-
Welcome to BirdForum, the internet's largest birding community with thousands of members from all over the world. The forums are dedicated to wild birds, birding, binoculars and equipment and all that goes with it.
Please register for an account to take part in the discussions in the forum, post your pictures in the gallery and more.
Use of image stabilization (1 Viewer)
- Thread starter rferrieri
- Start date
DanC.Licks
AKA Daniel Bradley
E-30 on SW80/600, on tripod with 10 second delay (AntiShake, mirror up) and wireless remote.
IS off
IS on
I chucked most of the tests as soon as I saw what I wanted to see, but forgot to delete these. They are crops of these full frames:
Taken through the window and snowing and bad light, but the results were consistent.
IS off
IS on
I chucked most of the tests as soon as I saw what I wanted to see, but forgot to delete these. They are crops of these full frames:
Taken through the window and snowing and bad light, but the results were consistent.
Tord
Well-known member
Hi Dan,
For a test to reveal the effect of IS I believe you need to conduct the test in a different way, mimic real conditions by shooting while holding the camera body, no shutter delay/mirror lock (meaning introducing some vibrations which effect the IS is supposed to attenuate)
Cheers
/Tord
For a test to reveal the effect of IS I believe you need to conduct the test in a different way, mimic real conditions by shooting while holding the camera body, no shutter delay/mirror lock (meaning introducing some vibrations which effect the IS is supposed to attenuate)
Cheers
/Tord
DanC.Licks
AKA Daniel Bradley
I have done plenty of those and they are much more in favor of IS, especially if the camera and scope are free to move. I wanted to know what the difference is under the best of conditions, i.e., whether the IS actually causes blurring with the camera and scope locked down.
DanC.Licks
AKA Daniel Bradley
No idea. I have zero experience with P+S cameras. I know that the IS causes problems with my 200mm lens mounted and locked down, but it seems to be the other way around when I get up to 600 or more. I will do some more tests and post them.
Tord
Well-known member
Now I follow.
/Tord
DanC.Licks
AKA Daniel Bradley
Well, I did some more tests and they show more than anything else that I don't know what I am talking about:-C
These are at ISO 400 and 1/125, E-30 on the SW 80/600. Hands on!
Maybe I have gotten different results at different shutter speeds, can't remember. But this is an ISO setting I need with the E-30, and a shutter speed on the lower limits. Leads me to think that the best thing is to get the ISO up high, especially for birds, and forget the IS.
These are at ISO 400 and 1/125, E-30 on the SW 80/600. Hands on!
Maybe I have gotten different results at different shutter speeds, can't remember. But this is an ISO setting I need with the E-30, and a shutter speed on the lower limits. Leads me to think that the best thing is to get the ISO up high, especially for birds, and forget the IS.
Last edited:
Tord
Well-known member
Hi Dan
I don't think you can draw conclusions from these few samples.
If you read through the description of the method advocated by Wrotniak <link> you will realize that there is a need for a number of samples to mitigate stochastic variations. I think he recommends series of 20. I did the test with an arbitrary amount of 10 shots for each setting in the graph posted on previous page.
At 1/125s I guess you are outside the safe zone (providing virtually 100% probablity for sharp picture) and one single frame is not enough.
As soon as I find the time I will run the test on my SW80 on a gimbal head, hands on.
/Tord
I don't think you can draw conclusions from these few samples.
If you read through the description of the method advocated by Wrotniak <link> you will realize that there is a need for a number of samples to mitigate stochastic variations. I think he recommends series of 20. I did the test with an arbitrary amount of 10 shots for each setting in the graph posted on previous page.
At 1/125s I guess you are outside the safe zone (providing virtually 100% probablity for sharp picture) and one single frame is not enough.
As soon as I find the time I will run the test on my SW80 on a gimbal head, hands on.
/Tord
DanC.Licks
AKA Daniel Bradley
These are each the best of seven frames. I did the same test twice to confirm my observations.
