p.9 #1 · DotTune: New AF tune technique, no photos required
I had thought you were getting both arrows at nearly the same time for a given AF tune value, which means something different. If you're getting just one arrow then that means AF tune values are not confirmed. Silly question but can you verify that AF fine tune is enabled? One gotcha on Nikon bodies is that it'll let you set/change individual lens values to your heart's content but if the AF fine-tune global setting is disabled then the values wont apply. Exposure doesn't matter provided you have a decent exposure for the AF system (ie, not too dark).
p.9 #2 · DotTune: New AF tune technique, no photos required
Yup my af fine tune option is set to ON. But yah I am continuously getting a flicker and never a solid dot for more than 5 seconds. I feel like it's close because the dot is solid for longer periods of time when I am at a certain range. Not sure if I am just doing something entirely wrong though.
p.9 #3 · DotTune: New AF tune technique, no photos required
You may be doing it entirely correctly - there are occasions where a body+lens combination wont tune anywhere within the +/- 20 range, although it doesn't happen too often. How many lenses have you tried?
p.9 #5 · DotTune: New AF tune technique, no photos required
Same thing with the 85. No solid dot achieved for 5 seconds. I just watched your video on youtube and saw the real live example you did at the end. I'm unable to get the solid dot like you did. What could that mean? Should I send it in for service?
Often times, I get both left and right flickers. How is that possible?
p.9 #6 · DotTune: New AF tune technique, no photos required
snapsy wrote:
Large ranges for certain lenses aren't abnormal. Btw since your range included +20 the range might be a little wider (past +20); to know for sure you can try the slightly-defocused workaround to find the full width of the range and then apply it to the midpoint calculation.
I'm trying the defocus workaround. But I keep going too far with the defocus (even though I'm trying hard to barely turn the lens focus ring) and it causes the results to go off the scale in the other direction. All it takes is just a hair of a turn and it goes off in the opposite direction.
p.9 #7 · DotTune: New AF tune technique, no photos required
MRomine wrote:
I'm trying the defocus workaround. But I keep going too far with the defocus (even though I'm trying hard to barely turn the lens focus ring) and it causes the results to go off the scale in the other direction. All it takes is just a hair of a turn and it goes off in the opposite direction.
It can be a bit tricky esp at certain focal lengths and the MF ring slop on a lot of modern AF-S lenses. I would try putting the AF target on a moveable platform like a cardbox box - that should give you more precision over slight defocusing vs defocusing with the lens.
p.9 #8 · DotTune: New AF tune technique, no photos required
snapsy wrote:
It can be a bit tricky esp at certain focal lengths and the MF ring slop on a lot of modern AF-S lenses. I would try putting the AF target on a moveable platform like a cardbox box - that should give you more precision over slight defocusing vs defocusing with the lens.
p.9 #9 · DotTune: New AF tune technique, no photos required
TightLines wrote:
Below are the calibration numbers for my three lenses:
105mm: -20
60mm: -17
16mm Fisheye: -20
Came to these by shooting a ruler in mm's at an angle and looking at 100% crop. Think this body is far enough out of whack that the dot tune method simply won't work.
Cope - thanks for the pricing details. $143 would be worth it on the body but I fear they would claim it's acceptable and not actually adjust anything.
Snapsy - quick question about the global adjustment values you linked me to in another thread. My take away was that each increment of the "default" value corresponded to roughly 2x of the adjustment for the individual lens. I spent quite a bit of time playing around with this on my camera and have come to the conclusion that changing the default settings has NO impact on my focus fine tuning. Really bizarre. Say I set the fine tune adjustment to -20 and my focus was ballpark ok, changing the global value on top of that definitely does not move my focus point (as confirmed in images). To further test things, I set my lens specific fine tune back to 0 and changed only the default to -20, again, no difference (left with a strong back focusing bias). ...Show more →
Snapsy - thanks for all the help throughout this.
I picked up a sturdier tripod and moved my tests to better lighting. Did a more manual process of comparing live view to actual tethered shots (from LR) and selected the fine tune adjustments that best mirrored the live view shots. Then confirmed focus tune using dot tune as you've described it. All is well and no lens is more than +/- 5. Bad technique had led me down the -20 road.
p.9 #10 · DotTune: New AF tune technique, no photos required
Snapsy- thanks for all of your hard work. I just want to verify things before I do my fine tuning. On zoom lenses, do I go per focal length or do you I go from wide to middle to tele? Also, do I have to move the tripod back if I go from wide to middle to tele? Thanks in advance. More power to you.
p.9 #11 · DotTune: New AF tune technique, no photos required
lottel wrote:
Snapsy- thanks for all of your hard work. I just want to verify things before I do my fine tuning. On zoom lenses, do I go per focal length or do you I go from wide to middle to tele? Also, do I have to move the tripod back if I go from wide to middle to tele? Thanks in advance. More power to you.
My pleasure, thanks. Since Nikon bodies don't support multiple AF tune values for zooms like Canon, you naturally must use only a single tune value. I suggest tuning twice on the lens, one at each extreme end of the focal range. The best case scenario is that both ends tune within 5 or fewer ticks of each other - in that case you can simply use the average as your final AF tune value. If however they tune much greater than 5 then you either have to still use the average (which may not give great results at either end) or use the tune value at whichever end of the focal range you plan to use most often with the lens.
p.9 #12 · DotTune: New AF tune technique, no photos required
Snapsy, thank You very much! Only now I found out your method, tried it with d810 and 200-400 with teleconverters: it works perfectly!! After so many headaches with different unsiccesful tries i saw the light... Thanks again and Best Regards from a very satisfied pixelpeeper.
p.9 #13 · DotTune: New AF tune technique, no photos required
I have D800E that never gets a solid dot for more than a second. Most of the time both arrows blink and the focus hunts. I have tried the Dot Tune Method and changed F-stops to get a wider field of view; nothing helps. Same blinking from -20 to +20. This is a longer lens with a teleconverter on it. Any thoughts?
p.9 #14 · DotTune: New AF tune technique, no photos required
Hello all,
Tried yesterday to fine tune with this method my D750 and the Nikkor 50mm f/1.8, as also the Nikkor 24-120 f/4.
