Luminous Landscape Forum
Equipment & Techniques => Medium Format / Film / Digital Backs – and Large Sensor Photography => Topic started by: spassig on June 09, 2019, 05:27:42 am
-
Hey
I shoot lot of pictures in Tuscany some weeks ago.
I use PO XF with IQ3.
I learn using the system.
A very small group of pictures don't have sharpness.
Example in screenshot.
I use a tripoid.
I use delay max 4 sec after shutter release.
I dont know the autofocus distance. The value isn't in EXIF.
Normally I use autofocus and when the autofocus is OK (house) I switch lens to manually and move to the crop.
Can I analyse the RAW to find out the cause of the RAW blur?
I want to improve my work to minimize the committee. :)
Jochen
-
Can I analyse the RAW to find out the cause of the RAW blur?
Hi Jochen,
Only if the blur has a very distinct character, it might become clear what caused it.
In this case, it could be a combination of factors.
When I try to improve the results in postprocessing (using Topaz Sharpen AI), I can achieve the most recovery with a combination of Stabilization and defocus Blur correction. So it looks like some (mostly horizontal) vibration was involved, and some defocus. In addition, the narrowish aperture added some diffraction blur. The Diffraction Blur you can correct with the built-in Capture One Diffraction correction.
I don't know if any wind was involved during the exposure time, to cause some vibration in the tripod setup.
Cheers,
Bart
-
Some years ago, Popular Photography magazine ran a test of shutter speeds and vibration. Turns out, your chosen shutter speed falls into the range that is most susceptible to vibrations, even when on tripod.
-
Some years ago, Popular Photography magazine ran a test of shutter speeds and vibration. Turns out, your chosen shutter speed falls into the range that is most susceptible to vibrations, even when on tripod.
Thanks.
If so than I will carry out some own tests.
Did you know which Magazin?
Jochen
-
... Did you know which Magazin?
No, sorry, it was years ago.
Their conclusion was, and I am speaking from memory, that the range between 1s and 1/15s is the most susceptible to vibrations from mirror slap, shutter bounce, or outside causes. Faster that that would freeze the movement, and slower than that would allow enough time for camera to stabilize.
But, as you said, it is best to conduct your own tests, with your specific camera.
-
No, sorry, it was years ago.
Their conclusion was, and I am speaking from memory, that the range between 1s and 1/15s is the most susceptible to vibrations from mirror slap, shutter bounce, or outside causes. Faster that that would freeze the movement, and slower than that would allow enough time for camera to stabilize.
But, as you said, it is best to conduct your own tests, with your specific camera.
I seem to remember that the original Sony A7 had this problem at just these shutter speeds.
-
Shake means lack of stability. I use the same system as you, and outside a studio every step of the system is fraught with shake! Are you locking up your mirror as part of the 4s delay? The tripod/head system is crucial - it should have more inertia than your camera, don't succumb to convenience because you'll pay for it with lost shots; likewise the head too needs to provide real down-force not just be a symbol of notional stability. If windy, try and stand where you can block it. You'll fix it - it all comes down to physics.
-
Why not use the electronic shutter function with the mirror up if the subject doesn't more too much?
Regardless, all images should be captured with the mirror permanently up, using live view. With leaf shutter lenses this should cause zero issues.
At least I am not facing any sharpness problems in the field with y H6D-100c with the aforementioned technique, a good series 2 RRS tripod and lenses up to 210mm f4.
Cheers,
Bernard
-
It looks for me as the house is primarily not in focus.
Did you use the focus trim function to adjust every lens in your XF?
(https://captureintegration.com/xf-focus-trim/)
Against vibration I use mirror up and the "Vibration Delay"
with the seismograph in my XF and a heavy tripod.
Regards,
Ben
-
@all
Thanks for feedback with recommendations and hints.
I will take into consider in future.
Jochen
-
Looks like defocus (misfocus) rather than shake. Shake generally has more effect on one axis, such that tall-thin detail is sharp (because there is very little left-right vibration) and wide-short detail is soft (because of the vertical angle change from vibration).
The advice you got about focus trimming is good advice, but in this case the part of your post where you said "A very small group of pictures don't have sharpness" leads me to believe the issue was not trimming.
I would suggest using the XF's Hyperfocal Focus Tool (tuned to your own standards) for shots like this. It will eliminate the need to focus at all, and decrease the likelihood of mis-focus to near zero.
The advice you're getting for camera shake is good general-purpose photographic advice. However, since you're sporting an XF with an IQ3 100mp it's not terribly relevant. Simply turn on XF's vibration mode (setting the max delay to infinity) and switch to using the IQ3's sensor-based electronic shutter. This will make it nearly impossible to get any vibration in the image. It's the absolute best anti-vibration workflow I've ever seen.
