Back/front-focus

If you would like to post, you'll need to register. Note that if you have a BCG store account, you'll need a new, separate account here (we keep the two sites separate for security purposes).

S

Studioff

New member
Member's gallery
Anyone out there who knows the answer to this?
For example, I have a Nikon 200-500 that I suspect is not focusing rigt. I have run some tests and need to run some more but I think I see aconstant frontfocus issue even if I adjust the fine tuning in my D500. I ran several auto fine tune tests and got results in the -11 to -6 area and I have ran tests manually against a "autofocus-calibration-target" (you know, the ones with a ruler) but I think I see frontfocus issues no matter how I adjust it.

Two questions.
1. If I change the fine tuning, does the D500 try to focus by that number automatically or do I need to restart the camera?
2. When having a frontfocus issue that results in maybe 0,5-1cm focuserror against a target some meters away, does the frontfocuserror-marginal increase in length on a longer distance. Let's say I'm shooting something twice the distance from the distance I used when testing against the target, does the error distance increase to maybe 2-3 cm?
 
Last edited:
Member's gallery
Once you determine the AFFT value, you need to save the value in the camera menu item for lens fine tuning. Then, whenever that lens is mounted onto the camera, the camera will use that value to adjust focus.

When determining the needed AFFT value, you should set the target at the distance you normally shoot that particular lens. If you you typically shoot the 200-500 @ 500mm.....then you should test focusing for 500mm...... Also be sure you are testing wide open, on a tripod, fully locked down and with no wind and plenty of light.
 
Member's gallery
Studioff said:
2. When having a frontfocus issue that results in maybe 0,5-1cm focuserror against a target some meters away, does the frontfocuserror-marginal increase in length on a longer distance. Let's say I'm shooting something twice the distance from the distance I used when testing against the target, does the error distance increase to maybe 2-3 cm?
Click to expand...
This is an interesting question and I'd like to know the answer too! I think what you're asking is, once you've got AFFT dialed in, does it compensate for front/back focus whether you're shooting at 20 yards or 40 yards or 80 yards?
 
Member's gallery
MikeJ said:
This is an interesting question and I'd like to know the answer too! I think what you're asking is, once you've got AFFT dialed in, does it compensate for front/back focus whether you're shooting at 20 yards or 40 yards or 80 yards?
Click to expand...
I don't believe the camera "compensates" for distance. That is the reason you should test at the distance you typically use that lens. The camera uses the AFFT value that the user entered.

Watch Steve Perry's video HERE....
 
Since the fine tune adjusts the distance between the mirror and the focusing sensor it shouldn't vary by distance but it should vary by focal length. You could try shooting the target at different distances using the viewfinder and again in live view which shouldn't be impacted at all by the adjustment.
 
A lot of people like the dot tune method which there are plenty of videos on. Or even better go mirrorless, problem solved!
 
Member's gallery
bleirer said:
Since the fine tune adjusts the distance between the mirror and the focusing sensor it shouldn't vary by distance but it should vary by focal length. You could try shooting the target at different distances using the viewfinder and again in live view which shouldn't be impacted at all by the adjustment.
Click to expand...
True. Some cameras allow the user to set multiple AFFT values for different focal lengths of a zoom lens.
 
Though this video says distance can make a difference so now I'm not sure.

 
Last edited:
Member's gallery
bleirer said:
Though this video says distance can make a difference so now I'm not sure.

Click to expand...
It really depends on the lens design whether distance is a factor and by how much. The big primes seem less affected, the zooms a little more so.
Were either the lens or body serviced (AF adjusted) by Nikon or an authorized service center?
 
Member's gallery
If your camera is front focusing then the adjusted values should be + I believe. This is from Nikon....

Make the adjustment in the appropriate direction, (+) to move the focal point away from the camera and (-) to move the focal point towards the camera depending on the results of your focus test.