Unfortunately, I often need such shutter speeds due to the relativly poor high ISO noise levels in the E-30. I am happy when I can get up to 1/500 at ISO 400.
Unfortunately, I often need such shutter speeds due to the relativly poor high ISO noise levels in the E-30. I am happy when I can get up to 1/500 at ISO 400.
DanC.Licks
AKA Daniel Bradley
Tord,
I see what you are saying. You are suggesting a more statistical approach with larger samples evaluated (good, bad, ugly) and averaged out (or the percentage of "keepers" calculated) to plot on a graph. No doubt the right way to go about it.
I rather doubt that the results would be that much different though. My overall impression is that the shots I did yesterday with IS on were clearly much worse that the shots with it off, at least at 1/125. The "best" ON shot was below the average Off shot. I had about 3 good OFF shots to choose from, whereas I had a hard time finding a "good" ON shot. That is enough for me to decide whether to turn it on or not.
The light is getting better, so maybe I can do some more today at faster shutter speeds, up to about 1:1 (1/600 with the 80/600).
I am sure lots of people have done these tests, but somehow we seem to feel the need to do them ourselves to be sure. I believe what I see, but I don't believe everything I read!
I see what you are saying. You are suggesting a more statistical approach with larger samples evaluated (good, bad, ugly) and averaged out (or the percentage of "keepers" calculated) to plot on a graph. No doubt the right way to go about it.
I rather doubt that the results would be that much different though. My overall impression is that the shots I did yesterday with IS on were clearly much worse that the shots with it off, at least at 1/125. The "best" ON shot was below the average Off shot. I had about 3 good OFF shots to choose from, whereas I had a hard time finding a "good" ON shot. That is enough for me to decide whether to turn it on or not.
The light is getting better, so maybe I can do some more today at faster shutter speeds, up to about 1:1 (1/600 with the 80/600).
I am sure lots of people have done these tests, but somehow we seem to feel the need to do them ourselves to be sure. I believe what I see, but I don't believe everything I read!
Tord
Well-known member
Hi Dan,
What I did was quite similar to what Wrotniak did but slightly different since I have "problems" with arbitrarily giving 1 point for a sharp picture and 1/2 point for a semi-sharp picture.
What I instead did was to include the collection of sharp pictures into the semi-sharp collection. My reasoning was that I would like to find out what shutter speed I need to use to get, with a certain probability, a sharp shot or an acceptable record shot. Sometimes the situation gives you the opportunity to shoot many frames, sometimes you only have a small window of opportunity and will be able to take one picture only.
When using scopes with manual focus, probably you take a sequence of several frames with slight focus adjustment between to make sure at least one has is shot at the desired optimal focal plane.
If you have the opportunity probably you want take several pictures at a given focus in order to reduce the risk of shake blur if you are out of the safe zone. For a sharp pictures both conditions need to be met.
/Tord
What I did was quite similar to what Wrotniak did but slightly different since I have "problems" with arbitrarily giving 1 point for a sharp picture and 1/2 point for a semi-sharp picture.
What I instead did was to include the collection of sharp pictures into the semi-sharp collection. My reasoning was that I would like to find out what shutter speed I need to use to get, with a certain probability, a sharp shot or an acceptable record shot. Sometimes the situation gives you the opportunity to shoot many frames, sometimes you only have a small window of opportunity and will be able to take one picture only.
When using scopes with manual focus, probably you take a sequence of several frames with slight focus adjustment between to make sure at least one has is shot at the desired optimal focal plane.
If you have the opportunity probably you want take several pictures at a given focus in order to reduce the risk of shake blur if you are out of the safe zone. For a sharp pictures both conditions need to be met.
/Tord
DanC.Licks
AKA Daniel Bradley
IS-1 on the E-30 is vertical and horizontal stabilization. I set the IS to 600mm.
