[Leumas]

I'm curious if anyone has tested this, but I was just doing a macro photowalk and I turned off my IBIS 'cause my 50mm f4 Macro has SOOOOOOO much focus breathing that it's true focal length is unknown. lol

Since the IBIS relies on a set focal length to calculate it's movement, I'm wondering if anyone has experimented with IBIS actually degrading image quality due to either focus breathing or perhaps a lens that doesn't report it's true focal length to the camera accurately.

Thoughts? GO!

[swanlefitte]

That is interesting because focus breathing always reduces and ibis is said to be better under not over with manual input lenses.

[photoptimist]

Good observation! You are right about macro focus breathing and the SR setting.

The SR system uses the reported or entered "focal length" to relate angular pitch and yaw motion to the estimated up/down and left-right shifts in the image which it then corrects.

In this regard, the correct "focal length" for a macro lens would change with magnification. A 50mm lens racked out to 1:1 magnification acts like a 100mm lens for SR purposes.

However, the effect of setting SR to "50mm" and then going to 1:1 magnification is that the SR system will under-correct the pitch and yaw motion by 50%. It will still perfectly correct any roll motion. Thus, you'll still get at least 1 stop of SR performance.

For macro lenses, SR will never degrade the image, it just won't do as good a job as when the lens was at ∞.

For zoom lenses, it may be a different story. Many zooms show a decrease in effective focal length when zoomed to the max and focused close up. If the actual focal length is less than the reported or entered value, the SR system over-compensates any motion. But the net result won't cause degradation unless the effective focal length is less than half the reported focal length.

[othar]

I'm curious if anyone has tested this, but I was just doing a macro photowalk and I turned off my IBIS 'cause my 50mm f4 Macro has SOOOOOOO much focus breathing that it's true focal length is unknown. lol

Since the IBIS relies on a set focal length to calculate it's movement, I'm wondering if anyone has experimented with IBIS actually degrading image quality due to either focus breathing or perhaps a lens that doesn't report it's true focal length to the camera accurately.

Thoughts? GO!

While SR won't work as efficient in macro mode as you might have expected it doesn't add more motion blur because it will actually undercompensate the movement and not overcompensate.
I usually increase shutterspeed when doing macro handheld to compensate this behavior.
The same problem occurs when using extension tubes (there was some discussion before on the forum related to this topic)

[othar]

I am not sure this is related to focus breathing in general though, but has more to do with the reached magnification when you enter closeup or macro photography

[UncleVanya]

Good observation! You are right about macro focus breathing and the SR setting.

The SR system uses the reported or entered "focal length" to relate angular pitch and yaw motion to the estimated up/down and left-right shifts in the image which it then corrects.

In this regard, the correct "focal length" for a macro lens would change with magnification. A 50mm lens racked out to 1:1 magnification acts like a 100mm lens for SR purposes.

However, the effect of setting SR to "50mm" and then going to 1:1 magnification is that the SR system will under-correct the pitch and yaw motion by 50%. It will still perfectly correct any roll motion. Thus, you'll still get at least 1 stop of SR performance.

For macro lenses, SR will never degrade the image, it just won't do as good a job as when the lens was at ∞.

For zoom lenses, it may be a different story. Many zooms show a decrease in effective focal length when zoomed to the max and focused close up. If the actual focal length is less than the reported or entered value, the SR system over-compensates any motion. But the net result won't cause degradation unless the effective focal length is less than half the reported focal length.

The 60-250 acts like a 135 when focus is close... But I've never noticed a problem with SR at those distances. Interesting, I'm going to look for it in the future.

[Leumas]

Good observation! You are right about macro focus breathing and the SR setting.

The SR system uses the reported or entered "focal length" to relate angular pitch and yaw motion to the estimated up/down and left-right shifts in the image which it then corrects.

In this regard, the correct "focal length" for a macro lens would change with magnification. A 50mm lens racked out to 1:1 magnification acts like a 100mm lens for SR purposes.

However, the effect of setting SR to "50mm" and then going to 1:1 magnification is that the SR system will under-correct the pitch and yaw motion by 50%. It will still perfectly correct any roll motion. Thus, you'll still get at least 1 stop of SR performance.

For macro lenses, SR will never degrade the image, it just won't do as good a job as when the lens was at ∞.

For zoom lenses, it may be a different story. Many zooms show a decrease in effective focal length when zoomed to the max and focused close up. If the actual focal length is less than the reported or entered value, the SR system over-compensates any motion. But the net result won't cause degradation unless the effective focal length is less than half the reported focal length.

Interesting! Good to know that's how it works! On a side note I have had to turn off horizon correction for composite mode, it introduced a lot of image blur.

I am not sure this is related to focus breathing in general though, but has more to do with the reached magnification when you enter closeup or macro photography

I dunno....the focal length is changing based on focusing...I would consider that focus breathing.

The 60-250 acts like a 135 when focus is close... But I've never noticed a problem with SR at those distances. Interesting, I'm going to look for it in the future.

Well according to photoptimist...that might introduce some additional motion blur with SR rather than help. Let us know what you find

[UncleVanya]

Well according to photoptimist...that might introduce some additional motion blur with SR rather than help. Let us know what you find

I rarely shoot that lens at short distances but I will keep this in mind.

