New Sigma 17-50 f2.8 OS on D7000. Backfocus?

Started Dec 16, 2012 | Discussions
gen123
New MemberPosts: 4
Like?
New Sigma 17-50 f2.8 OS on D7000. Backfocus?
Dec 16, 2012

Hi all,

I just bought a new Sigma 17-50 f2.8 OS from Amazon and did the battery test to see if there were any front/back focus issues.

I used a tripod + timer, and focused on the second battery from the left. The third battery appears to be more focused at 17mm and 26mm, while the correct battery is in focus at 38mm and 50mm.

I also took photos of a lens test chart, where again it appears to be backfocusing at 17mm.

I'm fairly new at all of this, so I'd like to get your opinions. Do you guys see a backfocusing issue as well? Would I be able to correct this using the AF fine tune in my D7000? Should I exchange the lens / send it to Sigma for calibratoin?

Any help is appreciated!

17mm

17mm crop

26mm

26mm crop

38mm

38mm crop

50mm

50mm crop

17mm

50mm

Nikon D7000
If you believe there are incorrect tags, please send us this post using our feedback form.
SteveL54
Senior MemberPosts: 2,610
Like?
Re: New Sigma 17-50 f2.8 OS on D7000. Backfocus?
In reply to gen123, Dec 16, 2012

gen123 wrote:

Hi all,

Hi Gen,

Welcome to the forum!

I just bought a new Sigma 17-50 f2.8 OS from Amazon and did the battery test to see if there were any front/back focus issues.

I used a tripod + timer, and focused on the second battery from the left. The third battery appears to be more focused at 17mm and 26mm, while the correct battery is in focus at 38mm and 50mm.

I also took photos of a lens test chart, where again it appears to be backfocusing at 17mm.

It does look like the lens/camera is back focusing at the short end of the lens's focal length.

Have you tried the new lens on larger subjects further away? The reason I ask is that lens focus errors can be influenced by target distance. Both of these tests are good for close-in AF testing but they often exercise the lens near its minimum focus distance. I'm not familiar with this Sigma lens, but characteristics of a few lenses interact with the camera's phase detect AF unit and cause it to mis-focus only at close-in distances.

If you send the lens in for AF calibration the test rig used by the service center will be set up to test the lens in the neighborhood of 40 times the lens focal length.

I recommend a test like this that is carried out at a further distance.

http://www.pbase.com/slockhart54/auto_focus_test_procedure

I notice from the EXIF that you are using a D7000. There have been reports that the ambient light color temperature influences the AF accuracy. These reports have mentioned that incandescent or low color temperature lighting causes back focusing. I don't know if it's been confirmed, but it's worth trying tests outdoors or with higher color temperature daylight CFLs to see if that influences the AF test results.

I'm fairly new at all of this, so I'd like to get your opinions. Do you guys see a backfocusing issue as well? Would I be able to correct this using the AF fine tune in my D7000?

The AF fine tune may help if it corrects the AF at the short end of the focal length without affecting the long end as much. I have noticed this with a couple of my zooms, so AFFT worked well with my lens that had a tendency to back focus at the short end, but not at the long end. But it's best to fine tune at the distance you intend to most use the lens.

Should I exchange the lens / send it to Sigma for calibratoin?

The problem could be the lens or the camera. Have you tried the camera with another lens under similar conditions?

Even if I was sure it was the lens (not the camera) causing the BF problem I'm not sure which course of action to recommend here. If you send the lens off for calibration then your window of opportunity to return/exchange the lens will close should the lens not be fixed to your satisfaction. If you exchange, then it's another roll of the dice.

Steve

Reply   Reply with quote   Complain
HSway
Senior MemberPosts: 2,658
Like?
Re: New Sigma 17-50 f2.8 OS on D7000. Backfocus?
In reply to gen123, Dec 16, 2012

I wouldn’t worry. You are not likely to be shooting at 17mm at 2.8 these ranges. For similar scenarios especially and on tripod the LV (focus) is also an alternative.

You can fine tune a zoom lens and in case of different values at various FL settings, you can optimize for the best result in general or for your preferred setting. If you are getting results you simply are not happy with, or it shows a bad variation, exchange the lens.

