p.28 #2 · Dot-Tune: Autofocus Fine Tuning in under 5 minutes
Anyone care to speculate why Canon has not released their own Firmware automation equivalent for their cameras that provide MFA? To me, the algorithm is so basic, the firmware has all the hooks already to the camera operation commands, I can't think of why they haven't done it. Especially when they could directly reduce "bad AF / bad lens" claims and warranty service burdens.
p.28 #3 · Dot-Tune: Autofocus Fine Tuning in under 5 minutes
I'm not sure when you updated the tutorial video but it's different from the first one I watched a few years back. The new chart is great too.
I wanted to fine-tune a couple of my lenses with Focal but the 1D C body is not supported. I tried Dot-tune last night and it gave me very accurate results. It really only takes about 3 minutes to fine tune each lens.
This is a great free tool and I highly recommend it.
Fred
p.28 #4 · Dot-Tune: Autofocus Fine Tuning in under 5 minutes
gschlact wrote:
Anyone care to speculate why Canon has not released their own Firmware automation equivalent for their cameras that provide MFA? To me, the algorithm is so basic, the firmware has all the hooks already to the camera operation commands, I can't think of why they haven't done it. Especially when they could directly reduce "bad AF / bad lens" claims and warranty service burdens.
This will be the main selling point of the 5D mk 5 and the 1DZ.
p.28 #5 · Dot-Tune: Autofocus Fine Tuning in under 5 minutes
Fred Miranda wrote:
I'm not sure when you updated the tutorial video but it's different from the first one I watched a few years back.
I wanted to fine-tune a couple of my lenses with Focal but the 1D C body is not supported. I tried Dot-tune last night and it gave me very accurate results. It really only takes about 3 minutes to fine tune each lens.
This is a great free tool and I highly recommend it.
Fred
Glad it was helpful Fred. I've only done the one video - the only revisions I've made are to the comment part of the info section (underneath the video) and some on-screen comment boxes that apply to Nikon owners.
p.28 #6 · Dot-Tune: Autofocus Fine Tuning in under 5 minutes
Fred Miranda wrote:
Honestly I always used Focal but yesterday, out of necessity, I took Dot-tune seriously.
I'm glad I did and will be using it with all my bodies now. The best part is that you don't need to be tethered.
Any tips for fine-tuning the 100-400L ? Always use the 50x rule?
Here is what Focal recommends:
I recommended 50x in the video because that was the prevailing convention but in my testing since then I've found that tuning @ infinity produces the best results for most lenses, particularly those which exhibit distance-specific tuning variability. For such lenses I found that the AF tune value at MFD-to-50x doesn't work well @ intermediate and infinity distances but the tune value infinity works well at all distances. For infinity I use a natural target outside, like a roof or a tree.
p.28 #7 · Dot-Tune: Autofocus Fine Tuning in under 5 minutes
snapsy wrote:
I recommended 50x in the video because that was the prevailing convention but in my testing since then I've found that tuning @ infinity produces the best results for most lenses, particularly those which exhibit distance-specific tuning variability. For such lenses I found that the AF tune value at MFD-to-50x doesn't work well @ intermediate and infinity distances but the tune value infinity works well at all distances. For infinity I use a natural target outside, like a roof or a tree.
Great tip! Especially for lenses like the Sigma 35 f/1.4 art which fall in the category of 'distance-specific' tuning.
p.28 #8 · Dot-Tune: Autofocus Fine Tuning in under 5 minutes
snapsy wrote:
I recommended 50x in the video because that was the prevailing convention but in my testing since then I've found that tuning @ infinity produces the best results for most lenses, particularly those which exhibit distance-specific tuning variability. For such lenses I found that the AF tune value at MFD-to-50x doesn't work well @ intermediate and infinity distances but the tune value infinity works well at all distances. For infinity I use a natural target outside, like a roof or a tree.
This is interesting. I went through the dot tune process with all of my lenses and TC combinations a year or so ago, at the recommended (i.e. 50x) distances, and found the results unimpressive - most lenses ended up better without MA than with the numbers that I got using dot tune. I may try again at infinity or threreabouts.
p.28 #11 · Dot-Tune: Autofocus Fine Tuning in under 5 minutes
I've had good results with MFA and dot-tune just inside of infinity focus. I'm curious now, and will re-check a few lenses. Is it possible to get examples of lenses that fall into the 'distance-specific' category? Since all focusing inside of infinity is 'distance-specific' I'm a bit confused.
p.28 #12 · Dot-Tune: Autofocus Fine Tuning in under 5 minutes
jcunwired wrote:
I've had good results with MFA and dot-tune just inside of infinity focus. I'm curious now, and will re-check a few lenses. Is it possible to get examples of lenses that fall into the 'distance-specific' category? Since all focusing inside of infinity is 'distance-specific' I'm a bit confused.
Thanks.
Distance-specific means that the optimal AF tune value changes with focus distance for a given lens.
p.28 #14 · Dot-Tune: Autofocus Fine Tuning in under 5 minutes
Fred Miranda wrote:
Honestly I always used Focal but yesterday, out of necessity, I took Dot-tune seriously.
I'm glad I did and will be using it with all my bodies now. The best part is that you don't need to be tethered.
Any tips for fine-tuning the 100-400L ? Always use the 50x rule?
Here is what Focal recommends:
Are those distances from the front of the lens or the actual sensor? I've been doing it from the sensor.
