Idea on benchmarking image stabilization ?

[Kolia]

I'd like to compare the OS in my Sigma lenses to the Steady Shot in my camera body.

Any suggestions on how I could set that up ? Things to look for ?

Thanks !

 

[Overread]

It's difficult.

If you just want to compare it at home the best way would be to put yourself in a fixed lighting condition, such as bright sunlight or heavy overcast - ie conditions where its not changing.

Then set your aperture and ISO to fixed values with the ISO on its lowest setting.

After that I would take a series of test photos at varying shutter speeds close to the handholding limit for the lens (Based on the rough 1/focal length for shutter speed rule of thumb). Note that you'd want to take a burst of a few frames at each shutter speed so that you can ensure that the blurr you see is from handholding and not your finger pressing on the shutter button.

That would let you see how slow you can take the shutter speed and, comparing the photos in the computer, show you at what point the blur from handholding becomes to extreme.

Of course you'll want to do this without being fatigued and you'll want to rest between bursts as needed so that you're hold of the camera is as equal as possible each time.

 

[pgriz]

^^^ looks like a good suggestion. I'd add that putting a bright point-like source (bare flashlight bulb?) in the image will help detecting whether there is any motion when you blow up the image on the computer.

 

[nanhi]

Hello Kolia Sir, here is a simple very accurate method that uses the DSLRs mirror and shutter slap vibration. Cheaply and easily made too. Also the results are not affected by you or your hands as you use a remote to fire the camera. Here is what you do: make a flat tin plate spring by folding a 6 inch wide by 36 inch long tin sheet. Fold a 6 inch length once up, then the next 6 inch length down and so on - fan fold style. Place it on a table and set you camera on your contraption. Touch the camera gently to see if you get a good swing / bobbing motion. Now use the widest aperture and a shutter speed of 1/4 sec. Put some beer bottles as the test subject. Use the remote - wired or wireless - to fire the camera.
Make sure you have the lens SR (shake reduction) ON and the camera SR off, and then vice versa. Now compare the photos. Don't forget to take some shots w/o using the contraption. Enjoy.
Regards. Bipin - from that far away enchanting land.

 

[DiskoJoe]

Hello Kolia Sir, here is a simple very accurate method that uses the DSLRs mirror and shutter slap vibration. Cheaply and easily made too. Also the results are not affected by you or your hands as you use a remote to fire the camera. Here is what you do: make a flat tin plate spring by folding a 6 inch wide by 36 inch long tin sheet. Fold a 6 inch length once up, then the next 6 inch length down and so on - fan fold style. Place it on a table and set you camera on your contraption. Touch the camera gently to see if you get a good swing / bobbing motion. Now use the widest aperture and a shutter speed of 1/4 sec. Put some beer bottles as the test subject. Use the remote - wired or wireless - to fire the camera.
Make sure you have the lens SR (shake reduction) ON and the camera SR off, and then vice versa. Now compare the photos. Don't forget to take some shots w/o using the contraption. Enjoy.
Regards. Bipin - from that far away enchanting land.

This is not going to work. The stabilizer is specifically designed to be used handheld and using it in a fixed location such as a tripod or anything else of the sorts will actually have negative effects on the picture quality.


But I have that same lens as you Kolia. Trust me there is not much difference. Personally I choose to use the in-camera stabilization for a few reasons.

1. you dont have to remember to turn it off. With the sigma stabilizer if you forget to turn it off and give it time for the lens motor to turn off you can damage the lens and the AF motor. it takes a while to turn off too.

2. The in-camera stabilizer has a read out that can be seen on the viewscreen and give you a visual reference about when you are holding the camera more steady where the lens stabilizer does not do this so you still dont know when it is working best.