The Phase One dealer from whom you purchased can walk you through the Hyperforcal Focus Tool, the Vibration Delay mode, the Sensor-Based ES (https://www.dtcommercialphoto.com/phase-ones-electronic-shutter-in-detail/) as well as various in-camera focus checking. It's literally their job to make sure you're getting the most out of this hot rod kit! :)
-
OP, could you clarify what you mean when you say you switch to manual and move to the crop? Are you using a zoom, and then zooming in after you autofocus?
-
OP, could you clarify what you mean when you say you switch to manual and move to the crop? Are you using a zoom, and then zooming in after you autofocus?
I don‘t use a zoom.
The AF focus field is in the middle of the „sensor“.
I use the AF to the house.
Then I switch on the 150mm lens from autofocus to manualfocus.
Then I move the camera slightly right and slightly up.
I fix all connections (I hope so).
I press the trigger.
Jochen
-
...I use the AF to the house.
Then I switch on the 150mm lens from autofocus to manualfocus.
Then I move the camera slightly right and slightly up. ...
If I understood correctly, you are using autofocus and recomposing. You are not doing any additional manual-focus corrections, right?
If so, that might explain the results that "a very small group of pictures don't have sharpness. Autofocus is not absolutely precise, but has a range of acceptable sharpness instead.
I do not know your camera, but most modern cameras with live view have a manual-focus magnifying feature. That is the best way to focus precisely.
-
It really does not look like anything is sharp, so it may not be just focus.
Big cameras need solid tripods as mentioned.
Also mirror up and cable or timed release also as mentioned.
> "Normally I use autofocus and when the autofocus is OK (house) I switch lens to manually and move to the crop."
Explain how you focussed again.
Autofocus is sometimes guess focus.
Focus and then recompose does not work. It is mathematically impossible. Read up on "True Focus" as used by Hasselblad.
Focus and then zoom does not work on most zooms.
-
Focus and then recompose does not work. It is mathematically impossible.
The distance between house and camera is approximativ 800 m.
See screenshot 1.
The calculation depth of field say from very near until infinite.
See screenshot 2.
Why can I not change recompose after autofocus by fix lens distance?
Jochen
-
Focus and then recompose does not work. It is mathematically impossible. Read up on "True Focus" as used by Hasselblad.
Focus and then zoom does not work on most zooms.
Bob, the XF also has a dedicated mode for this called AFr (same idea as True Focus).
But it's 100% irrelevant to the OP's question. He's using a wide angle on a far away subject. The geometry difference (that AFr or True Focus address) is completely negligible here.
If I had to guess, the OP was following [AF mode >>> focus on subject >> switch to MF >> recompose >> capture] throughout the shoot without issues, but on this specific sequence of images he forgot the [switch to MF] step and so the XF tried (and failed) to refocus on the sky, resulting in several out of focus images.
If this theory is correct, it highlights that the default settings are not great for the workflow he is using. I would suggest changing the workflow (as explained in my previous post, for example, to use hyperfocus mode), but if this workflow is what he prefers I'd suggest a minor tweak: switch the XF UI to remove autofocus from the shutter release and add it to the rear button. This "AF on demand" tweak would remove the need to switch between AF and MF mode on the lens, which would also eliminate the chance that lens focus changes during that switch.
Again, I'd direct you to your P1 dealer. I would rather have expected they would have walked you through these powerful and practical features of the XF (and the many many many others) during your pre-purchase evaluations or your post-purchase training session, but if not, they can do so now.
-
Jochen
I think AF is not a good tool to make landscape photographs with a tripod. Why don't you use live view to nail manually focus?
It's a fast workflow with your camera.
With the vibration delay of the XF you will be almost 100% sure to make the picture you want.
I use Af only for handhold work.
Regards,
Ben
-
He's using a wide angle on a far away subject. The geometry difference (that AFr or True Focus address) is completely negligible here.
I did not actually realise until he shared it that he is actually 800m from the house. A wide angle lens at 800m? Why? With that magnification (and possibly heat haze) it is probably as good as it gets. The slightest vibration would kill it.
And raw is raw. Normally processing should improve it.
Is the Afr automatically applied? On the Hassy you actually have to enable it and know how to use it.
-
... A wide angle lens at 800m? Why?
It was a 150mm lens, a medium telephoto (93mm equivalent).
-
It was a 150mm lens, a medium telephoto (93mm equivalent).
Indeed. Must have misread the screen grab in my sleep deprived state.
Still, recomposition is not a meaningful factor at the distance with this amount of recomposition.
-
Nobody has mentioned it, so here is another possible cause of the softness: air/heat shimmer.
(Edit: apologies, just noticed that BobShaw mentioned it a few posts back)
Shoot faster to freeze it. Or, if the light isn't changing too quickly, take a few shots and pick the one with the least shimmer
-
@all
Thanks for feedback.
Today I have send the RAW to PO and discuss the picture.
PO is sure it's not a problem with focus.
It's vibration.
We also discuss some recommendations from this thread.
I will consider them in the future.
I hope make better technical pictures in Danmark were I go tommorow.
