Willie73478
Well-known member
I probably would return the camera to Best Buy except the 18-140 lens that came with it is spot on with zero AF fine tuning required, and sharp, too.
-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
Back focus in artifical light
- Thread starter Willie73478
- Start date
-
- Tags
- nikon d7000
Leonard Shepherd
Forum Pro
Since when was 2 typing errors "handfuls"?
Leonard Shepherd
Forum Pro
I have never said there are no defects in any products. On the other hand I have never had artificial light focus issues using Nikon AF in 16 years using Nikon.
I agree, as is common on forums, there is often very much talk about very little.
I recall Henry Posner of B&H posting at the time of the D200 minor banding issue "there is a lot more talk about banding than there is banding to talk about"
For those saying they would like to see a repeat test with an 80a filter why not first find a camera and lens that, with good technique, appears to mis focus in artificial light, and then repeat with an 80a filter? For anyone owning an SB-900 when supplied new there is a resin filter suitable for this type of test.
My D710, D750, D800 and D810 do not mis focus in artificial light and my previous 6 different Nikon AF bodies did not mis focus either. This is not the same as saying an occasional body cannot have a focus issue which it should not have.
If equipment is defective when new my logic is, as I have paid good money to buy it, I expect good performance or an exchange of a faulty item.
MisterHairy
Senior Member
If I am being generous to you, I can see six, straight away. However, this is not a "being able to adequately convey one's message" discussion, nor is it even a discussion about being a poorly programmed auto-generator of text, so I suspect that this is taking more than a slight turn off topic.
And in case you do have the facility to understand, I did not simply call you a "rogue", as in a scoundrel, I called you a "rogue Turing test", as in a computerised conversation partner which is not functioning as intended. That seems fair, given your commentary to date.
Last edited:
Willie73478
Well-known member
I redid the tests with flat peices of paper with the letter A of various fonts glued to the batteries and got the exact same results under artificial light.
I also tried stopping down and the 35mm f/1.8 and the target was still out of focus at f/2.8, f/4 and even f/5. On the other hand the 18-140 lens was spot on wide open at 35mm and f/4.2.
I also tried stopping down and the 35mm f/1.8 and the target was still out of focus at f/2.8, f/4 and even f/5. On the other hand the 18-140 lens was spot on wide open at 35mm and f/4.2.
Last edited:
Willie73478
Well-known member
OK, so yesterday I spoke to Nikon Service. They advised me to reset the camera using the two green buttons (+/- and + with the magnifying glass). The results in artificial light looked pretty much the same, although I was using pieces of paper with A's of various fonts taped to the batteries instead of the batteries themselves as subjects.
Then this morning something interesting happened. It is sunny outside, where I think it was probably cloudy during previous "natural light" tests. Now it appears that -10 AF seems to be acceptable for both the 35 and 50 mm lens in both artificial and natural sunlight (indirect through low-E glass).
Cropped images:
35mm f/1.8 artificial AF=-10
35mm f/1.8 AF=10
The ideal was probably a little more than -10 for artificial light and a little less for natural light, but this seems to be a good enough compromise.
- Messages
- 30,186
- Solutions
- 112
- Reaction score
- 12,428
I like that setup. Nice tip regards nullifying the issue of shooting rounded targets.
In practical terms...I think that works well and more along the lines of what I would expect in the case of those primes. May be no need for service. Your everyday shooting results may not practically benefit from a trip to Nikon Service.Then this morning something interesting happened. It is sunny outside, where I think it was probably cloudy during previous "natural light" tests. Now it appears that -10 AF seems to be acceptable for both the 35 and 50 mm lens in both artificial and natural sunlight (indirect through low-E glass).
Cropped images:
35mm f/1.8 artificial AF=-10
35mm f/1.8 AF=10
The ideal was probably a little more than -10 for artificial light and a little less for natural light, but this seems to be a good enough compromise.
--
My opinions are my own and not those of DPR or its administration. They carry no 'special' value (except to me and Lacie of course)
Retzius
Leading Member
This is the same situation I was in. The 35 1.8 works fine in bright daylight... kind of defeats the purpose of having a fast prime lens. In this situation you might as well just use a zoom and have the convenience of multiple focal lengths if light levels are not an issue.
- Messages
- 30,186
- Solutions
- 112
- Reaction score
- 12,428
I never said you did. Why would you suggest I did by reply to my post here?