In case your lens and camera is Nikon and both still within the warranty – you can send it in, Nikon must adjust it whether it’s a body or the lens.

But I mainly add to very good advice of the above poster because I had 2 of these Sigmas and both were spot on d7000. but I think I didn’t even go up to 17mm. My 24-70 Nikkor for example (quite recent purchase) was also spot on on d600 from 70 to 35 @2.8, at 24mm it was off. I didn’t mind it at all. That is quite good for a zoom, I have seen much more peculiar results for this sort of variability and at the same time many zoom lenses that were easy to fine tune using the camera.

-- hide signature --
Reply   Reply with quote   Complain
Guidenet
Forum ProPosts: 12,454Gear list
Like?
Re: New Sigma 17-50 f2.8 OS on D7000. Backfocus?
In reply to gen123, Dec 16, 2012

Doesn't look bad to me. If you don't like it, fine tune it a smidgin.

But look here. What in the world are you doing? Get out there and take real pictures. Enjoy that excellent camera and lens and quit fussing around. I know of few people who take battery pictures. It would bore me to death. It should you as well.

There's a lot of crazy measurabators on these forums who don't ever seem to be into photography but into super testing glass. I don't know why. If you're not into photography, I can't see why it matters. If you are into photography fine tune it or work around it. No lens/camera combination is perfect and you don't know what might be a very slight bit off. Sending it back over and over again would be an effort in frustration and further detract from the primary goal of having a good time.

The best tests on new glass is to try them out in real situations to see if they perform well. If they do, keep them. If not trade them in. If you're trading more lenses than keeping, something is wrong with you. This is supposed to be fun.

Good luck and have a great holiday season taking real pictures and creating real images.

-- hide signature --

Cheers, Craig
Follow me on Twitter @craighardingsr : Equipment in Profile

 Guidenet's gear list:Guidenet's gear list
Nikon D300 Nikon D700 Nikon D3S Nikon D800 Nikon AF-S DX Nikkor 18-135mm f/3.5-5.6G ED-IF +24 more
Reply   Reply with quote   Complain
gen123
New MemberPosts: 4
Like?
Re: New Sigma 17-50 f2.8 OS on D7000. Backfocus?
In reply to SteveL54, Dec 17, 2012

Thanks for the welcome and reply!  I've been lurking around these forums for a long time, and have always been impressed by the helpfulness of folks like yourself :).

It does look like the lens/camera is back focusing at the short end of the lens's focal length.

Have you tried the new lens on larger subjects further away? The reason I ask is that lens focus errors can be influenced by target distance. Both of these tests are good for close-in AF testing but they often exercise the lens near its minimum focus distance. I'm not familiar with this Sigma lens, but characteristics of a few lenses interact with the camera's phase detect AF unit and cause it to mis-focus only at close-in distances.

I brought the lens out today to take some pictures with bigger subjects in both natural and artificial light.  To be fair, these were hand-held, thus user-error can always come into play... but the results have been fairly consistent as far as I can tell.  I always used the center focus point, and did not do any focus-and-recomposing.  Check out these photos:

17mm, the drink is not in focus at all.  it's possible that the camera locked focus on the wrong thing though.

17mm (not LV), appears that the poster behind the subject is in focus (I put the AF point in between her eyes)

17mm (LV), subject is in focus

I tried a couple other FL, and to me, it seems the backfocus persists from 17mm to about 28mm.  From 28mm to 50mm, the focus was more or less correct.

If you send the lens in for AF calibration the test rig used by the service center will be set up to test the lens in the neighborhood of 40 times the lens focal length.

I recommend a test like this that is carried out at a further distance.

http://www.pbase.com/slockhart54/auto_focus_test_procedure

I notice from the EXIF that you are using a D7000. There have been reports that the ambient light color temperature influences the AF accuracy. These reports have mentioned that incandescent or low color temperature lighting causes back focusing. I don't know if it's been confirmed, but it's worth trying tests outdoors or with higher color temperature daylight CFLs to see if that influences the AF test results.

I did the battery test again this morning in natural light, with similar results.

