How many of GX9's "5-axis IBIS" are actually IB?

Juan Muchos Jarros wrote:

Tom Caldwell wrote:

Helen wrote:

Juan Muchos Jarros wrote:

pardon me for asking this if it is clear to all, but it is unclear to me from the descriptions i've read what portion of the IBIS is actually built in to the sensor, and what is only available when using OIS Panny lenses. it looks like (to my uneducated mind) the sensor only includes the same 2-axis stabilization (simple shift up-down-left-right?) as was part of the GX7, with the rotational & fore-aft stabilization being performed by the lens? or vice versa?

i guess my question is, if i were to buy they GX9 & hook it up to my Oly 12-40, what kind of stabilization would i see?

5-axis is what you’d get, like on any current OM-D body. From the GX80/85 onwards, Panasonic has been using a magnetic levitation IBIS system rather like the Olympus one. Neither system does fore-aft movement, as that would interfere with focusing.

I think that the OP might have been referring to whether or not the “5-axis” stabilisation included the dual or triple extra stabilisation from any lens OIS available. Somehow perceiving that the 5-axis stabilisation needed the help from lens OIS to achieve that effect.

thanks very much, that was exactly what i was wondering about. i guess what threw me was this phrase from Panny's press release: "A 5-axis Dual I.S. (Image Stabilizer) in the LUMIX GX9 effectively suppresses blur. Combining an O.I.S. (Optical Image Stabilizer, 2-axis) and a B.I.S. (Body Image Stabilizer, 5-axis)," it just seemed an odd way to phrase it, given that those 2 axes are already covered by the BIS.

Two of the axes are covered by both the lens and the body, and the three others are body-only. Of the shared two axes, combining the BIS and the LIS will give a larger range than either form alone.

These IS operations, on every axis, are limited by some range beyond which the compensation movements cannot reach. On reaching an edge of a range, that axis must be reset, so any motion beyond that won't get properly compensated and you tend to get double image motion blur. Increasing the IS movement range increases how far the user's hand shake can go before compensation fails.

In order to properly control a combined effort from both body and lens, you need higher compute capability and higher comms bandwidth between the body and lens. Hence Sync-IS is not available for every Oly body+lens combination and Dual IS is not available for every Pana combination. And Panasonic even has different versions of Dual IS which might mean 1) different compute capabilities in the cameras and 2) different IS ranges in the BIS units.

A simpler form of 3-axis body+lens combined IS also exists - I think only in some purely Olympus body+lens combinations. In this method, only the roll element in the BIS is active and the LIS takes care of pitch and yaw. There is no shifting compensation because shifting must be computed together with pitch and yaw and these body+lens combinations are not capable of that degree of compute collaboration.

I am totally surprised that this simplified 3-axis combined IS is not possible cross-brand. This shows that the excuses they make of not being able to combine across brand boundaries (they claim fundamental algorithmic differences) are lies. Roll is totally independent of any of the other axes and can be compensated independently. Hence a lack of co-stabilization is a marketing agreement. This body+lens 3-axis compensation is very useful as yaw and pitch are the most important things needing compensation followed by roll, in normal distance shooting, and pitch and yaw are usually better compensated for by LIS.