Oyvind Tryti
Posts: 17
Location: Oslo, Norway
Registered: 14 May 2007
|
Focusing the Sigma 8/3.5
Posted: 6 Apr 2008 at 23:05 GMT
|
|
I have used the Sigma 8/3.5 on an EOS 400D for a year or so, and been satisfied with it.
My brother in law recently came from Phoenix, Arizona to Oslo to attend to our silver wedding. Since his P&S really seemed to be a POS, he was asking for advice on what to replace it with. The only logical thing for me to do, was to give him the 400D.
This, of course, gave me a good reason to get a 40D.
The 40D has live view, and I decided to have a look at the focusing of the Sigma 8/3.5. I autofocused it on something far away (the barrel stopped almost at the infinity mark), switched the lens to manual focus and checked the live view at 10x magnification. It was somewhat sharp, but not as good as I had expected. Then I adjusted the focusing until it was as sharp as possible. I looked at the distance marking, and was flabbergasted. It was set at just this side of 1 m! The test was repeated several times - autofocus would set the lens at infinity, manual refocusing would set it at just under 1 m.
I tested several Canon lenses too, none of them had this problem. It is possible to get more exact focus by using live view, but none of them were consistently far off like the Sigma was.
I photograph panoramas with my lens set for infinity (it is actually taped down). I want everything at a far distance to be in focus - this way anything larger than 2 mm will be recognizable as long as the lens can resolve it. See luminous-landscape.com/tutorials/digital%20focusi... for an explanation.
So, for the last year I have been using my Sigma focused PAST infinity! ARGH! No wonder my tests showed that I needed to stop down to at least f5.6, preferrably f8!
New tests show that best image quality is obtained at F4-5.6.
Owners of Sigma 8/3.5 might want to do this test on their own copy - if you have a camera with live view, it is a matter of minutes checking it. If not, a series of test images, where focus is changed a little for each image will tell you if this is a problem.
I trusted the distance scale, and did not test for focusing accuracy. The field of depth is so large at 8 mm, that it is not possible to see in the viewfinder that it not exactly right.
The fact that my copy of the lens does not autofocus correctly, and that the distance scale is wrong is of little significance to me, as long as I know what setting I should tape the distance scale at to obtain infinity focus..
|
|
alert moderator
|
|
mhc1
Posts: 159
Location: Hamburg, Germany
Registered: 29 Dec 2005
|
Re: Focusing the Sigma 8/3.5
Posted: 8 Apr 2008 at 7:04 GMT
|
|
Oyvind: there are plenty of posts in the DSLR-Forum regarding the monitor of the 40D.
--> it should not be used to set focus when magnified.
They complain about the bad quality of the monitor.
Use the remote control- there is a live-view that´s supposed to be more exact.
I use a 40D / Sigma8 like you too, may focus is infinity like you said app. 8 when useing AF.
I can see, that the scale is about true when using AF.
But I use manual focus and try to focus hyperfocal to get the most out of it.
So I set the infinity mark just before the F-stop I use.
Not sure , if this is advicable - but it works good for me. My problems are likely to be found in my workflow I guess.
Ciao
Mike
www.360de.de
|
|
alert moderator
|
|
frozenjaZz
Posts: 34
Location: France
Registered: 8 Apr 2008
|
Re: Focusing the Sigma 8/3.5
Posted: 8 Apr 2008 at 8:02 GMT
|
|
Hi there,
this is funny, I just did the exact same test this morning with my 450D. I just bought a Sigma, and found that the results were quite soft, even with the distance set to 1m...
I did a series of shots with distance set to 0.5, between 0.5 and 1, 1m, and between 1 and infinity.
The best one was between 0.5 and 1...
But this morning I tried with live view... wow, it is so easy to get a perfect focus !! I tried on distant trees... the setting was very slightly before 1m (anything above 1m gives very soft results, even for very distant objects...)
So I get results similar to yours... and yes, autofocus is unusable for distant objects. However it works nicely for very close objects (I mean, very very close).
|
|
alert moderator
|
|
Mike Jensen
Posts: 20
Location: Sacramento, CA, United States
Registered: 8 May 2006
|
Re: Focusing the Sigma 8/3.5
Posted: 8 Apr 2008 at 18:55 GMT
updated: 8 Apr 2008 at 18:58 GMT
|
|
Thanks so much for your tests.
I also have the Sigma 8mm 3.5, and have run some tests this morning. My lens is sharpest dead on at the 1m focus mark- getting softer towards 0.5 and again towards the infinity mark. I was amazed that adjusting these small focusing increments (from .5 to infinity) made MORE difference in the image sharpness than the fstop! At the 1m mark, I now have a sharp lens
Mike Jensen
Jensen Wedding Films
www.JensenVideo.com
|
|
alert moderator
|
|
frozenjaZz
Posts: 34
Location: France
Registered: 8 Apr 2008
|
Re: Focusing the Sigma 8/3.5
Posted: 8 Apr 2008 at 19:02 GMT
|
|
Good news for you !