The AF fine tune may help if it corrects the AF at the short end of the focal length without affecting the long end as much. I have noticed this with a couple of my zooms, so AFFT worked well with my lens that had a tendency to back focus at the short end, but not at the long end. But it's best to fine tune at the distance you intend to most use the lens.

Should I exchange the lens / send it to Sigma for calibratoin?

The problem could be the lens or the camera. Have you tried the camera with another lens under similar conditions?

I also using my Nikon 35mm f1.8 and Tamron f2 60mm, and both were spot on with the focus.

Even if I was sure it was the lens (not the camera) causing the BF problem I'm not sure which course of action to recommend here. If you send the lens off for calibration then your window of opportunity to return/exchange the lens will close should the lens not be fixed to your satisfaction. If you exchange, then it's another roll of the dice.

Steve

Reply   Reply with quote   Complain
gen123
New MemberPosts: 4
Like?
Re: New Sigma 17-50 f2.8 OS on D7000. Backfocus?
In reply to Guidenet, Dec 17, 2012

I agree that we should all be taking real pictures rather than boring battery tests.  However, given some of the stuff I've read about this (and other) lens, I wanted to make sure I did my due diligence on it by testing it with as few variables as possible.  In regular usage, I find that it's not always easy to tell whether a shot is bad because of user error or because of the hardware.

In this case, I wanted to consult with the more experienced users on this forum to see whether the lens I got is a "bad copy".  As you said, no lens/camera combination is perfect, but some are worse than others.  I really don't know the extent of how bad these camera/lens issues can get, and I figured it wouldn't hurt to have a few more people to look at it.   It's hard to have a good time taking photos when things don't work the way you want them to!

Guidenet wrote:

Doesn't look bad to me. If you don't like it, fine tune it a smidgin.

But look here. What in the world are you doing? Get out there and take real pictures. Enjoy that excellent camera and lens and quit fussing around. I know of few people who take battery pictures. It would bore me to death. It should you as well.

There's a lot of crazy measurabators on these forums who don't ever seem to be into photography but into super testing glass. I don't know why. If you're not into photography, I can't see why it matters. If you are into photography fine tune it or work around it. No lens/camera combination is perfect and you don't know what might be a very slight bit off. Sending it back over and over again would be an effort in frustration and further detract from the primary goal of having a good time.

The best tests on new glass is to try them out in real situations to see if they perform well. If they do, keep them. If not trade them in. If you're trading more lenses than keeping, something is wrong with you. This is supposed to be fun.

Good luck and have a great holiday season taking real pictures and creating real images.

-- hide signature --

Cheers, Craig
Follow me on Twitter @craighardingsr : Equipment in Profile

Reply   Reply with quote   Complain
Guidenet
Forum ProPosts: 12,454Gear list
Like?
Re: New Sigma 17-50 f2.8 OS on D7000. Backfocus?
In reply to gen123, Dec 17, 2012

gen123 wrote:

I agree that we should all be taking real pictures rather than boring battery tests. However, given some of the stuff I've read about this (and other) lens, I wanted to make sure I did my due diligence on it by testing it with as few variables as possible. In regular usage, I find that it's not always easy to tell whether a shot is bad because of user error or because of the hardware.

In this case, I wanted to consult with the more experienced users on this forum to see whether the lens I got is a "bad copy". As you said, no lens/camera combination is perfect, but some are worse than others. I really don't know the extent of how bad these camera/lens issues can get, and I figured it wouldn't hurt to have a few more people to look at it. It's hard to have a good time taking photos when things don't work the way you want them to!

That's true and I understand, my friend. There's a good article over at Lens Rentals about this point you might want to read. Basically, it said that you just can't tell what the cause is for a little front or back focus. Just being a few putzometers off in the mirror box or mount can affect a lens but not another. It might not be the lens at all, but the combination of factors you just can't control. The most perfect lens in the world might not be so on your camera.

Moreover, the tests change a little based on humidity, pressure and other factors from time to time. You might be correcting something that really doesn't exist. Like I said, your shots looked ok to me in most cases. A little fine tuning might get you as close as possible with your camera. I don't know. Also, there are work arounds we've delt with for years. After a while you know where your lens / camera combination likes to be. You learn to be at that sweet spot.