If your focus is exactly correct you should have as much DOF in front of the focal point as behind the focal point. Normally you focus fine tune at a close distance and maximum aperture to get a small DOF. A 500mm at f5.6 focused at 35ft has a DOF of about 4 inches, this is an amount that you should be able to observe on a ruler type calibration target. At 75ft the DOF becomes 18" which is harder to see consistently on a ruler type target. If your AF is correct I would expect the lens to provide 50% of the DOF in front and in back of the focal point. I am not sure this is exactly accurate but if your focal point is 1 inch in front at 35 ft then your DOF would have 3 inches in front and 1 inch behind. At 75ft the DOF should be 13.5 inches in front and 4.5 inches behind, so your focus point would be off by 4.5 inches.

I do know that focus consistency can also be an issue. The better the lens the better the focus consistency. My 70-200 f2.8 FL E series produces consistent results when fine tuning no matter what method I try. My 200-500 f5.6 provides fairly consistent results when fine tuning, but my 70-300 is much harder to AF fine tune consistently.
 
Member's gallery
Thanks for all the replies!

I hope you understand me in my questions.

1. I wonder if when testing and adjusting the -20 to +20 value, if the camera directly focuses by that new number or if I have to restart the camera for the adjustment to take action.

2. I wonder if the lens frontfocuses by lets say a centimeter when I test at lets say 10 meters and then I shoot in reallife and the distance is 25 meters, does the errorgap get longer oor does it stay the same at 1cm frontfocus? The reason why I'm asking is that this is what I think I see when shooting. With this lens it's impossible to get a decent shot at subjects far away. A small bird at 6m away looks..not ok..but close.

The lens or body hasn't been to a nikon servicecenter but if the following tests will show the same unadjustable frontfocus I will have to contact them. As someone suggested I can do testing on this too but I wonder if anyone here knows for a fact what happens at subjects at a different distance.
 
Member's gallery
Studioff said:
Thanks for all the replies!

I hope you understand me in my questions.

1. I wonder if when testing and adjusting the -20 to +20 value, if the camera directly focuses by that new number or if I have to restart the camera for the adjustment to take action.
Click to expand...
No restart necessary
2. I wonder if the lens frontfocuses by lets say a centimeter when I test at lets say 10 meters and then I shoot in reallife and the distance is 25 meters, does the errorgap get longer oor does it stay the same at 1cm frontfocus? The reason why I'm asking is that this is what I think I see when shooting. With this lens it's impossible to get a decent shot at subjects far away. A small bird at 6m away looks..not ok..but close.
Click to expand...
Error increases with distance.
 
Member's gallery
The reason I ask about the service is that the alignment rigs have a known error compared to assembly line alignment. Any lens that has AF adjusted using one of the service rigs will get 7 AFFT units of front focus as compared to the factory.
Something in the last OP post caught my eye, are you sure you are having a focus issue? Could you be seeing atmospheric effects and maybe lens hood induced atmospherics as Steve verified last month?
 
Member's gallery
Some people report success with the auto fine tune function in the D500. I could never get it to produce useable results. Also with the 200-500 don't expect perfectly spot-on, repeatable results. You should be able to get it close enough to keep the target within DOF but not necessarily dead center in DOF. You should be shooting multiple frames at each adjustment value and taking the average. Due to the lack of repeatability if you use single values you will chase your tail forever w/out a solution.
 
Member's gallery
Warren D said:
The reason I ask about the service is that the alignment rigs have a known error compared to assembly line alignment. Any lens that has AF adjusted using one of the service rigs will get 7 AFFT units of front focus as compared to the factory.
Something in the last OP post caught my eye, are you sure you are having a focus issue? Could you be seeing atmospheric effects and maybe lens hood induced atmospherics as Steve verified last month?
Click to expand...
I'm sure several times has been to atmospherics. Absolutely! But I can never get a decent shot with this lens at greater distance.
 
Member's gallery
NorthernFocus said:
Some people report success with the auto fine tune function in the D500. I could never get it to produce useable results. Also with the 200-500 don't expect perfectly spot-on, repeatable results. You should be able to get it close enough to keep the target within DOF but not necessarily dead center in DOF. You should be shooting multiple frames at each adjustment value and taking the average. Due to the lack of repeatability if you use single values you will chase your tail forever w/out a solution.
Click to expand...