Mine seems to be soft whatever the distance or the f-stop...
I have a problem with Sigmas !! All the sigma lenses I ever bought in my life were bad...
This one is no exception, my Peleng is way sharper ! But the Peleng has a big problem with flare, this is why I wanted to upgrade to Sigma, but given the price paid, I am extremly disappointed !
|
|
alert moderator
|
|
mhc1
Posts: 159
Location: Hamburg, Germany
Registered: 29 Dec 2005
|
Re: Focusing the Sigma 8/3.5
Posted: 8 Apr 2008 at 19:25 GMT
|
|
Mike:
there is NO difference in DOF ????
Mine just behaves as expected when setting the 8 mark on the F-stop (or the 1m mark -- I must test again on the WE !!!)
But using 0.5m at F5.6 is definately unsharp at 20cm and at 10m.
I did some Panos, where there where objects ab. 10cm away - these I wanted to be sharp --> NO problem at all, but infinity at the same time: impossible.
I used Iso 400 and F11 to get a sharpness for some 1 -2 meters.
I don´t complain- the DOF is immense - but when I convert to MOV or JPG --> sharpnes is gone - hurray !!
So what can I do ??
I have a really good TIF and bad MOVs & JPGs.
Ciao
mike
www.360de.de
|
|
alert moderator
|
|
Oyvind Tryti
Posts: 17
Location: Oslo, Norway
Registered: 14 May 2007
|
Re: Focusing the Sigma 8/3.5
Posted: 8 Apr 2008 at 21:55 GMT
updated: 8 Apr 2008 at 22:17 GMT
|
|
Mike (mhc1), I find the monitor to be plenty good enough for critical focusing at 10X magnification. It is just a case of moving focus back and forth, and you will quickly find the correct point for optimum sharpness. I tried this with several lenses up to 500/4, and it is always easy to see when optimum focus has been achieved.
See this article about using live view for critical focusing: www.naturephotographers.net/articles0408/dw0408-1...
Unless you are making panoramas with very close details, you might want to try making one where the lens is focused on whatever is farthest away. You will stop down to f4 or f5.6 in any case for best quality, and if the closest object is the ground the tripod is standing on, you might be surprised at the quality of the panorama.
Hyperfocal focusing is really an outdated idea that was useful for film, where the size of the Circle of Confusion was used to determine the depth of field, and thus also what the hyperfocal distance is. The size of the CoC was determined by the final magnification needed. Modern digital cameras require a smaller CoC, because the magnification is larger, and the corresponding calculated depth of field is smaller. So, by setting the lens to the hyperfocal distance calculated for film, objects at infinity will actually be outside the depth of field. And in a panorama (except 'claustro-panos') you want distant objects to be as sharp as possible.
|
|
alert moderator
|
|
John Houghton
Posts: 2133
Location: Hitchin, United Kingdom
Registered: 17 Jan 2005
|
Re: Focusing the Sigma 8/3.5
Posted: 9 Apr 2008 at 6:08 GMT
|
|
Oyvind Tryti said: Hyperfocal focusing is really an outdated idea that was useful for film, where the size of the Circle of Confusion was used to determine the depth of field, and thus also what the hyperfocal distance is. The size of the CoC was determined by the final magnification needed. Modern digital cameras require a smaller CoC,
Hyperfocal focusing is as valid now as it was in the past. You may well need to use different CoC values, but that does not invalidate the technique. You just need to be aware that the particular hyperfocal distances engraved on the lens mount may not be applicable in your digital setup. Use a DoF calculator such as this one for a range of Nikon cameras, and which also allows you to enter a specific CoC of your own:
www.nikonians.org/html/resources/guides/dof/hyper...
John
|
|
alert moderator
|
|
Oyvind Tryti
Posts: 17
Location: Oslo, Norway
Registered: 14 May 2007
|
Re: Focusing the Sigma 8/3.5
Posted: 9 Apr 2008 at 6:47 GMT
|
|
John Houghton said: Hyperfocal focusing is as valid now as it was in the past.
Agreed - as long as one knows what one is doing. But just blindly following the field of depth markings on the lens, which normally are calculated for film use, will lead to a hyperfocal distance that is too close to the camera.
The new lenses for reduced frame digital cameras may have field of depth markings using a smaller CoC - do you know if this is the case, John?
In any case, I think panographers that focus at the hyperfocal distance owe it to them selves to try out focusing at infinity, and compare the result (see the link in my first post for the rationale behind this statement).
