Are subjects softer the further they are away? UPDATED: Lens sent for repair

thanks guys, will try tonight when I'm home.
I may also be able to aquire a tripod for the evening.
 
Ok just had this from Sigma

Dear Ben,

Thank you for your E-mail.

Is hard to tell from images provided as there is no depth to them so cannot see if there is focus deviation, please supply images at 45 degrees to the subject with depth in front and behind and forward in full size high res Jpegs so we can comment further.

Many thanks.

So what exactly do they want here?
The focus chart test?
 
Yea, something like this:
http://www.focustestchart.com/chart.html

Lining say, five items up along a diagonal plane with a ruler at 45degrees to them will help.
Focus on the middle item and see if it's sharp (or if the item in front/behind is sharper) will show front/back focus issues. The ruler is there to see how many mm (not really an issue) or cm (an issue!) the focus is out.
 
jamin100 said:
Ok just had this from Sigma

Dear Ben,

Thank you for your E-mail.

Is hard to tell from images provided as there is no depth to them so cannot see if there is focus deviation, please supply images at 45 degrees to the subject with depth in front and behind and forward in full size high res Jpegs so we can comment further.

Many thanks.

So what exactly do they want here?
The focus chart test?
Click to expand...

No. Not that friggin test chart linked above. It causes so many problems and is no good for a 200mm lens - you need to get some sensible real world distance in there, or it will show errors for sure.

They want a 3D subject, so they can see what was focused on, and also where the focus has actually landed, either in front or behind. Cerial box and ruler, shot from about 3-4m at 200mm would be fine. It's exactly the same as this expensive device http://spyder.datacolor.com/product-cb-spyderlenscal.php#

Focus calibration is a compromise, especially with zooms. Errors inevitably creep in at the extremes and if you're too close, problems will be exagerated. When you correct for them, focus is thrown out at normal shooting distances.
 
jamin100 said:
http://i776.photobucket.com/albums/yy50/jamin100/Screenshot2011-04-06at215455.png[/IMG]
Click to expand...

Well the focus square was on his face, but also partly on the bg - not by much admittedly, but enough to cause the AF system to hunt back and forth between the two planes - his face and the bg. You need to make sure that the AF rectangle falls completely over a single plane to achieve positive focus.
 
Last edited:
HoppyUK said:
No. Not that friggin test chart linked above. It causes so many problems and is no good for a 200mm lens - you need to get some sensible real world distance in there, or it will show errors for sure.

They want a 3D subject, so they can see what was focused on, and also where the focus has actually landed, either in front or behind. Cerial box and ruler, shot from about 3-4m at 200mm would be fine. It's exactly the same as this expensive device http://spyder.datacolor.com/product-cb-spyderlenscal.php#

Focus calibration is a compromise, especially with zooms. Errors inevitably creep in at the extremes and if you're too close, problems will be exagerated. When you correct for them, focus is thrown out at normal shooting distances.
Click to expand...


Ok thanks for the help.

So if i setup a scene like you posted above, where would be the best place to focus? on the ruler?
 
CT said:
Well the focus square was on his face, but also partly on the bg - not by much admittedly, but enough to cause the AF system to hunt back and forth between the two planes - his face and the bg. You need to make sure that the AF rectangle falls completely over a single plane to achieve positive focus.
Click to expand...

Thanks, I will keep that in mind
 
jamin100 said:
Ok thanks for the help.

So if i setup a scene like you posted above, where would be the best place to focus? on the ruler?
Click to expand...

No. You focus on the box.

When it is set up square to the camera, it is a perfectly flat surface with plenty of detail for the AF to get its teeth into. Zero room for error.

Then you view by zooming in on the ruler, and that tells you where the focus in fact is - in front or behind, and how far out. If your camera has a decent LCD, you can actually see well enough on that. Takes about two minutes once you get the hang of it.
 
HoppyUK said:
No. You focus on the box.

When it is set up square to the camera, it is a perfectly flat surface with plenty of detail for the AF to get its teeth into. Zero room for error.

Then you view by zooming in on the ruler, and that tells you where the focus in fact is - in front or behind, and how far out. If your camera has a decent LCD, you can actually see well enough on that. Takes about two minutes once you get the hang of it.
Click to expand...

Argh ok, so I need to be straight on to the cereal box, Because the shot above looks like its at an angle.

I have borrowed the tripod from work so hopefully I'll be able to perform this test tonight (kids permitting)
 
jamin100 said:
Argh ok, so I need to be straight on to the cereal box, Because the shot above looks like its at an angle.

I have borrowed the tripod from work so hopefully I'll be able to perform this test tonight (kids permitting)
Click to expand...

Haha! Yes it's at an angle, to show what the set up looks like ;) You shoot it square on.
 
jamin100 said:
Ok here's another test. I dont have a tripod so can only hand hold but I made sure a was as steady as possible. ISO was 200, F2.8 200mm and shutter speed was very high
Click to expand...

