Nikkor UW on Sony Mirrorless the write up

1 hour ago, Alex_Mustard said:

We could do with merging the two Nikonos 15mm threads. If a moderator could help with that? I used to know how to do it, but can’t find any controls to do it now.

Age kicking in, Alex?

Done......

17 hours ago, TimG said:

Age kicking in, Alex?

Done......

Something is. I still don’t see any moderation controls. It is possible that I don’t actually have any practical moderation permissions on my log in. 

19 hours ago, Isaac Szabo said:

 Focus peaking with my setup was not precise enough for my standards. I had to use magnified manual focus.

Since both you and @Interceptor121 comment on peaking being hard to use. I wonder if this is because the lens is stopped down already - making peaking less effective. It might be worth setting it to f/2.8 to check if peaking works better. If it does, you could then dial in the focus at the start of the dive, then shut the lens down and trust it is in focus. 

13 minutes ago, Alex_Mustard said:

Something is. I still don’t see any moderation controls. It is possible that I don’t actually have any practical moderation permissions on my log in. 

Since both you and @Interceptor121 comment on peaking being hard to use. I wonder if this is because the lens is stopped down already - making peaking less effective. It might be worth setting it to f/2.8 to check if peaking works better. If it does, you could then dial in the focus at the start of the dive, then shut the lens down and trust it is in focus. 

I took shots between f/2.8 and f/11 maybe one at f/16 broadly I was at f/5.6

Peaking is problematic on the Sony because in order to work properly it needs the live view effect to be set to on

However if you set it to on the entire exposure reflects the settings dialed in the camera which means you may not be able to see a thing

With a native lens if you set live view on and close down the aperture peaking reflects the additional focus range however the nikkor is manual so the lens is physically stopped down to whatever you set and therefore peaking should reflect the aperture and work better when you are closed down so I am not sure what was going on there

It could simply be that when the ambient light is really low as it is the case of the pool I use for training contrast detection becomes less effective and peaking suffers

I did get the yellow coloring but even when it was like that the shots were not in perfect focus

Now that I know how to use the focus magnifier with the manual lens I can give it another go however I have other 5 use cases before I return to the Nikkor so it may take a while

5 hours ago, Alex_Mustard said:

Since both you and @Interceptor121 comment on peaking being hard to use. I wonder if this is because the lens is stopped down already - making peaking less effective. It might be worth setting it to f/2.8 to check if peaking works better. If it does, you could then dial in the focus at the start of the dive, then shut the lens down and trust it is in focus. 

With my A7R II peaking works a little better with the aperture wide open, but it is still very imprecise/unreliable. The problem (at least with the A7R II) is that the camera is basing the peaking on the downscaled image being displayed on the LCD/EVF instead of the full resolution image from the sensor. And when an image is downscaled much more appears to be in focus than is actually in focus when the image is viewed at 100% at full resolution. So the peaking implementation has a fundamental flaw that makes it very imprecise. This is especially problematic with wide angle lenses since out of focus areas only blur slightly and are still lit up by the peaking on the downscaled image. It’s not quite as bad with macro lenses since out of focus areas blur much more dramatically. Obviously the right way to implement peaking would be to base it on the full resolution image from the sensor, but I’m not sure whether or not there are other camera models that do it that way.

5 hours ago, Interceptor121 said:

Peaking is problematic on the Sony because in order to work properly it needs the live view effect to be set to on

That is not the case on my Sony.

That is not the case on my Sony.

Try setting live view on and off

You will see that with it on the depth of field reflects in the frame

With it off it doesn’t

Focus needs to be on standard or the lens stays open

The a1 has 3 megapixel evf it would be adequate to nail focus however it wasn’t

I need to try with live view on but then i need light…

Peaking is implemented through the video pipeline it has nothing to do with the sensor read but the display

So evf peaking is better than lcd

For wide angle even low resolution is fine the camera doesn’t focus at higher resolution anyway when using autofocus

It will only readout at full resolution when exposing the frame so that you t can up the frame rate to absurd values to make AF fast

The A1 may he an exception to this rule as it reads 200 fps

Sent from my iPhone using Tapatalk

We are talking about fully manual lenses here. The aperture will stay at whatever it is set on the lens. Camera settings have no influence on it. So yes, with lenses like the Nikonos 15mm focus peaking will show the same DOF regardless of live view display setting, focus settings, etc. 

I've already tried every relevant setting in regards to focus peaking on my camera. What I said above comes from direct experience. Unlike you I'm not guessing. Why don't you actually try your ideas before telling other people they're wrong?