Finally, a good compelling image with good composition trumps most any fine tuing to the lens you can do, providing the lens is pretty good anyway and yours certainly seems to be. Afterall, cheaper cameras can't fine tune and these minor issues are worked around or ignored.

The problem is you can never get it perfect. If you try, you just get frustrated. When you get frustrated with a lens / camera combo you lose confidence in it and if that happens, might as well sell it all and start over. You'll never be happy. That's why I say it might be better to only test under field conditions. Don't look for corrections that can never be perfect. Don't let it ever get to that point. Everything will eventually become a "bad copy." Blame bad images on yourself only. That's the real culprit anyway.

I recently bought that lens you just bought for my daughter who's a wedding pro and still shoots DX. The Nikon 17-55 f/2.8 is just too heavy for her to love it. She loves the Sigma and creates wonderful images with it.. absolutely beautiful.

Good luck and have a wonderful holiday season.

-- hide signature --

Cheers, Craig
Follow me on Twitter @craighardingsr : Equipment in Profile

 Guidenet's gear list:Guidenet's gear list
Nikon D300 Nikon D700 Nikon D3S Nikon D800 Nikon AF-S DX Nikkor 18-135mm f/3.5-5.6G ED-IF +24 more
Reply   Reply with quote   Complain
SteveL54
Senior MemberPosts: 2,610
Like?
Re: New Sigma 17-50 f2.8 OS on D7000. Backfocus?
In reply to gen123, Dec 18, 2012

gen123 wrote:

I brought the lens out today to take some pictures with bigger subjects in both natural and artificial light. To be fair, these were hand-held, thus user-error can always come into play... but the results have been fairly consistent as far as I can tell. I always used the center focus point, and did not do any focus-and-recomposing. Check out these photos:

These photos do tend to support the battery and slant chart tests that the AF error extends to normal distances, but by themselves are not that conclusive. This is typical of many "real" subject photos. On the other hand, if these two mis-focused photos are typical of nearly all your shots between 17-24mm then I'd say there's a calibration/adjustment problem with the AF system. Unfortunately it doesn't narrow the source to either the lens or the camera body.


17mm, the drink is not in focus at all. it's possible that the camera locked focus on the wrong thing though.

In the first photo it's possible that the high contrast grout line on the tile background influenced the PDAF system's correlation function. It's close enough to the selected target (cup) to possibly show up on the central sensor. The AF sensors are influenced by high contrast detail immediately outside of the brackets in the view finder. As a rule of thumb I try to have a buffer which extends 1/2 the width of the marked bracket around all sides of the selected AF bracket. If there's anything in that buffer that's high contrast and not at the same distance as the desired AF target then there's an increased risk of AF error.


17mm (not LV), appears that the poster behind the subject is in focus (I put the AF point in between her eyes)

The poster is not in focus either. The best focus is behind the woman's face but in front of the poster. The woman's face is framed in the photo such that it covers a greater area for AF sensors than the cup in the first photo so I think the issue described above is less likely in this case, but the very dark jacket hood which forms a high contrast rim against the light poster might also influence the AF system if it's in the central AF sensor's area of influence. The rim which extends behind her head is better focused than her face.

But these are just elements of the supplied photos that might influence the AF. Chances are slim that these issues come into play in all the other photos you've taken between 17-24mm. Even if you do get a very occasional in-focus photo at these settings does not prove you don't have an error either. Sometimes things happen that cause a canceling error and you end up with an in-focus photo. (A few errors occur even when the system is properly calibrated.)

If you send the lens in for AF calibration the test rig used by the service center will be set up to test the lens in the neighborhood of 40 times the lens focal length.

I recommend a test like this that is carried out at a further distance.

http://www.pbase.com/slockhart54/auto_focus_test_procedure

I notice from the EXIF that you are using a D7000. There have been reports that the ambient light color temperature influences the AF accuracy. These reports have mentioned that incandescent or low color temperature lighting causes back focusing. I don't know if it's been confirmed, but it's worth trying tests outdoors or with higher color temperature daylight CFLs to see if that influences the AF test results.

I did the battery test again this morning in natural light, with similar results.