NorthernFocus said:
No restart necessary

Error increases with distance.
Click to expand...
Thanks! That is exactly what I was after!
 
Member's gallery
NorthernFocus said:
Some people report success with the auto fine tune function in the D500. I could never get it to produce useable results. Also with the 200-500 don't expect perfectly spot-on, repeatable results. You should be able to get it close enough to keep the target within DOF but not necessarily dead center in DOF. You should be shooting multiple frames at each adjustment value and taking the average. Due to the lack of repeatability if you use single values you will chase your tail forever w/out a solution.
Click to expand...

Yes the auto fine tune feels like quite random and I have tried that as well...at the right distance and about 12-15 times. I have gotten close with my Tamron 150-600 G2 and ended upon +2 after auto fine tune AND going through the whole -20 to +20 with a total of 41 shots and then checking the difference in the shots. I could narrow it down to +2 and I hope it works. My problem with the Tamron is another one...it's not that reliable when focusing. Lots of out of focus shots but sometimes good ones. The 200-500 feels more accurate BUT...is it focusing right...

I have tried the auto fine tune and different tests and yes maybe I have stared at too many tests now or I'm chasing my tail ...or maybe it can't be calibrated within the -20 to +20 but needs hardware calibration. Right now I don't know. I have gone through -20 to +20 with it on both my D500 and D7500 (2 steps between shots though) and ...yeah now I don't know.

If anyone has "sparetime" and likes to try to help me you can download the jpegs here:


Filenamnes D500 (1) is -20 and goes to D500 (21) which is +20 and the samewith D7500...and it's a homemade target but I have checked so everything should be quite good aligned.
 
Member's gallery
Have you tried focusing with Live View and taking a shot and comparing it to a viewfinder focused shot? Live View focuses off of the sensor and should be accurate. Just use a flat target for starters.
 
Member's gallery
Although it is optimum to test at the same range you expect to be shooting it is often not practical. Lensalign/focustune recommend a test range of 25x focal length. So 40 ft for 500mm. That's what I've always used and results translated well to field. My recommended procedure is as follows:

Target at 25x FL.
Camera on tripod, mirror up mode, delayed release set to 3s, AF-S, single focus point, VR off, 1/1000s, max aperture, manual ISO, jpeg, monochrome
Shoot five frames at each step, -20, -15, -10...., +20. Manually offset focus between each shot and refocus with AF.
Identify best then repeat test at 2 step intervals from -5 to +5 either side of best. e.g. if best result is +10 then repeat test from +5 to +15, five shots each step at 2 step intervals i.e.-5, -3, -1...,+5.

While I have the lensalign target with ruler I don't find it all that helpful. Using a single target with the above procedure works better for me. For longer range test and/or for a quick and dirty test in the field I've done this using a road sign as a target and basically going through an abbreviated version of the same test at 100ft range or so(500mm). With the road sign test there's no need to be perfectly precise setting up perpendicular to the sign.

It is important to have as steady lighting as possible when doing these tests because you are basically evaluating which image has highest contrast. Varying light make that difficult. Even with the Focustune software variable lighting can make it difficult because that's what the software is doing is comparing contrast between shots.
 
Member's gallery
Studioff said:
Yes the auto fine tune feels like quite random and I have tried that as well...at the right distance and about 12-15 times. I have gotten close with my Tamron 150-600 G2 and ended upon +2 after auto fine tune AND going through the whole -20 to +20
Click to expand...
Had similar problem with D7500 and Tamron 150-600mm G2. Far distance focus was horrible/inconsistent. Tried auto fine focus on the camera, Focal software, all with variable results, no consistency.
Sent both the tami and nikon in for maintenance, nothing wrong found on the Tamron, and it was the Nikon maintenance that finally got it working, although nothing noted from Nikon with the camera return. Was then able to get some consistency. I still ended up setting up my own focus system, took lots of picts at different fine focus adjustments comparing them, to finally get where I feel its as good as its going to get.
 
You must log in or register to reply here.
Back
Top