D7000 or 70-300 VR focus problem

[Goatch]

I thought I would post up a couple of pics of an issue that I discovered with my 70-300VR , I've always been a bit suss on this lens since I've bought it and it wasn't until I stuck it on the front of my brand new D7000 that I realised it had a focus issue which seemed to be there right through the aperture range , well from 4.5 up to 11 that I tested anyway , I've been hesitant to use to Autofocus fine tune up till now but have been a bit stunned with the results of a little tweaking , It may well be that I have become much more attuned to whether a focus point is pin sharp too , both shots are straight out of the camera , RAW cropped to 50% and converted to JPEG , the first one is how the focus point was initially and the second is with -10 Fine tune applied , there is a big difference , I also tried the same test with my 24-70 and 70-200 VR11 but found no issues that I could see , the focus point in both shots is the middle E in Amberley , have others found this 9196591966
Is that what is called Back Focusing ?

[Film Street]

It was certainly rear focusing there. Looks much better now.

[WhoDo]

Yep, a bit of back focus in the lens Goatch. I had to do the same with my 55-300 on the D7000 and the results were instantaneous, although I only needed -3 on the fine tune for that one. None of the other lenses required any alteration that I could see. Maybe it's a function of ultra-zooms that they perhaps go out of alignment or something after shipping? :confused013

[Goatch]

Yeah Waz you might be on to something there , it's funny after doing that fine tune the lens is still no better at full zoom , it may be even a little worse , very hard to judge though , but it has definitely improved it out to around 250mm

[WhoDo]

... it's funny after doing that fine tune the lens is still no better at full zoom , it may be even a little worse , very hard to judge though , but it has definitely improved it out to around 250mm
I think the super zooms are even more prone to soft focus at the extremities than a short zoom, but I'm no expert. That last 50mm between 250mm and 300mm is usually a problem in my experience, such as it is. :confused013

[arthurking83]

I don't think it a problem with shipping, it's probably just that particular lens!
I had similar issues with my 80-200/2.8, when I had that lens, and still have backfocus on the Tammy 28-75/2.8 too, but it's not worth the hassle to AF fine tune it.

Some thing to watch for when doing Focus fine tune via the camera is that:

1. don't test the lens with the focus distance set too close, unless you do more shooting up close and less on distant subjects.
Try and get a gauge of roughly how far the average shot would be when using this lens, and make sure it's focusing right at that focused distance.
It's not always the case, but with some lenses the focus distance can have an bearing on how accurate focusing can be.

I noticed this with the 80-200/2.8 where it was annoyingly inaccurate at anything less than about 4 or 5m away, but seemed to be more accurate if the subject was 20m or more away.
It wasn't only due to the increased DOF with increasing distance, but the actual plane of focus was more accurately placed as it focused further away.
The Tammy doesn't do this, tho, but it's inaccurate only at the very long end of the focal length scale.

which leads to point number 2!
When AF fine tuning, you should also take into account the other focal lengths.. especially the shorter end of the scale.

I'd reckon it would be better to have the lens adjusted to perform better natively, and only use AF fine tune as an emergency procedure, and/or as the means of proof that the lens needs adjustment.

[old dog]

I`d like to fine tune my 70-300vr as well......but........I havn`t because......you have to set it up at least 50 times the focal length to test. That equates to 15 metres if set at 300mm FC. I cant set the camera up that far away. I was using the test chart on the net that was posted a little while ago on Flickr. Apart from that I will have to print off a different chart that I can use or just do the test like Goach has done. Add lib until I am satisfied. Any thoughts.

[Goatch]

Mmm, I think I might revisit this today as my test/tune was done at about 5mtrs , I might try pushing it out to 15-20 mtrs as Arthur has suggested and see how we go , I also think I will def go down the path of getting the lens tuned by Nikon , Cheers for the input guys , learning is always good .

[arthurking83]

..... you have to set it up at least 50 times the focal length to test. That equates to 15 metres if set at 300mm FC. I cant set the camera up that far away. I was using the test chart on the net that was posted a little while ago on Flickr. .....

Firstly! Forget about test charts and just use normal everyday run'o'th mill subject matter.

Test charts have their place in photography, but what you really want is just to quickly test to see how the lens works(or the camera/lens combo) works for ya in real life.

Don't get me wrong .. I have one or two test charts that I sometimes refer back too, but they're not vital or even helpful in any special way to sort out something simple such as focus fine tune.

All you need is a subject that will fill the entire focus square at a reasonable distance away.
So if you test at 15m, then maybe a road sign, eg. a speed limit sign or stop sign or whatever, but making sure that there are other object close by fore and aft of the plane of focus, so that you can see how the DOF applies for that particular situation.
Fence posts or whatever else can also do the trick.

I have yet to be convinced of this 50x the focal length theory too.
I just tested my lenses(so far only two of them) at focus distances that matter!
For the 80-200/2.8 at 200mm(where the issue was) I just tested at approximately 5m, as this was a reasonable close in range for portrait shooting.
That is, a good distance where detail sharpness can impact on the final image if it's spot on or not. At 5m, there is clearly a difference between the sharpness of an eye or an ear and it can be visible in the final image.

Going much further than that distance, eg. 10m or more, DOF will usually take care of any slight inaccuracies of the AF system.
So that, at 10m the DOF is naturally deeper to begin with, so the difference between the ear and the eye being sharply rendered is much less so that if shot at 5m focus distance.
Also at 10m or more, the ability to differentiate between a sharp ear or eye is much harder to perceive on the final image compared to the closer in 5m shot.

With the 80-200/2.8 is eventually turned off any AF fine tune simply because it became annoying at focal lengths below 200mm(where the issue was) and just had to force myself not to shoot very close in subjects.
Bummer, but the replacement lens I eventually got(Tammy 70-200) doesn't have these issues, so the problem was solved the hard way.
Would have loved to keep that lens too(I'm more of a lens buyer, not seller!! :p) as it had some awesome ability, but it's focusing system(AF-D), and not the focus issues was the reason why I sold it off.
Otherwise I'd have kept it and got it properly re set by Nikon.
(now I'm a massive fan of AF-S type lenses only!!)

With the Tammy 28-75mm which I still have, I just put up with it. That is, it back focuses on my D300. I know this and it's easy to work with.
With this lens, you can force the lens against the cameras focus motor, which you shouldn't do, and I don't do. But as the focus ring still has the ability to rotate, there is a very slight amount of slack in the focus drive system. I found that this amount of backlash is exactly how far off the lens is off it's proper focus mark.
A very light twist of the focus ring in the right direction on the 28-75mm, to take up the slack, gets me pretty spot on focus.
On the Nikon 80-200mm the focus ring is locked hard! .. so you can't use the same method.

Once again, I should have tried to get it sorted out properly, so that I don't have to 'jump through hoops' with the Tammy 28-75, but the reality is that it's not a major problem.
And now with the new 24-70/2.8 VC lens released, it's time to update that lens anyhow .. whoot .. finally with an AF-S type focusing system too.
I contemplated getting either a Nikon 24-70, or the older 28-70, but they're so damned expensive(for my purposes).

[old dog]

All you need is a subject that will fill the entire focus square at a reasonable distance away.
So if you test at 15m, then maybe a road sign, eg. a speed limit sign or stop sign or whatever, but making sure that there are other object close by fore and aft of the plane of focus, so that you can see how the DOF applies for that particular situation.
Fence posts or whatever else can also do the trick.



thanks Arthur for the explanation. I`ll try that out for the 70-30vr and see what I come up with.