At least that seems to eliminate the old color temp question.

The reason I recommended the type of test in the link above is that if one follows the procedure and the lens back focuses, it's hard to come up with a "user error" excuse for the result. The test uses an ideal AF target. If significant AF errors occur with that test then the AF system (body and/or lens) needs adjustment.

The AF fine tune may help if it corrects the AF at the short end of the focal length without affecting the long end as much. I have noticed this with a couple of my zooms, so AFFT worked well with my lens that had a tendency to back focus at the short end, but not at the long end. But it's best to fine tune at the distance you intend to most use the lens.

Should I exchange the lens / send it to Sigma for calibratoin?

The problem could be the lens or the camera. Have you tried the camera with another lens under similar conditions?

I also using my Nikon 35mm f1.8 and Tamron f2 60mm, and both were spot on with the focus.

Do you have access to another lens in the 17-24mm focal length range? How about another camera body?

If not, then I suppose you could exchange for another new lens before your exchange privilege runs out and see how it behaves.

edit: Almost forgot to ask if you tried AF fine tuning?

Steve

Reply   Reply with quote   Complain
gen123
New MemberPosts: 4
Like?
Re: New Sigma 17-50 f2.8 OS on D7000. Backfocus?
In reply to SteveL54, Dec 18, 2012

Do you have access to another lens in the 17-24mm focal length range? How about another camera body?

If not, then I suppose you could exchange for another new lens before your exchange privilege runs out and see how it behaves.

edit: Almost forgot to ask if you tried AF fine tuning?

Steve

I have a 18-105mm and a 18-200mm (which I want to sell and replace with the Sigma) that I can take some shots with. I've had a limited time shooting with them in the past (the 35mm f1.8 stays on most of the time), but I haven't noticed any obvious focusing issues. I'll definitely try both of them in the 18-24mm range now that you mention it though.

As for AF fine tuning on the Sigma 17-50... I played around with it last night in increments of 5, and seemed to get the sharpest (handheld) photos with a AF fine tune of -20. Strangely, even the shots at 50mm became slightly sharper with the adjustment. I wonder how much of a difference AF fine tune makes across different FL? I plan on testing more systematically (i.e. with tripod and inanimate subjects) to see if I can get the focus even better. My concern is that the proper AF fine tune is inconsistent across the zoom range, or that the ideal AF fine tune adjustment is more than -20.

I came across this guide regarding AF fine tuning. Would you happen to have any other resources to recommend on this subject?

Again, thanks for your reply and advice  .

Reply   Reply with quote   Complain
rhlpetrus
Forum ProPosts: 23,086Gear list
Like?
Same with mine
In reply to gen123, Dec 19, 2012

It can be an amazing lens, but it needed +20 AF finetune to work alright. All my Nikkors work spot on with AF FT = OFF.

I will never, likely, buy 3rd party lenses again.

-- hide signature --

Renato. http://www.flickr.com/photos/rhlpedrosa/ OnExposure member http://www.onexposure.net/ Good shooting and good luck (after Ed Murrow)

 rhlpetrus's gear list:rhlpetrus's gear list
Nikon D7000 Nikon 1 V1 Nikon AF-S DX Nikkor 12-24mm f/4G ED-IF Nikon AF Nikkor 35mm f/2D Nikon AF-S Nikkor 70-300mm f/4.5-5.6G VR +3 more
Reply   Reply with quote   Complain
SteveL54
Senior MemberPosts: 2,610
Like?
Re: New Sigma 17-50 f2.8 OS on D7000. Backfocus?
In reply to gen123, Dec 19, 2012

gen123 wrote:

Do you have access to another lens in the 17-24mm focal length range? How about another camera body?

If not, then I suppose you could exchange for another new lens before your exchange privilege runs out and see how it behaves.

edit: Almost forgot to ask if you tried AF fine tuning?

Steve

I have a 18-105mm and a 18-200mm (which I want to sell and replace with the Sigma) that I can take some shots with. I've had a limited time shooting with them in the past (the 35mm f1.8 stays on most of the time), but I haven't noticed any obvious focusing issues. I'll definitely try both of them in the 18-24mm range now that you mention it though.