Again, what I said above about the inadequate resolution of focus peaking applies to the A7R II. Other cameras like your A1 could behave differently. Then again, you've already said it's inaccurate and can't be relied on.

Why don't you just test it at home? It's very easy and only takes a few minutes. Put on a wide angle lens and make a mental note of some of the things focus peaking is saying are in focus. Then zoom in to 100% and check if all of those things are actually in focus. With my A7R II it's immediately apparent that focus peaking can't be relied upon for critical focus (by my standards anyway).

7 hours ago, Isaac Szabo said:

We are talking about fully manual lenses here. The aperture will stay at whatever it is set on the lens. Camera settings have no influence on it. So yes, with lenses like the Nikonos 15mm focus peaking will show the same DOF regardless of live view display setting, focus settings, etc. 
I've already tried every relevant setting in regards to focus peaking on my camera. What I said above comes from direct experience. Unlike you I'm not guessing. Why don't you actually try your ideas before telling other people they're wrong?
Again, what I said above about the inadequate resolution of focus peaking applies to the A7R II. Other cameras like your A1 could behave differently. Then again, you've already said it's inaccurate and can't be relied on.
Why don't you just test it at home? It's very easy and only takes a few minutes. Put on a wide angle lens and make a mental note of some of the things focus peaking is saying are in focus. Then zoom in to 100% and check if all of those things are actually in focus. With my A7R II it's immediately apparent that focus peaking can't be relied upon for critical focus (by my standards anyway).

You seriously need to drink some camomile

I explained to Alex why the issue was not the aperture I had the lens open then made a comment on peaking for photography
You replied to my comment peaking is problematic with for me it isn’t
What followed is a general explanation of why peaking doesn’t work correctly unless you set live view effect on and how peaking works which has to do with the displayed view
Even on a manual lens the display brightness compensation with live view off will change the contrast the camera will detect regardless of the aperture being or not fixed because it works on the display not the actual exposure
But my point remains that in general terms peaking for photography is problematic unless you shoot ambient light scenes because you can’t work with live view on
For video where you shoot how you see it instead it works well, this is not unique to sony it is the same for pretty much all camera system. I have planned to get this Nikkor lens for video and we will see how it does there.
I think the fact that the camera was compensating the display together with the dark conditions of the pool made peaking work less well than it should and I could not work with live view off as it was pitch black.
The rest of what you wrote about peaking not using the full sensor resolution is not the issue as you can nail focus with reduced resolution in almost all situations especially wide angle

Now it would be good to go back on theme
Peaking in dark conditions doesn’t seem to work 
Is focus magnifier better net of the distracting Effect?


Sent from my iPhone using Tapatalk

You guys really need to stop trying so hard to be right on the internet. 

Guys can we attack the proposition rather than the person, it helps to keep things civil here.

Guys can we attack the proposition rather than the person, it helps to keep things civil here.
 

1 minute ago, Interceptor121 said:

It is very easy to see at which point in this discussion that happened
When someone starts talking directly to someone else instead of the topic at hand
It is always the same two people in my case that get excited and do it regularly


Sent from my iPhone using Tapatalk

The request is in general - to all parties to the topic.  If someone attacks please just ask them to stick to the topic at hand and respond to their proposition.

22 minutes ago, ChrisRoss said:

The request is in general - to all parties to the topic.  If someone attacks please just ask them to stick to the topic at hand and respond to their proposition.

My objective is to try to have a point of view or state facts on how things work I have no interest in battles

There were a number of imprecise statements that need correcting

With my A7R II peaking works a little better with the aperture wide open, but it is still very imprecise/unreliable. The problem (at least with the A7R II) is that the camera is basing the peaking on the downscaled image being displayed on the LCD/EVF instead of the full resolution image from the sensor. 

Adobe has a good explanation of focus peaking

https://www.adobe.com/creativecloud/photography/discover/focus-peaking.html

Focus peaking starts in videography normally on a field monitor with manual focus you use peaking to check focus or magnifier

In a video situation the exposure on the monitor is what you will record

The challenge with photography is that many brands Sony included can have a setting to turn live view off.

This is useful for dark scenes or high contrast scenes because the display does not reflect the exposure settings anymore so you can clearly see and frame.

