GH5 GH5 issues on Ronin M

D

dtocco

Active member
Hey all, just got my GH5 and having some issues on my Ronin-M. I shoot a lot of fisheye stuff and since switching to the 5 from the GH4 my "running" shots have noticeable vertical shake. This has never been an issue on the GH4, and I'm pretty stumped at the moment. I'm hoping my lack of knowledge on optics is the culprit, but I've experimented with different focal length settings and keep running into the same problem.

I'm shooting with a Fotasy EOS-M4/3 converter and an 8mm fisheye. I've experimented with the focal length set to 8, 16, and everywhere around those numbers. Stabilization is obviously off for the fisheye, and I've turned my stiffness up pretty high on the Ronin.

I appreciate any insight!!
 
Is it a focal reducer? You need to put in the multiplication number of the reduced image if it is. So at .71. 8mm is 5.68

dtocco said:
Hey all, just got my GH5 and having some issues on my Ronin-M. I shoot a lot of fisheye stuff and since switching to the 5 from the GH4 my "running" shots have noticeable vertical shake. This has never been an issue on the GH4, and I'm pretty stumped at the moment. I'm hoping my lack of knowledge on optics is the culprit, but I've experimented with different focal length settings and keep running into the same problem.

I'm shooting with a Fotasy EOS-M4/3 converter and an 8mm fisheye. I've experimented with the focal length set to 8, 16, and everywhere around those numbers. Stabilization is obviously off for the fisheye, and I've turned my stiffness up pretty high on the Ronin.

I appreciate any insight!!
Click to expand...
 
soarprod said:
Is it a focal reducer? You need to put in the multiplication number of the reduced image if it is. So at .71. 8mm is 5.68
Click to expand...

Your math is wrong or my understanding of the gimbal is wrong. Assuming the gimbal is programmed for full frame, the math is:

lens focal length x crop factor x focal reducer factor = equivalent full frame focal length

8 x 2 x .71 = 11.36mm

This also assumes a 2x crop in the selected video mode which may or may not be true.
 
Greg, I think he's talking about the focal length to input for IBIS which should not include the crop factor. The equivalent focal length would be 11.36mm but that is not what Mitch from Panasonic said to use on the gh5 facebook page.

On a side note dtocco, make sure to turn off the e stabilizer in the menu. It is an electronic post effect in cam. it has proven to be very problematic and causes all kinds of issues with distortion and jumping around of the image.

I use IBIS at all times on a mechanical Steadicam. It works great. I would never suggest turning it off for use on your ronin.

Cheers,
Pete
 
Thanks for the responses guys. I guess I wasn't specific enough but yes, all IBIS and other stabilization is off. IBIS is a complete nightmare for shooting with a fisheye lens. The image vignettes and warps terribly, unlike on a normal lens.

It's my understanding that the focal length setting is only to ensure that the IBIS is being used to the best of its ability, so if it's off, it shouldn't really matter, is that correct?
 
dtocco said:
IBIS is a complete nightmare for shooting with a fisheye lens. The image vignettes and warps terribly, unlike on a normal lens.
Click to expand...

So with the converter are you obtaining a circular fisheye image or a cropped-to-the-corners fish-eye image ?

I own a Samyang 7.5mm cropped-to-the-corners fish-eye lens, but I haven't tried using it with GH5 IBIS yet.
 
TheDingo said:
So with the converter are you obtaining a circular fisheye image or a cropped-to-the-corners fish-eye image ?

I own a Samyang 7.5mm cropped-to-the-corners fish-eye lens, but I haven't tried using it with GH5 IBIS yet.
Click to expand...


Not really sure what you mean here...but I have a full framed barreled fisheye image, with no vignetting. If you use IBIS with a fisheye lens you're going to end up with a "warped stabilizer" look. With the parts of the image bulging and warping, plus you'll most likely catch glimpses of your lenses' sunshade in your image. It's bad.
 
dtocco said:
Not really sure what you mean here...but I have a full framed barreled fisheye image, with no vignetting.
Click to expand...

That's what I have with the native Samyang 7.5mm lens. I imagine that you're going to run out of image if the sensor moves beyond the image circle the lens projects. Not sure I understand why the image would bulge or warp if you don't have GH5 electronic stabilization turned on. Seeing the sun-shade makes sense as you are looking beyond the normal image area of the cropped ( full-frame ) fish-eye lens. ( I've heard of people cutting off the built-in sun-shade from their fish-eye lens to reduce this effect, but then it won't be there when IBIS is turned off )

I will try and test this with my 7.5mm fish-eye tomorrow.
 
Assuming some vertical shake/bounce from running is being transmitted through the Ronin into the camera- which is possible as gimbals can only correct for vertical tilting, not vertical travel (aka booming or jibbing)- this is probably the result of the sensor bouncing a bit inside the camera (even with IBIS turned off)- something I was disappointed to discover does indeed happen with my G85 (and I would assume the GH5 would be the same).