As for AF fine tuning on the Sigma 17-50... I played around with it last night in increments of 5, and seemed to get the sharpest (handheld) photos with a AF fine tune of -20. Strangely, even the shots at 50mm became slightly sharper with the adjustment. I wonder how much of a difference AF fine tune makes across different FL? I plan on testing more systematically (i.e. with tripod and inanimate subjects) to see if I can get the focus even better. My concern is that the proper AF fine tune is inconsistent across the zoom range, or that the ideal AF fine tune adjustment is more than -20.

I came across this guide regarding AF fine tuning. Would you happen to have any other resources to recommend on this subject?

Again, thanks for your reply and advice .

Those two lenses should give you an idea if the body is significantly contributing to the AF error at 18-24mm. They aren't f/2.8 but they should be f/3.5 to f/4.0 in that range. That should provide a DOF shallow enough for a comparison.

I imagine the Sigma has a distance scale on the lens. That helps speed the AF fine tuning against a live view benchmark distance. See the shortcut section near the bottom of my procedure. The one thing to watch out for when comparing image sharpness between settings is any mirror slap or vibration. Cheap tripods -- especially their heads -- can be suseptable to vibration. Using the lens distance scale to check AF takes this potential variable out of the picture -- so to speak.

If you want to consider another type of rig for AF tuning Mako has recommended this procedure:

http://forums.dpreview.com/forums/post/50199802

If your systematic testing indicates -20 is required (or more) then something definitely needs to go in for service or be replaced.

BTW I just checked the setting I have in my 18-200mm. It's -4. That value isn't perfect throughout it's entire range of focal length and focus distance, but I've been happy with its performance on my D300. My 28-70mm f/2.8 uses an AF tuning setting of -5.

Steve

Reply   Reply with quote   Complain
motobloat
Regular MemberPosts: 497
Like?
Re: New Sigma 17-50 f2.8 OS on D7000. Backfocus?
In reply to SteveL54, Dec 19, 2012

SteveL54 wrote:

If your systematic testing indicates -20 is required (or more) then something definitely needs to go in for service or be replaced.

BTW I just checked the setting I have in my 18-200mm. It's -4. That value isn't perfect throughout it's entire range of focal length and focus distance, but I've been happy with its performance on my D300. My 28-70mm f/2.8 uses an AF tuning setting of -5.

Steve

Indeed. Steve's result of -4 or -5 AF Fine Tune (AFFT) are what you would normally see with a "healthy" lens and camera combo.

If the problem is with the camera body, you'll tend to need AFFT values of more than what is allowed (i.e. less than -20 or greater than +20) and almost all lenses will need the same correction AND the correction will be in the same direction.

If the problem is with one lens, you'll only need AFFT on that lens, and it won't be correctable with -20 or +20, AND - this is crucial - all other lenses you test on the body don't need a severe AFFT correction. If you have this issue, Sigma will correct the lens for you under warranty, but you will probably want to send your camera body to them so that they can match the lens perfectly to your body (they usually ask for the body from what I've heard).

...

My D7000 is at the Nikon service center right now to get the backfocus issue fixed. It used to focus pretty well, but in the last few months it has become worse and worse, to the point where nine different lenses all needed -20 or more AFFT (except for one, which only needed -18 lol). So I sent it in a few weeks ago. The repair bill was $155. Still waiting for it to come back.

You can read other people's results after sending their camera in for repair all over the place, but here are a couple of threads that were easy to find.

http://forums.dpreview.com/forums/thread/3287410#forum-post-50108572
http://nikonrumors.com/forum/topic.php?id=5557

The D7000 back focus issue is a real thing - although there are certainly people who claim to have a back focus problem where none exists, and they just don't know how to set up their AF settings. If it is a camera issue, it can be corrected at the repair center with an adjustment of the AF Sensor and the mirror. Or at least, most repair tickets for back focus issues contain one or more of these items:

- ADJ MIRROR ANGLE
- ADJ AF UNIT
- ADJ AUTO FOCUS OPERATION
- RPL MIRROR DRIVING UNIT
- RPL AF GEAR SHAFT

Reply   Reply with quote   Complain
Keyboard shortcuts:
FForum MMy threads