505V Auto Focus Challenge

Since disappointed badly with the birds, I have been in doubt about the 505V AF. With Marcel's great trick ( locating the cross marker on the high contrast spot ), we entered the new chapter of the issue. Now, can I go back to the birds to revenge with this new trick ? I simulated the setting at my backyard instead ;-)

• Multiple branches, high contrast background. And low contrast subject about. The balls are very similar to the birds in terms of the contrast and the colors. As you see AF was confused.

Then, I tried spot metering. I finally got the focus after several trys. I rotated the frame so that the spot marker was located as shown below. After the Green-light lock, I rotated the camera back.



The key was to try on the multiple high-contrast spots until the focus was found. One down side of this method is that the exposure could go noticeably wrong.

Pondria said:

I know you are tired of this - Again ? I thought I knew how to work
around the auto focus problem. I went out to a small zoo where they
showed colorful birds outside. I took pictures of them like crazy. Then,
I realized that I didn't get the right focus. About 30 or so shots were
all out of focus. So, I tried all kinds of techniques that I could think
of. Only manual focus worked. I mean , I guessed the distance. It was so
shocking to me as the birds were so COLORFUL. They have body areas in
very different colors. Here is one of the out-of-focus picture.

http://www.zing.com/picture/p505417199eddb697de3e6c506120b0bc/ff640255.jpg.orig.jp g
As you see the focus was on the trees at the back. But the birds have
enough distinctive regions for focus to work. Why not ? Here goes my
theory again. I think that the auto focus uses only the black-and-white
contrast ! It's just like looking throgh the old B/W camcoder view
finder. Here is the same picture in b/w.

http://www.zing.com/picture/pddb6d3fb3f7552e876a3cdbe6a5a3471/ff640254.jpg.orig.jp g
Although the birds look very clear in colors, they are very dull in b/w.
And the branches at the back look very vivid. This explains why
sometimes the kids wearing colorful T-shirt couldn't get the focus (
Curt A. would you agree ? ). So, when you frame the shot, you should
really think in terms of b/w to avoid the potential focus problems. Too
bad, isn't it ?

Click to expand...

Pondria said:

Since disappointed badly with the birds, I have been in doubt about the
505V AF. With Marcel's great trick ( locating the cross marker on the
high contrast spot ), we entered the new chapter of the issue. Now, can
I go back to the birds to revenge with this new trick ? I simulated the
setting at my backyard instead ;-)

• Multiple branches, high contrast background. And low contrast subject

about. The balls are very similar to the birds in terms of the contrast
and the colors. As you see AF was confused.



Then, I tried spot metering. I finally got the focus after several trys.
I rotated the frame so that the spot marker was located as shown below.
After the Green-light lock, I rotated the camera back.



The key was to try on the multiple high-contrast spots until the focus
was found. One down side of this method is that the exposure could go
noticeably wrong.

Click to expand...

I really like your new parrots

I did some more tests and found that after selecting an edge you better don't move the camera to much. There seems to be some sort of delay, probably the delay to adjust the focussing. This results in another disadvantage. Moving objects are difficult to catch. I don't think this is an extra disadvantage as you have to deal with the lag of the camera already. A non-frozen LCD will be a great help. As a matter of fact I find this the worst behavior of the camera.

Anyway the spotmetering solves many non-focus problems and I am glad that I could help you all to get better results as I think the camera really is a great instrument, especially comparing with the results I had with non-digital but expensive camera's.

Member said:

Pondria said:

I know you are tired of this - Again ? I thought I knew how to work
around the auto focus problem. I went out to a small zoo where they
showed colorful birds outside. I took pictures of them like crazy. Then,
I realized that I didn't get the right focus. About 30 or so shots were
all out of focus. So, I tried all kinds of techniques that I could think
of. Only manual focus worked. I mean , I guessed the distance. It was so
shocking to me as the birds were so COLORFUL. They have body areas in
very different colors. Here is one of the out-of-focus picture.

http://www.zing.com/picture/p505417199eddb697de3e6c506120b0bc/ff640255.jpg.orig.jp g
As you see the focus was on the trees at the back. But the birds have
enough distinctive regions for focus to work. Why not ? Here goes my
theory again. I think that the auto focus uses only the black-and-white
contrast ! It's just like looking throgh the old B/W camcoder view
finder. Here is the same picture in b/w.

http://www.zing.com/picture/pddb6d3fb3f7552e876a3cdbe6a5a3471/ff640254.jpg.orig.jp g
Although the birds look very clear in colors, they are very dull in b/w.
And the branches at the back look very vivid. This explains why
sometimes the kids wearing colorful T-shirt couldn't get the focus (
Curt A. would you agree ? ). So, when you frame the shot, you should
really think in terms of b/w to avoid the potential focus problems. Too
bad, isn't it ?

Click to expand...

Click to expand...