I was able to test this by using a fisheye with an undersized image circle (gives a semi-circular image) and shaking the camera fairly hard. The image did indeed bounce/shake at the end of my movements. It's not the lens moving on the camera, so the only explanation is that the sensor is moving. Apparently the magnetic IBIS system isn't able to completely "lock down" the sensor, as it obviously would be in a non-IBIS camera (i.e. your GH4).

Now, this would be pretty much a non-issue on any non-fisheye (or possibly an extremely wide rectilinear lens), because the amount of shaking needed to move the sensor is way past the point where you can still get usable footage. So by the time the sensor shakes, your footage is well into earthquake-o-vision anyways, so who cares.

But with a fisheye, the AOV is so incredibly wide that you can get away with shaking the camera a fair bit, and the image won't move enough in relation to the frame to be unusable. In fact it's usually still pretty damned usable. This is part of the reason why the action sports love fisheye lenses: they allow for simple one-handed camera operating, while skating/skiing/etc, with acceptably steady results, without much practice. With practice, way more than acceptable; even glidecam-like for a skilled filmer.

The other reason we like fisheyes is why everyone else hates them: the distortion. It lets us exaggerate speed, size, impact, distance, etc. Very useful for making our sports look as impressive as possible. But in this case, the distortion is also probably revealing the movement of the sensor, as the image moves through it, which would be what gives the Bad-Warp-Stabilizer-type effect.
 
aldolega said:
Assuming some vertical shake/bounce from running is being transmitted through the Ronin into the camera- which is possible as gimbals can only correct for vertical tilting, not vertical travel (aka booming or jibbing)- this is probably the result of the sensor bouncing a bit inside the camera (even with IBIS turned off)- something I was disappointed to discover does indeed happen with my G85 (and I would assume the GH5 would be the same).

I was able to test this by using a fisheye with an undersized image circle (gives a semi-circular image) and shaking the camera fairly hard. The image did indeed bounce/shake at the end of my movements. It's not the lens moving on the camera, so the only explanation is that the sensor is moving. Apparently the magnetic IBIS system isn't able to completely "lock down" the sensor, as it obviously would be in a non-IBIS camera (i.e. your GH4).

Now, this would be pretty much a non-issue on any non-fisheye (or possibly an extremely wide rectilinear lens), because the amount of shaking needed to move the sensor is way past the point where you can still get usable footage. So by the time the sensor shakes, your footage is well into earthquake-o-vision anyways, so who cares.

But with a fisheye, the AOV is so incredibly wide that you can get away with shaking the camera a fair bit, and the image won't move enough in relation to the frame to be unusable. In fact it's usually still pretty damned usable. This is part of the reason why the action sports love fisheye lenses: they allow for simple one-handed camera operating, while skating/skiing/etc, with acceptably steady results, without much practice. With practice, way more than acceptable; even glidecam-like for a skilled filmer.

The other reason we like fisheyes is why everyone else hates them: the distortion. It lets us exaggerate speed, size, impact, distance, etc. Very useful for making our sports look as impressive as possible. But in this case, the distortion is also probably revealing the movement of the sensor, as the image moves through it, which would be what gives the Bad-Warp-Stabilizer-type effect.
Click to expand...

This, more than anything I have been able to come up with, seems like a logical answer. And it is a very disappointing one. If anyone out there has any thoughts on what I can do to fix this issue, any info is appreciated.
 
The only fix would be to switch to a camera with a fixed sensor (no IBIS), or to alter your gimbal technique or setup to keep the vertical shake from reaching the camera. Meaning:

-run smoother

-use skates/skateboard/Segway/skis/hoverboard, etc instead of running

-mount the gimbal on a Steadicam vest/arm, which should isolate the camera from your vertical motion.
 
aldolega said:
The only fix would be to switch to a camera with a fixed sensor (no IBIS), or to alter your gimbal technique or setup to keep the vertical shake from reaching the camera. Meaning:

-run smoother

-use skates/skateboard/Segway/skis/hoverboard, etc instead of running

-mount the gimbal on a Steadicam vest/arm, which should isolate the camera from your vertical motion.
Click to expand...

Definitely aware of these things, but thank you for the response. It's really unfortunate to me that my brand new GH5 is underperforming my GH4 in these situations. If you need the stabilizer off, it should be off; that sensor should be locked in. I've been filming fisheye for 15 years, it's well known that when the lens is on you turn the stabilizer off. At this point I need a GH4 with 4k 60p. Very disappointed.
 
Yea, it's a bit of a bummer for me too (only a bit as I don't have a GH5 yet, only the G85). I agree that the sensor should be totally locked down when off. To be fair to Panasonic though, this is a very rare/specific usage case. I'm disappointed but not surprised that they never caught this defect.

For now I'm using the Samyang 7.5mm, which has plenty of room in its image circle for the sensor to move a bit without black corners creeping in. I haven't noticed any of the warping happening either, so it works for now- only problem being that it's not really wide/distorted enough for my tastes. At some point this summer I'm going to get one of the Meike 6.5mm f2 fisheyes off Amazon and use it for a couple days, to see if my typical usage causes the sensor to shake or not.
 
You must log in or register to reply here.
Back
Top