More testing using a slightly different method. I have been using an ISO 12233 chart. In the lower left corner is a cross type resolution scale. Because IS is supposed to take care of both vertical and horizontal movement, it stands to reason that there would be differences in the resolution depending on which way the lens moves. That is, if there is vertical movement, the vertical scale will be better than the horizontal scale. So here is what I did.
I took groups of 10 shots at different ISO settings and with IS-1 on and off, with the crossed scale in the center of the frame. Then I averaged out the resolution of the v and h scales of each shot. Say a shot was clear at 10 on one axis but only clear up to 8 on the other, I gave it a 9. This made evaluating the shots much less subjective. Then I added up the total points and divided by 10.
In general the average came out about half a point better without IS off (around 7.5 to 8), BUT, there was a much larger range. With IS off I was getting a mix of 6s to 10s, whereas with it on I was getting 7s to 9s. So even though the averages were not all that much different, the only really clear shots were with IS off, at the expense of more really bad ones. IS on was more even, but none as good as the best with IS off.
At ISO 1600 there is no point in testing because the noise in the sensor completely kills the resolution. 800 was OK, 400 pretty good and 200 best, but naturally with more bad ones the lower the ISO goes.
I am more convinced than ever that high ISO is the way to go, although the IS on the OM-D is supposed to be much better. Give Carlos (cango) something to do in his spare time.
More testing using a slightly different method. I have been using an ISO 12233 chart. In the lower left corner is a cross type resolution scale. Because IS is supposed to take care of both vertical and horizontal movement, it stands to reason that there would be differences in the resolution depending on which way the lens moves. That is, if there is vertical movement, the vertical scale will be better than the horizontal scale. So here is what I did.
I took groups of 10 shots at different ISO settings and with IS-1 on and off, with the crossed scale in the center of the frame. Then I averaged out the resolution of the v and h scales of each shot. Say a shot was clear at 10 on one axis but only clear up to 8 on the other, I gave it a 9. This made evaluating the shots much less subjective. Then I added up the total points and divided by 10.
In general the average came out about half a point better without IS off (around 7.5 to 8), BUT, there was a much larger range. With IS off I was getting a mix of 6s to 10s, whereas with it on I was getting 7s to 9s. So even though the averages were not all that much different, the only really clear shots were with IS off, at the expense of more really bad ones. IS on was more even, but none as good as the best with IS off.
At ISO 1600 there is no point in testing because the noise in the sensor completely kills the resolution. 800 was OK, 400 pretty good and 200 best, but naturally with more bad ones the lower the ISO goes.
I am more convinced than ever that high ISO is the way to go, although the IS on the OM-D is supposed to be much better. Give Carlos (cango) something to do in his spare time.
Tord
Well-known member
Dan,
You method has elements that I haven't thought of and are worth exploiting. I am thinking about measuring the X and Y resolution and better evaluating the sharpness of a picture.
Have you though about and made calculations for what "perfect sharpness" would mean in terms of resolution, in our application area (bird photography)? And what resolution a reference picture should have, to qualify for "sharp" and "acceptable". Again, in the context of bird photography.
/Tord
You method has elements that I haven't thought of and are worth exploiting. I am thinking about measuring the X and Y resolution and better evaluating the sharpness of a picture.
Have you though about and made calculations for what "perfect sharpness" would mean in terms of resolution, in our application area (bird photography)? And what resolution a reference picture should have, to qualify for "sharp" and "acceptable". Again, in the context of bird photography.
/Tord
DanC.Licks
AKA Daniel Bradley
Tord,
Such things are all so relative. Naturally we all want pictures as sharp as possible, but there are so many factors involved that in the end we have to pick the best of what we have.
Here is how I checked the x and y resolution:
good
different
Such things are all so relative. Naturally we all want pictures as sharp as possible, but there are so many factors involved that in the end we have to pick the best of what we have.
Here is how I checked the x and y resolution:
good
different