A little camera shake perhaps, holding technique a bit off perhaps.......:shrug:

I was having a similar problem................
http://www.talkphotography.co.uk/forums/showthread.php?t=309640

I have improved my camera holding technique, it seems to be working for me ;)

I've also invested in a very sturdy Manfrotto monopod. Now I'm looking for a wheelbarrow to carry it in. :thumbs:

In all seriousness.....I returned a 10-24 Nikkor due to soft focus. It was'nt the lens it was my technique.

I have just taken delivery of a second 10-24 and it's fine, as no doubt the first one was.....It was me all along.

D in W
 
ok guys hows this?
Lens was mounted on a tripod, shot at 200mm F2.8. fired off a timer to avoid camera shake.

No processing has be done, just exported to jpg in lightroom. Focus is on the BOLD word Freeing

20110409-DSC_2045.jpg
 
I took a shot across to the far side of the Thames today, from Essex into Kent. It was a warm day and there was a haze in the air. The sun was providing backlighting for this shot....

Full frame, unedited.....

20110410_105832_4016_LR.jpg


100% crop, unedited....

20110410_105832_4016_LR-2.jpg


There is no doubt that distance can cause a problem if the atmosphere is bubbling with heat, or just filthy, but this long distance example seems OK to me.


As for the AF calibration test in the post above, I would want to examine for critical sharpness at 100%. Is the shot above a crop or the whole frame? With respect to the ruler, used for judging the sharpest point of focus, I would place it immediately next to the box, touching, and be very certain about which figure on the ruler was lined up with the front of the box. It's hard to tell from your shot above just which figure on the ruler ought to be sharpest. As well as using the self timer you might as well add mirror lockup into the equation too.

Here's my version of that setup. The 13" mark is lined up with the front surface of the box and is my reference mark for judging focus position. It is hard to make a mistake in that because the ruler is resting on the edge of the box. Also note that my focus target is pretty well squared up with the camera, it is well lit and contrasty, with good detail for the AF to detect, and there is ample space around the active AF point to ensure the AF point sees nothing but the target it is to focus on.

20110412_010008_000.jpg


Closer up....

20110412_010137_000.jpg
 
Last edited:
jamin100 said:
ok guys hows this?
Lens was mounted on a tripod, shot at 200mm F2.8. fired off a timer to avoid camera shake.

No processing has be done, just exported to jpg in lightroom. Focus is on the BOLD word Freeing

20110409-DSC_2045.jpg
Click to expand...

If that small red dot on the rule is in line with the front of the box, then you've got a bit of back focus there. Do the same shot using live view AF with the magnification function, which should nail it absolutely, and compare.

That's assuming that the test is kosha. What was the distance? It looks a bit close, but should be okay. As Tim says, ideally the rule should be right next to the box to guard against field curvature but that's more of an issue with shorter lenses. Also, when you get very critical, the box must be dead flat and not bulging in the middle.

But we're not being super-critical here. That's good enought to show that the lens is indeed out by a signifcant degree.
 
ok sorry this has taken a while but here are my new images.
The top of the ruler was aligned directly with the front of the box. The focus was on the bold words "freeing the". Taken on a tripod with remote shutter release. taken from 10 feet away.

This is probably the worst example at 70mm f2,8 which I think shows that it is focused on the box behind.

7028100crop.png
 
Definite back focusing. I assume the D90 has AF microadjustment (or whatever Nikon calls it) so maybe you can effect your own fix, but with zoom lenses you might find the adjustment required varies across the zoom range. Then, unless you can find an acceptable compromise you may need to send your gear in for manufacturer adjustment.
 
well I finally got round to sending the lens off to sigma yesterday and have had the repair quote today :(

Apparently it needs:

REPLACE MOUNT
ADJUST CIRCUIT
CHECK OPTIC
RESET, CHECK & TEST

All for the bargin price of £130 :(

Hopefully it'll come back all better though :)
 
£131.99 is the standard service tariff for the 70-200 (unless its a 5 min job then they'll quote under that). good news is though that those parts, labour and shipping back to you would probably cost you more. i had most of the internals on my 120-300 replaced as the previous owner had bent them in a drop for the measily price of £167.

i'll race you though, i just approved some work this afternoon on one of our canon mount 70-200's :D
 
My documentation doesn't state an estimate just

REPAIR IS SUBJECT TO AVAILABILITY OF PARTS
 
neil_g said:
once youve sent the acceptance sheet back just drop them a quick email asking for an eta :)
Click to expand...

Been told to expect 2 to 3 weeks, so should be back in time for half term
 
neil_g said:
mines back! :D
Click to expand...

Well guess what :)



... I still dont have the lens back :(
I was hoping to get this back before we go to cosford airshow this weekend so i have just emailed my contact at sigma but have been told that;

I have spoken to the technician dealing with this and the circuit that was pre-written especially for you, did not bring the lens in enough to be within tolerances for this reason we have had to retake images and Japan will make another circuit for us.​

So it looks like I have a few more weeks of waiting and the longest reach i'll have at cosford is 75mm :(
 
New Update:
Spoke to sigma yesterday and the new circuits from Japan didnt work so they have to get some more sent :(

Still waiting
 
You must log in or register to reply here.
Back
Top