The method for the 50mm, gave me a interval in focus of +2 till +16, giving me a +9 as middle point. After executing i took several test shots for +9, 0 and -9 just to check the results. The -9 was clearly out of focus. But the strange was that i was unable to check in all the test shots a difference between +9 and 0. Suggestions?
For the 24-120, i found that distance to target really impacts the results, as i first runned the exercise with all focal distances, but at the same place, and it gave me one results. Then i made it again in 24mm at 1,20mts, and in 50mm at 2,5mts and found different results than before. Also i realized that low battery seems to influence the results also ... but it was late in the night, and i need to do it again.
p.9 #15 · DotTune: New AF tune technique, no photos required
HugoSousa wrote:
Hello all,
Tried yesterday to fine tune with this method my D750 and the Nikkor 50mm f/1.8, as also the Nikkor 24-120 f/4.
The method for the 50mm, gave me a interval in focus of +2 till +16, giving me a +9 as middle point. After executing i took several test shots for +9, 0 and -9 just to check the results. The -9 was clearly out of focus. But the strange was that i was unable to check in all the test shots a difference between +9 and 0. Suggestions?
For the 24-120, i found that distance to target really impacts the results, as i first runned the exercise with all focal distances, but at the same place, and it gave me one results. Then i made it again in 24mm at 1,20mts, and in 50mm at 2,5mts and found different results than before. Also i realized that low battery seems to influence the results also ... but it was late in the night, and i need to do it again.
You're seeing first hand why AFFT is often useless, since the value only works for one single combination of focal length and subject distance unless you get really lucky and your lens is out by the exact same amount at every combination. On your 24-120, for example, there is probably 30+ combinations - finding one value that snaps everything into perfect focus is unlikely. If you make the lens better at one combination, chances are it's now worse at another. If I had any lens out +16 I'd be exchanging it personally. I wouldn't worry about small values like 2 or 3 as it's within the tolerances of most PDAF systems but much more than that and you're going to notice it regularly if you're picky.
p.9 #16 · DotTune: New AF tune technique, no photos required
@Mark, any idea if Nikon repairs zooms with that issue? Just curious.
I have a Sigma 18-35, and I had to fine tune both ends of the zoom lens on the lens dock, but now it's good across the range. It was something like -10 up close, and +8 for distance. No wonder I could not get dot tune to work for that lens.
p.9 #17 · DotTune: New AF tune technique, no photos required
HugoSousa wrote:
Hello all,
Tried yesterday to fine tune with this method my D750 and the Nikkor 50mm f/1.8, as also the Nikkor 24-120 f/4.
The method for the 50mm, gave me a interval in focus of +2 till +16, giving me a +9 as middle point. After executing i took several test shots for +9, 0 and -9 just to check the results. The -9 was clearly out of focus. But the strange was that i was unable to check in all the test shots a difference between +9 and 0. Suggestions?
For the 24-120, i found that distance to target really impacts the results, as i first runned the exercise with all focal distances, but at the same place, and it gave me one results. Then i made it again in 24mm at 1,20mts, and in 50mm at 2,5mts and found different results than before. Also i realized that low battery seems to influence the results also ... but it was late in the night, and i need to do it again.
Try tuning at infinity. I haven't updated the video but in my testing, lenses that exhibit distance-specific AFMA behavior will focus well at all distances when tuned at infinity but that the inverse is not true (ie, tuning at closer distances will yield poor results at infinity). I typically use a natural target like a mountain. You can use anything you find outside at a distance, like a roof line.
p.9 #18 · DotTune: New AF tune technique, no photos required
mco_970 wrote:
@Mark, any idea if Nikon repairs zooms with that issue? Just curious.
I have a Sigma 18-35, and I had to fine tune both ends of the zoom lens on the lens dock, but now it's good across the range. It was something like -10 up close, and +8 for distance. No wonder I could not get dot tune to work for that lens.
As far as I know if you send the body and lens to them, they will calibrate it. They won't touch your Sigma stuff though, that will have to go to Sigma (Sigma has excellent customer service). You can see why the Sigma Dock requires you to calibrate so many different combinations I'm not sure why Nikon or any other company for that matter can't build this into their bodies, it must be more difficult than it sounds to us.
p.9 #19 · DotTune: New AF tune technique, no photos required
snapsy wrote:
Try tuning at infinity. I haven't updated the video but in my testing, lenses that exhibit distance-specific AFMA behavior will focus well at all distances when tuned at infinity but that the inverse is not true (ie, tuning at closer distances will yield poor results at infinity). I typically use a natural target like a mountain. You can use anything you find outside at a distance, like a roof line.
Thanks for the suggestion, I will do an extra round with my lens with better light conditions, to make sure i continue to have the same results. If they are still present i will try your suggestion and came back to report.
p.9 #20 · DotTune: New AF tune technique, no photos required
I am really interested to see what they did on the D500 to automate the process. The Sigma dock was time consuming, but the results look good. I would be happy not to ever tune another zoom, though.
CanadaMark wrote:
As far as I know if you send the body and lens to them, they will calibrate it. They won't touch your Sigma stuff though, that will have to go to Sigma (Sigma has excellent customer service). You can see why the Sigma Dock requires you to calibrate so many different combinations I'm not sure why Nikon or any other company for that matter can't build this into their bodies, it must be more difficult than it sounds to us.