Marianne Oelund
Veteran Member
First, I would like to extend my thanks to you, for taking the time to document your testing. Please excuse and ignore those who are unable to comprehend its implications.OK, so yesterday I spoke to Nikon Service. They advised me to reset the camera using the two green buttons (+/- and + with the magnifying glass). The results in artificial light looked pretty much the same, although I was using pieces of paper with A's of various fonts taped to the batteries instead of the batteries themselves as subjects.
Then this morning something interesting happened. It is sunny outside, where I think it was probably cloudy during previous "natural light" tests. Now it appears that -10 AF seems to be acceptable for both the 35 and 50 mm lens in both artificial and natural sunlight (indirect through low-E glass).
Cropped images:
35mm f/1.8 artificial AF=-10
35mm f/1.8 AF=10
The ideal was probably a little more than -10 for artificial light and a little less for natural light, but this seems to be a good enough compromise.
The issue you are studying is well known, and is principally due to design deficiencies of the AF module, which employs single-element molded-plastic lenses. CA for these lenses is rather high, causing focus shifts with color. The effect on AF accuracy is predictable and consistent.
On the other hand, how the camera may or may not compensate for the AF module's color shift is not as well understood, especially given the plethora of camera models in the field. Since recent Nikon models are equipped with color-sensitive metering, there is considerable latitude for a designer to add color compensation for the AF.
Given that, you might like to look for the effect of WB settings on AF. For example, you could try biasing auto WB with a blue/amber shift, or try the incandescent and daylight settings instead of using auto, or even try a range of manual color temperature settings, to see what the effect on AF accuracy might be.
--
Source credit: Prov 2:6
- Marianne
Willie73478
Well-known member
First, I would like to extend my thanks to you, for taking the time to document your testing. Please excuse and ignore those who are unable to comprehend its implications.OK, so yesterday I spoke to Nikon Service. They advised me to reset the camera using the two green buttons (+/- and + with the magnifying glass). The results in artificial light looked pretty much the same, although I was using pieces of paper with A's of various fonts taped to the batteries instead of the batteries themselves as subjects.
Then this morning something interesting happened. It is sunny outside, where I think it was probably cloudy during previous "natural light" tests. Now it appears that -10 AF seems to be acceptable for both the 35 and 50 mm lens in both artificial and natural sunlight (indirect through low-E glass).
Cropped images:
35mm f/1.8 artificial AF=-10
35mm f/1.8 AF=10
The ideal was probably a little more than -10 for artificial light and a little less for natural light, but this seems to be a good enough compromise.
The issue you are studying is well known, and is principally due to design deficiencies of the AF module, which employs single-element molded-plastic lenses. CA for these lenses is rather high, causing focus shifts with color. The effect on AF accuracy is predictable and consistent.
On the other hand, how the camera may or may not compensate for the AF module's color shift is not as well understood, especially given the plethora of camera models in the field. Since recent Nikon models are equipped with color-sensitive metering, there is considerable latitude for a designer to add color compensation for the AF.
Given that, you might like to look for the effect of WB settings on AF. For example, you could try biasing auto WB with a blue/amber shift, or try the incandescent and daylight settings instead of using auto, or even try a range of manual color temperature settings, to see what the effect on AF accuracy might be.
--
Source credit: Prov 2:6
- Marianne
Thanks for the feedback. I just checked the AF in artificial light with many different WB settings and I don't see any difference in the focussing,
Leonard Shepherd
Forum Pro
What are you like in a bad mood?
The first requirement of good testing is to eliminate other potential variables.
I pointed out the test was not done enough to a good enough standard to eliminate potential variable.
If you object to fair comment, as you seem to do - tough.
Leonard Shepherd
Forum Pro
I am pleased you redid the test - and learned you can get different results in different conditions.OK, so yesterday I spoke to Nikon Service. (snipped)
Then this morning something interesting happened. It is sunny outside, where I think it was probably cloudy during previous "natural light" tests. Now it appears that -10 AF seems to be acceptable for both the 35 and 50 mm lens in both artificial and natural sunlight (indirect through low-E glass).
Cropped images:
35mm f/1.8 artificial AF=-10
35mm f/1.8 AF=10
As Nikon say in your instruction book when AF cannot acquire good focus (as to some extent you have proved) consider manual focus for a better result.
Back to this test - sorry it is still not as good as it could be for a fully acceptable result.
There is quite a difference in the angle of the front horizontal line between the 2 images. With a good tripod that should not be there and ideally the line behind the batteries should be level.