Small picture without vibration ;)
Jochen
-
Not sure what to make of the new image - its soft - but is that because of the reduced size to post it? Is it possible to post some full size crops?
-
@all
Thanks for feedback.
Today I have send the RAW to PO and discuss the picture.
PO is sure it's not a problem with focus.
It's vibration.
Yes, that's what I concluded as well. But the strange thing is that it appeared to be more horizontal than vertical. That's strange for camera/shutter-induced motion. Hence my suspicion that wind or some other external force had something to do with it.
Have fun in Denmark.
Cheers,
Bart
-
Yes, that's what I concluded as well. But the strange thing is that it appeared to be more horizontal than vertical. That's strange for camera/shutter-induced motion. Hence my suspicion that wind or some other external force had something to do with it.
Have fun in Denmark.
Cheers,
Bart
You have eyes like a eagle :)
I don’t find the different between horizontal and vertical.
I don‘t remember of windy morning.
Maybe I don’t have fix all connections after first AF shoot.
Jochen
-
Not sure what to make of the new image - its soft - but is that because of the reduced size to post it? Is it possible to post some full size crops?
Thanks.
I show it that I can shoot sharper pictures, funny :)
Jochen
-
Perhaps the process of switching from AF to MF caused the focus to shift slightly? Some lenses use a push-pull mechanism of the focus ring to achieve that. I would suggest using the back button for AF, the. You can focus and recompose much more easily.
-
Hi,
The only way to achieve correct focus is using live view at actual pixels at shooting aperture.
Anything else is a proxy. May be a decent proxy or not.
Best regards
Erik
-
The only way to achieve correct focus is using live view at actual pixels at shooting aperture.
Anything else is a proxy. May be a decent proxy or not.
What a nonsense absolutest statement from a normally logical and non sensational guy.
There are many workflows that can consistently result in correct focus.
For example the XF Hyperfocal Tool will consistently be accurate to a single motor unit which is far finer than your hand during manual focusing in live view. It also has the benefit of being faster and easier to use and far more resistant to human error.
-
Hi Doug,
I would maintain that an AF-sensor in a different optical path is a proxy of the real sensor. For definition of proxy, check wikipedia: https://en.wikipedia.org/wiki/Proxy
Also with live view, you can choose any point to focus on. With AF on medium format you are pretty limited to a central part of the sensor. The focus/recompose method induces some geometric error. You can correct that with calculations in the camera using with gyros to detect rotation.
Most lenses also have field curvature, that varies with focus distance. Focusing using live view uses the actual image and you can focus on the exact point that you want to focus on. So, it is as good as it can get. But most lenses also have focus shift. Therefore optimum focus needs to be made at shooting aperture.
But:
- It is quite possible that Phase One has engineered around all those issues.
- The point you make on accuracy on focusing ring vs. AF actuator is an interesting one. It is quite true that many AF lenses are difficult to focus manually.
Best regards
Erik
What a nonsense absolutest statement from a normally logical and non sensational guy.
There are many workflows that can consistently result in correct focus.
For example the XF Hyperfocal Tool will consistently be accurate to a single motor unit which is far finer than your hand during manual focusing in live view. It also has the benefit of being faster and easier to use and far more resistant to human error.
-
Thanks, I know the definition of proxy.
I was disagreeing with your use of the word “only” in “live view is the only way to achieve correct focus”
It’s just flat out wrong to say that. Live view is very often a great way to focus. But there are many ways to achieve correct focus. In many cases live view isn’t even the fastest, easiest or most consistent.
-
Hi Doug,
Thanks for your comments.
That said, it seems that the OP missed something, which may be diagnosed as bad focus. Looking at the real image coming of the sensor at actual pixels, I don't think it can be bad advice.
I don't know how exactly the XF can focus, and I have not seen any test of focus accuracy on the XF.
With say the Fuji GFX we are in better shape, as Jim Kasson has made a lot of excellent testing on that camera.
Just as an example: https://blog.kasson.com/gfx-50s/fuji-63-2-8-focus-shift-and-autofocus-accuracy/
A system like Nikon D850 shows another picture:
(https://blog.kasson.com/wp-content/uploads/2018/01/d850-58-mf-coc-fixed.png)
From the same test. Sony A7rII has less issues:
(https://blog.kasson.com/wp-content/uploads/2018/01/a7riii-55-mf-coc-fixed.png)
I have never seen such tests with Phase One.
In the figures above, you wouldn't like to have the CoC exceed say 1.5 times the pixel pitch.
As I said, I have not seen any tests like that on Phase One. Jim only tests things he plans to buy.
Best regards
Erik
Thanks, I know the definition of proxy.
I was disagreeing with your use of the word “only” in “live view is the only way to achieve correct focus”
It’s just flat out wrong to say that. Live view is very often a great way to focus. But there are many ways to achieve correct focus. In many cases live view isn’t even the fastest, easiest or most consistent.