This has a side effect as many tools that need live view to reflect exposure start not working properly, this includes 

1. Focus peaking
2. Zebra
3. Waveforms
4. Other exposure or focus tools

Focus peaking works on the display not on the sensor and is unrelated to the camera autofocus. Generally unless you are focussing on a dot even 1280x720 has been sufficient for video to nail focus

The lower resolution of the display is not the issue here 

When the live view effect is off even if the aperture is fixed the camera pumps up the brightness of the display increasing gain and this changes the contrast and fools all the tools above. In some camera when you half press you get an updated view of the tools above that differs from what you were seeing

I have another issue with the previous statement as it assumes that autofocus works with full sensor resolution which is also generally not the case on phase detect that uses sensor display output only for subject detection

For focus it uses phase difference pixels that do not even form an image so the resolution in the EVF matters nothing to AF

It is important to avoid making confusion on how things work. if someone gets it wrong you can try and correct without getting personal it generally helps but at the end you can abandon the scene as you see fit

11 hours ago, Interceptor121 said:

You seriously need to drink some camomile

 

Gentlemen! Please, let us revert to the usual polite exchanges between us. WP has been a haven for moderate and calm discussion.

Let's keep it that way.

Gentlemen! Please, let us revert to the usual polite exchanges between us. WP has been a haven for moderate and calm discussion.
Let's keep it that way.

I'm talking to both of you...... 

As Chris has said, let's stick with the subject and not start to have a go at each other. It's not necessary.

My Nikonos 3 does not have focus peaking. Practice your underwater distance judging, and shoot within the dog plane.

For those who are actually interested in this, here is a quick/easy focus peaking accuracy test with the A7R II and a wide angle lens. Focus peaking was set to low. I did it with live view setting effect both on and off, but the results were identical, so there’s no point in posting more than one set of images. The results were also the same using the EVF instead of the LCD.

This is the full scene, and focus peaking says the test chart is in focus:

This is a 100% view of the test chart, which is clearly out of focus:

This is a 100% view of the test chart when properly focused:

Clearly not everything that focus peaking lights up as in focus is actually in focus, so for me it is not accurate enough to be relied upon.

11 hours ago, Isaac Szabo said:

For those who are actually interested in this, here is a quick/easy focus peaking accuracy test with the A7R II and a wide angle lens. Focus peaking was set to low. I did it with live view setting effect both on and off, but the results were identical, so there’s no point in posting more than one set of images. The results were also the same using the EVF instead of the LCD.
This is the full scene, and focus peaking says the test chart is in focus:

This is a 100% view of the test chart, which is clearly out of focus:

This is a 100% view of the test chart when properly focused:

Clearly not everything that focus peaking lights up as in focus is actually in focus, so for me it is not accurate enough to be relied upon.

This example is not really a good demonstration of how focus peaking is used or what is for
To begin with uw nikkor lenses are for underwater use only and don’t work properly on land as the correction impair severe field of curvature and aberrations
With regards to live view or not at the exposure settings of 1/4 i expect the scene is properly exposed so no it won’t make any difference but if you were at 1/125 try to look at the screen with live view on or off
Generally you need to have a clear subject for focus peaking and similar to what @Alex_Mustard was suggesting open the lens move the focus so you can see this clear target go in and out of focus and then close down because if you are already stopped down peaking generates false positives by definition
In my case in the pool I was not stopped down and I tried a test target that was flat to avoid any depth of field issues
On that target i tried the various settings and despite seeing the target going in and out of focus I didn’t achieve great results as I was in dark scene I expect that the brightness given by not using live view was giving me false positives
When I shoot a focus chart on land with a normal lens peaking does work well enough but if you think that you can rely on it to confirm the whole frame is in focus you will be disappointed
For me critical focus means a clear subject is in focus the rest may or not be depending on what I want to achieve i take plenty of photos and videos wide open
You can’t talk of critical focus on a scene like that in your example: peaking doesn’t work well for something like that.
Situation where there is a wide scene and you want a sub detail to be in focus need by definition a focus magnifier.
But this was not my use case I had divers in a pool and was trying to focus on their eyes mostly as it is was a normal photo on land
Part of the challenge was those guys were static at times moving other times so instead of changing focus I was swimming along
Peaking doesn’t fit well wide angle scenes with non clear subject nor moving targets where you can’t keep up with focus
When I use a focus pull on land peaking works fine and am at f/2.8 on long lenses with little depth of field in a properly exposed scene
Throw a fisheye lens in this scenario and no clear subject the results will be all seems in focus. Peaking not the tool for that but it never was you need to have a clear subject to put in and out of focus that you can see.
If you can’t even see it on the display because it is a tiny detail you are not in a good place
And of course magnifier is better but distracting so better for more static subjects
In conclusion i agree that to get something like in your example in focus this is not the tool but also that you don’t use peaking like that anyway


Sent from my iPhone using Tapatalk