An important aspect of testing with any target it to vary the focus distance to keep the target a constant size in the viewfinder. Did you vary focus distance or did you crop the wide angle result more?
We do not know where you aimed the AF point or how much you cropped the image.
At some point subject detail becomes too small for reliable AF as your camera instruction book explains. Whether the A is too small in the viewfinder for reliable AF is unclear.
If you send me your email by private message I can send you a target to be used at a specific multiple of focal length which incorporates resolution figures.
If this gives close to nil instead of -10 it may be your camera AF is having trouble accurately focussing on whichever letter A you used.
--
Leonard Shepherd
Is where you are in photography best defined by your recent photos or the equipment you own?
This does not make sense Leonard. If what you state is true, why is it that the outer non-x type sensor cannot grab focus when aimed at a single vertical target but as soon as one rotates (roll) the camera ca. 15 degrees the same sensor grabs focus?
Another thing, those "A"'s in the OP's test are good enough to attract the AF sensors' attention. This talk about AF needing very specific targets to work is highly overrated from your end.
Another thing, those "A"'s in the OP's test are good enough to attract the AF sensors' attention. This talk about AF needing very specific targets to work is highly overrated from your end.
- Messages
- 30,186
- Solutions
- 112
- Reaction score
- 12,428
Those other variables you speak of have no impact on this particular issue/test setup. In that regard, your input was irrelevant to the OP's problem.
Again, of no relevance in this case. Your input here is like someone asking you how best to fix a tire with a hole in it from a nail...and you explain the need to have proper air pressure to maximize gas mileage when testing to determine how far the car will go on a tank of gas.
--
My opinions are my own and not those of DPR or its administration. They carry no 'special' value (except to me and Lacie of course)
Last edited:
MisterHairy
Senior Member
Don't make me angry. You wouldn't like me when I'm angry. (You may need to look that one up.)
No. One first needs to identify the variables and then prove which ones can influence the test results. The use of controls can help with this. One never needs to remove ALL variables unless one has infinite resources and wishes to waste them. Your failure here is in identifying which variables are important and which are not.
VariableS. And while that may be true, I was dismissing some of your points on the basis that you do not even have the clarity of thought to express them coherently in words. The above sentence continues in that fashion. To further pick apart the pseudo technical aspects of your commentary is unfair as this has been done time and time again and with great aplomb by other correspondents. You just don't know what you don't know.
When you make one, let me know and we'll see.
By the way, selective quoting and responding to points which I was not making is rather poor form. I see that I am not the only poster to whom you have done this. It does not help your cause, such as it is.
[QUOTE="Marianne Oelund wrote... First, I would like to extend my thanks to you, for taking the time to document your testing. Please excuse and ignore those who are unable to comprehend its implications.
The issue you are studying is well known, and is principally due to design deficiencies of the AF module, which employs single-element molded-plastic lenses. CA for these lenses is rather high, causing focus shifts with color. The effect on AF accuracy is predictable and consistent.
On the other hand, how the camera may or may not compensate for the AF module's color shift is not as well understood, especially given the plethora of camera models in the field. Since recent Nikon models are equipped with color-sensitive metering, there is considerable latitude for a designer to add color compensation for the AF.
Given that, you might like to look for the effect of WB settings on AF. For example, you could try biasing auto WB with a blue/amber shift, or try the incandescent and daylight settings instead of using auto, or even try a range of manual color temperature settings, to see what the effect on AF accuracy might be.
The issue you are studying is well known, and is principally due to design deficiencies of the AF module, which employs single-element molded-plastic lenses. CA for these lenses is rather high, causing focus shifts with color. The effect on AF accuracy is predictable and consistent.
On the other hand, how the camera may or may not compensate for the AF module's color shift is not as well understood, especially given the plethora of camera models in the field. Since recent Nikon models are equipped with color-sensitive metering, there is considerable latitude for a designer to add color compensation for the AF.
Given that, you might like to look for the effect of WB settings on AF. For example, you could try biasing auto WB with a blue/amber shift, or try the incandescent and daylight settings instead of using auto, or even try a range of manual color temperature settings, to see what the effect on AF accuracy might be.
MisterHairy
Senior Member
Last edited:
Deliverator
Active member
Thanks for this thread. I've been experiencing some odd backfocus problems with my Sigma 17-70 C indoors in room light that don't show up outdoors in daylight. To a much lesser extent, I see it with my 35 f/2 and 50 1.8D, but it's nowhere near as bad as on my Sigma.
Keyboard shortcuts
- f
- Forum