Pondria said:

Since disappointed badly with the birds, I have been in doubt about the
505V AF. With Marcel's great trick ( locating the cross marker on the
high contrast spot ), we entered the new chapter of the issue. Now, can
I go back to the birds to revenge with this new trick ? I simulated the
setting at my backyard instead ;-)

• Multiple branches, high contrast background. And low contrast subject

about. The balls are very similar to the birds in terms of the contrast
and the colors. As you see AF was confused.



Then, I tried spot metering. I finally got the focus after several trys.
I rotated the frame so that the spot marker was located as shown below.
After the Green-light lock, I rotated the camera back.



The key was to try on the multiple high-contrast spots until the focus
was found. One down side of this method is that the exposure could go
noticeably wrong.

Click to expand...

Pondria, that was an interesting experiment. I am approaching this problem from a different direction having tried the "spot" method for focus then switching to manual mode to allow me to utilise the average exposure metering.

I have just ordered a "rangefinder" (popular in the sixties onwards for use with 35mm cameras that did not have a built in rangefinder) for use with my F505V one slight problem is that they are calculated in feet rather than metres, which is not a great problem.

The rangefinder will be used when the camera is in manual focus mode and is just a case of transferring the distance calculated by the rangefinder to the camera lens.

When finding the focus point it would appear preferable to have the camera lens set at maximum zoom, this gives 94 focus points which range from .8m to 10m which increase in increments of .1m. After the 10m setting it then goes to infinity.

By comparison the wide-angle setting of the lens gives 27 focus points (which is reasonable as the wide-angle settings gives greater depth of field) starting at .1m and also going to 10m and then infinity but from 1m onwards the focus steps are in metres, ie: 2m, 3m etc upto 10m.

Therefore it would be reasonable to assume that for criticial focus it is best to focus at the maximum zoom setting and then re-zoom to frame the picture you want.

I am sure this focussing problem could be overcome in software if only Sony would release a software update, it is not that hard to update software if Nikon can do it via USB I fail to see why Sony cannot do the same. Todate I have "flashed" my motherboard and video card with software updates and provided instructions are followed to the letter there is no problem. Utilising USB is should be even easier.

A number of people have not seen a "rangefinder" device before so once I have received it I will post a couple of images.

Aitch

Aitch said:

Pondria, that was an interesting experiment. I am approaching this
problem from a different direction having tried the "spot" method for
focus then switching to manual mode to allow me to utilise the average
exposure metering.

Click to expand...

Yes, I tried this, too. It needs more steps but works. If we are shooting a moving (slowly) subject, it becomes more difficult.

Member said:

I have just ordered a "rangefinder" (popular in the sixties onwards for
use with 35mm cameras that did not have a built in rangefinder) for use
with my F505V one slight problem is that they are calculated in feet
rather than metres, which is not a great problem.

Click to expand...

Let us know the result ! If you are willing to go through the extra steps to get the focus right, the device I made may be helpful, too. As Andreas suggested, you can drill a hole on the 52mm lens cap. I later found that the hole size should be 5-7 mm instead of 10mm. http://www.dpreview.com/forums/read.asp?forum=1009&message=457061

Member said:

When finding the focus point it would appear preferable to have the
camera lens set at maximum zoom, this gives 94 focus points which range
from .8m to 10m which increase in increments of .1m. After the 10m
setting it then goes to infinity.

By comparison the wide-angle setting of the lens gives 27 focus points
(which is reasonable as the wide-angle settings gives greater depth of
field) starting at .1m and also going to 10m and then infinity but from
1m onwards the focus steps are in metres, ie: 2m, 3m etc upto 10m.

Click to expand...

Very accurate observation ! Thank you for the info.

Member said:

Therefore it would be reasonable to assume that for criticial focus it
is best to focus at the maximum zoom setting and then re-zoom to frame
the picture you want.

I am sure this focussing problem could be overcome in software if only
Sony would release a software update, it is not that hard to update
software if Nikon can do it via USB I fail to see why Sony cannot do the
same. Todate I have "flashed" my motherboard and video card with
software updates and provided instructions are followed to the letter
there is no problem. Utilising USB is should be even easier.

A number of people have not seen a "rangefinder" device before so once I
have received it I will post a couple of images.

Aitch

Click to expand...

Good test Pondria! I continue to be amazed by the level of honest testing that goes on here in this forum.

Member said:

One down side of this method is that the exposure could go
noticeably wrong.

Click to expand...

But generally, the exposure will be suitable for the subject in question, right? It would seem that most situations with this sort of contrasting lighting will require the spot to properly expose the subject on hand. Of course, there could be all sorts of scenarios that would trick the exposure routines.

Hi there Marcel. Yes, you have contributed greatly with this one small but not insignificant finding.

MarcelX said:

I think the camera
really is a great instrument, especially comparing with the results I
had with non-digital but expensive camera's.

Click to expand...

Interesting. A point that we should keep in mind, especially considering that we usually have to wait a rather long period of time to get our results back with a film camera, whereas with the digital instrument, we get to pick apart and analyze the results almost immediately, allowing us to be much more picky about the results. It isn't perfect, but I get the results that I want far more often than not. And I'm learning to get even better with threads such as this.