[mikesbike]

well, this is very interesting- what actually is "focus breathing"? Does it actually change the focal length of a lens? Or does the lens "act like" a different focal length? In the case of APS-C vs. FF a lens gives a cropped image, making it "act like" a lens of a different focal length. In any case, if we are using modern AF lenses, the lens info is fed to the camera system. We don't have to feed in the FL manually. Would it not be possible for the SR system to be set up within the camera system to detect which lens is being used, its FL, whether APS-C or FF is being used, and what the properties of the lens may be, including the distance to subject and the "focus breathing" characteristics of the lens?

The "focus breathing" phenomenon has been known for a long time, even before SR came along.

[Kerrowdown]

Focus breathing

You mean my "Ladies" really are alive when they're in my hands...

[photoptimist]

well, this is very interesting- what actually is "focus breathing"? Does it actually change the focal length of a lens? Or does the lens "act like" a different focal length? In the case of APS-C vs. FF a lens gives a cropped image, making it "act like" a lens of a different focal length. In any case, if we are using modern AF lenses, the lens info is fed to the camera system. We don't have to feed in the FL manually. Would it not be possible for the SR system to be set up within the camera system to detect which lens is being used, its FL, whether APS-C or FF is being used, and what the properties of the lens may be, including the distance to subject and the "focus breathing" characteristics of the lens?

The "focus breathing" phenomenon has been known for a long time, even before SR came along.

The answer depends on:
1) the definition of focal length (refraction versus angle-of-view definitions); and
2) the type of lens (simple primes versus everything else).

A physicist's optical definition of focal length measures how the lens refracts light and defines it in terms of a mathematical relationship between the distance to the in-focus object out in the world and the distance to the in-focus image inside the camera. A physicist or optical engineer would say that the focal length of a simple prime lens DOES NOT change when you focus from far to near, even in the case of a macro lens.

However a photographer's angle-of-view definition of effective focal length (how wide is the view) and the definition of focal length for SR purposes does imply that "effective focal length" does change with the focusing distance and any focus breathing does involve a change in "effective focal length". The SR system needs to know the angle-of-view of a pixel so that it can convert a measured angle of shake motion into some number of pixels of corrective shift movement. (Note:; the SR system does not care whether the sensor is full frame or APS-C -- the amount of required sensor shift to correct pitch and yaw shake doesn't depend on the format size)

For zoom lenses and complex internal-focus primes, both the physical and the angular measures of focal length can change as the zoom and focus mechanisms shift the internal elements. Note that whereas the effective focal length of a prime tends to increase with decreasing distance, the effective focal length of many zooms at the long end tends to decrease with decreasing distance.

As for your final question about automating this, the answer is yes. For newer Pentax lenses, the data pin tells the camera about the model number, focal length, and approximate focus distance of the lens. One can presume (or hope) that Pentax engineers created lens-specific tables for what SR focal length to use for a given lens, zoom focal length, and focus distance setting.

[othar]

well, this is very interesting- what actually is "focus breathing"? Does it actually change the focal length of a lens? Or does the lens "act like" a different focal length? In the case of APS-C vs. FF a lens gives a cropped image, making it "act like" a lens of a different focal length. In any case, if we are using modern AF lenses, the lens info is fed to the camera system. We don't have to feed in the FL manually. Would it not be possible for the SR system to be set up within the camera system to detect which lens is being used, its FL, whether APS-C or FF is being used, and what the properties of the lens may be, including the distance to subject and the "focus breathing" characteristics of the lens?

The "focus breathing" phenomenon has been known for a long time, even before SR came along.

I am certain the SR system is fed with the momentary FL of the lens (but I think only for infinity focus as where FL is beeing determined for) the sensor size is in my opinion irrelevant for SR (that's why you type in the same focal lenght for all-manual lenses no matter if you use an APS-C or FF camera)
While the focus-breathing characteristic could be determined for every lens and beeing taken into acount this would make little to no difference in most shooting scenarios (except maybe for extreme cases like the 60-250mm lens)

edit: as long as you stay out of macro photography that is

[pentaxus]

Since focus distance is provided by the lens to the camera you would thing that that could be programmed into IBIS. Maybe since the distance info was designed for flash enhancement it isn't accurate enough for IBIS and ignored. Hmmm. Overthinking.

[kypfer]

Thanks to the OP for bringing this subject up and everyone else for the discussion These are symptoms I'd experienced but wasn't sure why!


My Tamron 18-200m exhibits this characteristic to what I would consider an extreme. Brief "out of the window" tests suggest maximum focal length acts more like a 100mm at close focus. Long term experience with this lens had demonstrated it wasn't quite the "jack-of-all-trades" I'd hoped for. Even though most of the time it's perfectly adequate, if I'm not space or weight restricted, I'll often use my DA 18-55mm and DA 50-200mm combination instead as my "stroll in the country" rig. The one has very adequate close-focus capability and the longer shows virtually no "breathing" at all. Combined weight is only a couple of ounces more than the Tamron

[Lowell Goudge]

I am not sure it will produce better results

true focus breathing changes the focal length, in fact it reduces it, so the compensation will be set slightly higher focal length than the actual. But considering your proximity to the subject, compared to normal shooting where the subject distance Is much greater, it is still better than nothing.

Also note SR does nothing to address the most sensitive motion in macro shooting, and that is the distance to the subject.