alignment of the focussing ring at the 1 m mark in the DOF scale and at the infinity mark. That seems possible because of the extreme depth of field. This was corroborated by Bob Atkins' DOF calculator.

 

>>> Samuel Parker - " I have had the same issue with wide angle lenses in the past. My working theory is that at those focal lengths, the depth of field is so great that minor differences in the actual infinity focus position won't really make any noticeable difference in the image. I'm not completely comfortable with that, but I haven't found a better solution."

 

Very true. Even major differences ( 1 m and infinity) don't seem to matter.

 

>>> Adam Rosser - " I would advise against focussing at infinity with a wide angle lens. Doing so greatly reduces your depth of field for nearer objects. With a wide-angle lens like the 10-20, even at 20mm, the depth of field is so extreme as to not require focussing at infinity."

 

This is also what Bob Atkins' DOF table shows. At 16 mm focal length (in 35 mm format, not 1.6 X), and an aperture of f/11, if I focus at infinity , the DOF will be from 0.775 m and infinity. At 16 mm focal length (in 35 mm format, not 1.6 X), and an aperture of f/11, if I focus at 1 m , the DOF will be from 0.441 m and infinity. That's very interesting.

 

>>> Alan Myers - "One other thing, have you tried using "focus confirmation" with your camera? Most EOS have this, shown by a green LED that lights up in the viewfinder, and often with a confirming "beep" (that some people find annoying and turn off)."

 

Yes, I do note the beep. But I wasn't very confident of the AF itself. Thank you for taking the trouble to post a long and detailed explanation.

 

>>> William W. (Retired)- "I am unclear in what way you were unhappy with the results. Does the AF not work accurately, (i.e. does it NOT focus of the focussed point?)"

 

I had the nagging doubt that the AF wasn't working properly. But after today's trials, I will attribute the lack of sharpness in those photos where I thought the AF wasn't accurate to such a mundane thing as hand-shake. Everything today was blazingly sharp (however, note that I didn't use AF, but I am sure that THAT isn't the issue).

 

Thank you all so much for enlightening me. Thank you, Bob Atkins for your wonderful DOF calculator. Photo.net is a great place to visit.

 

Regards,

Nilangsu

[William Michael]

>>> Thank you [Alan Myers] for taking the trouble to post a long and detailed explanation. <<< (NM)

 

Well Alan Myers certainly put in the yards: he deserves to shout, I think.

 

:)

 

 

(Strange that if you shoot the hole in one, you have to buy the beers?)

 

 

Well, what I a attempting to say, is that Alan Meyers hit the nail on the head and set it all up.

 

 

It is all about Hyperfocal Distance, (get to understand that concept, and you are good to go).

 

 

To that end: get a good chart or lap top programme (`f-calc` - search it) or get a disc.

 

 

http://www.photo.net/bboard/q-and-a-fetch-msg?msg_id=00N1Z6

 

 

Cheers.

 

 

WW