If this focus problem with my Sigma is generic, it may be that people focusing their Sigma at the hyperfocal distance engraved on the barrel are accidentally focusing closer to infinity, and for this reason are satisfied with the result.
|
|
alert moderator
|
|
mhc1
Posts: 159
Location: Hamburg, Germany
Registered: 29 Dec 2005
|
Re: Focusing the Sigma 8/3.5
Posted: 9 Apr 2008 at 8:08 GMT
|
|
I would expect, that a DG lens ist optimized for the use with digtal camere --> therefore the depth marks to be accurate.
The DOF- calculator: www.dofmaster.com/dofjs.html
shows very impressive that the CoC of Nikon (SD200/300) and Canon (40D ...) are very much the same : 0,19 and 0.20.
So I expect Sigmas 8mm lens to be accurate - it is not a
50 $ lens though.
As far as I see it focuses right: infinity is ( on the mark.
I did not proof the values of the DOF calculato - I just set the AF measured value onto the F- Stop - that´s it.
But I´ve seen so many panos that are just made by setting the focus with AF - switching to manual and the shoot.
And they were remakable sharp !
ciao
mike
www.360de.de
|
|
alert moderator
|
|
frozenjaZz
Posts: 34
Location: France
Registered: 8 Apr 2008
|
Re: Focusing the Sigma 8/3.5
Posted: 9 Apr 2008 at 12:04 GMT
|
|
|
What I can tell is, if I want the focus to be "infinite", I have to set it at the "1m" mark. If I use the "infinity" mark, then everything is out of focus (totally blurred).
|
|
alert moderator
|
|
Matt Rogers
[360 Precision]
Posts: 181
Location: Oxford UK, United Kingdom
Registered: 16 Jun 2005
|
Re: Focusing the Sigma 8/3.5
Posted: 9 Apr 2008 at 12:30 GMT
|
|
If this is the case then why would you keep the lens ? If my 8mm was blurry at infinity I'd return it for a new one.
Matt
|
|
alert moderator
|
|
Hans Nyberg
Posts: 713
Location: Denmark
Registered: 28 Aug 2005
|
Re: Focusing the Sigma 8/3.5
Posted: 9 Apr 2008 at 12:58 GMT
|
|
I guess you do not know where the infinity is.
Look below.
The infinity is the white vertical line with a horizontal line to the infinity mark.
If you put it on the infinity mark it is beyond infinity.
This is how most lenses are made and the reason is that focusing can change because the mtrl changes during extreme temperatures.
I did test the influence on my old Sigma8mmm and it only had little influence on focus at infinity, but the new Sigma 8mm is different.
See also my page here
www.panoramas.dk/technics/focus.html
Hans
|
|
alert moderator
|
|
Oyvind Tryti
Posts: 17
Location: Oslo, Norway
Registered: 14 May 2007
|
Re: Focusing the Sigma 8/3.5
Posted: 9 Apr 2008 at 13:13 GMT
|
|
mhc1 said: But I´ve seen so many panos that are just made by setting the focus with AF - switching to manual and the shoot.
And they were remakable sharp !
That's exactly my point! Focus correctly on far objects, and let the close ones be a little unsharp - the final result is a very sharp looking panorama.
I cannot use this method with my lens - in any case, the focus ring turns so easily that I tape it down to make sure that it is not moved accidentally while the camera is turned from one stop to the next.
Hopefully this is a rare problem that only affects a few of the Sigma lenses, but owners should test to make sure that their copy is good.
Matt Rogers said: If this is the case then why would you keep the lens ? If my 8mm was blurry at infinity I'd return it for a new one.
The thought has struck me. I will talk to the store, but I have had it for more than a year before discovering the fault. That might be a problem, even if we have two years guarantee here in Norway.
|
|
alert moderator
|
|
Oyvind Tryti
Posts: 17
Location: Oslo, Norway
Registered: 14 May 2007
|
Re: Focusing the Sigma 8/3.5
Posted: 9 Apr 2008 at 13:32 GMT
|
|
Hans Nyberg said: I guess you do not know where the infinity is.
Hans, your link to ephotozine is broken.
And yes, I do know how to set a lens to infinity. I have been using and focusing lenses manually for 40 years, and this Sigma is the first lens I have encountered to be far off at room temperature. And yes, I am also aware of focus changing with temperature - that is why lenses are able to focus past infinity.
The Sigma is the only autofocusing lens I have (of the ones I have tested so far) that consistently sets the focus distance more or less correctly (at infinity, the vertical white line is placed a fraction to this side of infinity), whereas manual refocusing placed it at slightly less than 1m.
I am planning to make a page similar to yours available on my own site, but I will include pictures of the actual setting of the distance scale along with the images, in order to show how large the discrepancy can be.
The 500/4 was very impressive - auto focus was dead on, any adjustment I made would make the image less sharp.
|
|
alert moderator
|
|
Prev