Ulysses said:

Member said:

I think the camera
really is a great instrument, especially comparing with the results I
had with non-digital but expensive camera's.

Click to expand...

Interesting. A point that we should keep in mind, especially considering
that we usually have to wait a rather long period of time to get our
results back with a film camera, whereas with the digital instrument, we
get to pick apart and analyze the results almost immediately, allowing
us to be much more picky about the results.

Click to expand...

Excellent point ! I absolutely agree.

Member said:

It isn't perfect, but I get
the results that I want far more often than not. And I'm learning to get
even better with threads such as this.

Click to expand...

As I started this thread, I owe the group the final result with the birds. I visited the birds again today with the new "Marcel's focus tool". Was I successful ? No. I took 96 pictures. I was very careful about the focus. I tried locating the spot meter market on various high-contrast edges. I had to throw away 51 as they were out of focus ! You may think I'm whining too much about the focus. Please, visit the local Zoo and try Macaw's. You will be surprised. Attached are the pictures that shows different type of focus challenges. Especially the 3rd one with Red bird was very difficult to use AF. AF just won't give any respect to the red object !

Pondria said:

• Multiple branches, high contrast background. And low contrast subject

about. The balls are very similar to the birds in terms of the contrast
and the colors. As you see AF was confused.

http://www.zing.com/picture/p61d2cbf6a1cfc0377a05b86e291dafb3/ff5f13b7.jpg.orig.jp g

Then, I tried spot metering. I finally got the focus after several trys.
I rotated the frame so that the spot marker was located as shown below.
After the Green-light lock, I rotated the camera back.

http://www.zing.com/picture/p9f5a907741f726b7fc1b769fb4edde9c/ff5f13b8.jpg.orig.jp g

The key was to try on the multiple high-contrast spots until the focus
was found. One down side of this method is that the exposure could go
noticeably wrong.

Pondria said:

I know you are tired of this - Again ? I thought I knew how to work
around the auto focus problem. I went out to a small zoo where they
showed colorful birds outside. I took pictures of them like crazy. Then,
I realized that I didn't get the right focus. About 30 or so shots were
all out of focus. So, I tried all kinds of techniques that I could think
of. Only manual focus worked. I mean , I guessed the distance. It was so
shocking to me as the birds were so COLORFUL. They have body areas in
very different colors. Here is one of the out-of-focus picture.

http://www.zing.com/picture/p505417199eddb697de3e6c506120b0bc/ff640255.jpg.orig.jp g
As you see the focus was on the trees at the back. But the birds have
enough distinctive regions for focus to work. Why not ? Here goes my
theory again. I think that the auto focus uses only the black-and-white
contrast ! It's just like looking throgh the old B/W camcoder view
finder. Here is the same picture in b/w.

http://www.zing.com/picture/pddb6d3fb3f7552e876a3cdbe6a5a3471/ff640254.jpg.orig.jp g
Although the birds look very clear in colors, they are very dull in b/w.
And the branches at the back look very vivid. This explains why
sometimes the kids wearing colorful T-shirt couldn't get the focus (
Curt A. would you agree ? ). So, when you frame the shot, you should
really think in terms of b/w to avoid the potential focus problems. Too
bad, isn't it ?

Click to expand...

Click to expand...

Pondria -

Heheh... I don't think you're whining. I do believe that it shouldn't have taken you 96 pictures to prove the point even to yourself.

A long time ago, you offered to me the suggestion with the BOCS problem that this was one situation in which we had an easily remedied solution by simply clicking the AE button and switching to any of the other modes when indoors and using a flash. That was a surprisingly simple answer and way of looking at what I perceived at that time as a real problem. If you'll remember, I was forgetting that I don't shoot many indoor pictures, and I therefore don't need the flash that often for this situation. But when I do, I can simply switch Program AE modes. Easy!

Now, you've been stabbing yourself all over with the bird feathers. You shot 96 images when perhaps 10 would have sufficed. Your experiment with the fluffy balls provided pretty much similar results. Even the Spot-focus method isn't fool-proof. But it works most of the time for most people in most situations. Yes, there are times when you can fool even this method.

However, the solution is very easy once again if we can check our image focus via the Preview feature. If we see that there is a pattern during that shooting session of not focusing on a specific subject, then it is simply time to flick the Focus switch from "Auto" to "Manual". Easily done. Doesn't detract from the camera too much. And I have a lot more time to shoot 96 in-focus images.

Now, one might say that this is too much of a burden in an $900 camera. But that's what Manual focus is for -- the situations where Autofocus doesn't work. When a better camera comes along for me, I'll probably buy it. For now, I know that the Autofocus doesn't work as well as my own hand, eye and brain. I'll work with it until some all-around camera makes significant improvements.

I know this doesn't help the frustration or the agony. But it is essentially the same hand that you offered me several months ago. I now enjoy things much more. I don't try to find difficult work-arounds to implement. I just try to enjoy it and concentrate on what my eyes see in the light. I'd like to be as accomplished a photographer as Phil or some of the others here. The camera allows me to work at this and have fun at the same time. It's not perfect. But neither am I.