Last week I dot tuned my sigma 35/1.4 and the before / after results are impressive. The sigma Dock is an impressive tool.
p.28 #15 · Dot-Tune: Autofocus Fine Tuning in under 5 minutes
graham_martin wrote:
I'm going to try this on an older Sigma 120-300. Should I set the lens at its mid range focal length of about 200mm?
I would do 120, 200, and 300 to get an idea of how consistent the focus is across the focal range. Hopefully all the results will be pretty close.
Then pick the average (unless you have a camera that lets you enter two values), or if there is a wide range, you will likely have to go with the adjustment that corresponds to the focal lengths you use the most.
p.28 #16 · Dot-Tune: Autofocus Fine Tuning in under 5 minutes
snapsy wrote:
Most cameras suffer from this, including the D800/E. AWB sensors probably aren't reliable enough to rely upon for integration into the PDAF logic.
I agree, the 5DM3/1DX appear to be unique in their consistent ability to operate at a zero MFA. My theory is that they've tightened the manufacturing tolerances of these bodies.
For bodies that need MFA I wouldn't suggest avoiding tuning just because PDAF operates differently in different lighting temperatures. I tune to my most used lighting temperature for a given lens and if I find myself in a lighting situation that differs significantly from what I tuned then I just tune again on location, which is fast and easy with DotTune but not so with other methods. I also have tungsten -> daylight deltas written down for my most used lenses (along with focus shift deltas) so mostly I just adjust the tune value by memory.
Snapsy, I've been reading through this entire thread and got up to page 10 (2 years ago)where this guy stated he was seeing a significant variance in his Dot Tune results based on the type (tempeture) of lighting used during the tuning process. Then I see where you replied to his post (above) and it looks like you are saying "Yes" the Dot Tune can be thrown off depending on the type of lighting used.
Is this correct? I know I'm digging back 2 years but has anything changed regarding this issue? Have you learned anything new since then about this phase detection focus shift due to light tempeture?
The reason I ask is that I experienced something strange after I did the Dot Tune AFMA to my EF-S 17-55. My wide end got worse but the long end improved. -3 and +3 respectively per my results of the Dot Tune. I had used a very bright tungsten shop light bouncing off the ceiling during the tuning process and am now wondering if this is what caused my issue?
p.28 #17 · Dot-Tune: Autofocus Fine Tuning in under 5 minutes
lighthound wrote:
Snapsy, I've been reading through this entire thread and got up to page 10 (2 years ago)where this guy stated he was seeing a significant variance in his Dot Tune results based on the type (tempeture) of lighting used during the tuning process. Then I see where you replied to his post (above) and it looks like you are saying "Yes" the Dot Tune can be thrown off depending on the type of lighting used.
Is this correct? I know I'm digging back 2 years but has anything changed regarding this issue? Have you learned anything new since then about this phase detection focus shift due to light tempeture?
The reason I ask is that I experienced something strange after I did the Dot Tune AFMA to my EF-S 17-55. My wide end got worse but the long end improved. -3 and +3 respectively per my results of the Dot Tune. I had used a very bright tungsten shop light bouncing off the ceiling during the tuning process and am now wondering if this is what caused my issue?
Sensitivity to light temperature is an inherent PDAF attribute and so naturally will manifest independent of the tuning method used. I generally don't find it to be a major issue in terms of AF variance between the bodies I use but others who shoot in a more varied set of conditions (or have different bodies) may struggle more with it.
Regarding your 17-55, was the bright lighting you used diffuse or high-contrast on your target? I've found high-contrast lighting can have a big effect on tuning and produce inaccurate results, esp for large-aperture lenses where fringing would affect what the AF sensors see. I generally do most of my tuning now outside during daytime and use natural targets with only moderate reflectivity.
p.28 #18 · Dot-Tune: Autofocus Fine Tuning in under 5 minutes
I've noticed that depending on how well the target is lit, AF will struggle more or less. However from my limited indoor tests, that was not an issue because if it struggles on one extreme it will struggle the same way on the other and the average will be the same. I will try dot-tuning outdoors as suggested and see if there is any dramatic difference in the numbers. With fast aperture lenses, a ND filter may be needed.
p.28 #19 · Dot-Tune: Autofocus Fine Tuning in under 5 minutes
Fred Miranda wrote:
With fast aperture lenses, a ND filter may be needed.
I dot-tuned the 85L II indoors. For fast aperture lenses it is best to first quickly determine whether it is front/back focussed (I use the Lensalign incline target for this) and then dot-tune it.
p.28 #20 · Dot-Tune: Autofocus Fine Tuning in under 5 minutes
I decided to have another go at dot tune, but I'm still getting inconsistent results.
I decided to adjust my 24-70 II. Tele end was no trouble - I tested at about 3-4m, and an adjustment of -1 was indicated. This adjustment gives me good results at various focal distances. Not that the pre-adjustment results were bad at 70mm, of course, given the small adjustment. It's the consistency that I'm interested in here.
A different story at 24mm. First, I did the dot tune test at about 1.2 - 1.5m (i.e. 50x focal length) - indicated a required adjustment of +10. I took some general shots of my son at a longer focus distance (i.e. normal environmental portrait distance of 2-4m), which were clearly OOF.
I then redid the dot tune test at about 4m - the same distance as I did for the long end. This gave an indicated adjustment of -4. Better results in the 2-4m general shooting test. But some testing at closer focus distances clearly showed front focus at an adjustment of -4.
My conclusion? At the wide end, my 24-70 needs different adjustment at different focussing distances. Is this normal? What do I do, just put the adjustment somewhere in between? This is what I have done, setting it back to zero.
That sigma hub allowing adjustment at different focussing distances sounds good . . .
