wow, so many questions! One answer: Yes I stabilize both the EM5 and the GH2 with the same settings but without skew/rolling shutter correction. It makes it much easier to spot the amount of jello because only the partial warping of the image is exposed, caused by the delay of the shutter when it rolls from top to bottom of the sensor.
I used CoreMelt Lock&Load X to pin the shots.
the GH2 has clearly more warping. The EM5 was mounted on top of the GH2.
Hi Alex,
Thanks for the explanation. It looks like the EM-5 is receiving a bit more image stabilization in post? Perhaps since the EM-5 is mounted on top of the GH2, it requires more stabilization in post since any camera shake might be amplified by the longer moment arm?
E-M5 in quadrocopter
E-M5 vs GH2 in quadrocopter:
other video from person suffer from essential tremor:
@Vitaliy_Kiselev is possible some day try to hack bitrate of Olympus cameras? (same E-M5),
or codification of firmware makes it unfeasible?
thanks
I like colors on EM5. GH2 shifts greens towards yellow. Those khaki firs in helicopter comparison are annoying.
Although E-M5 applies software-image-stabilization, it is impressively stable. It might be just more than software IS.
but, it is certain that the 5-axis IS in video does not work?
I would like to see if some of the IS smear would be gone if the shutter speed is set shorter.
Other good video E-M5 + Nokton 25mm F0.95
at least there is a clear difference between IS in stills mode and IS in video mode. You can hear when the 5-achsis IS goes to work. That does not happen in video mode. The way the IS reacts to panning is also different. In video mode there is quite of a big lag. When you turn on IS in video mode the EM5 zooms a little in on the sensor as it is typical for software IS. There was an official statement that IS is software in video mode (sorry, I don't have a link)
Whatever it is - it works fantastic. My guess is that unlike conventional software IS that detects motion based on the flat image sequence this thing might read 3D information from 5-achsis motion sensors to calculate the compensation for motion and rotation. Even though the sensor itself is not moving to compensate. At least this explains the unrivaled precision of Oly's IS in video mode.
+1 lack of 24p/25p is the only thing holding me back from buying this camera.
@Alex E-M5 might not move the 5 axis IS, but it might detect the shaking directions and feed the real-time info to in-body-software-IS which helps minimize the shake and the rolling shutter effect more effectively than post-processing corrections. You know... Nintendo Wii controller can detect x/y/z rotation movement. Just a speculation. But I sure wish Panasonic implements such feature.
... and that could be the reason why Olympus officially says 5-achsis IS is used for both, video and stills ...
Some user testing showed E-M5 has much better DR than Fuji X100. It will be great for timelapse.
Test video E-M5 vs Fuji X-Pro 1
More test DR and Noise: E-M5 have better values of Nex7:
http://www.dpreview.com/reviews/olympusem5/16
@Alex What about aliasing/moire on E-M5?
I didn't test for aliasing/moire in particular. I will brows my footage and see if I can spot some (returned my E-M5 today) ...
A friend of mine just got his new E-M5, I bet we'll have some fun time doing "shoot-outs" ;-)
(At last both of us may have bought the right camera, as he's more into stills, and I'm more into video.)
this is my favorite pro argument from the dpreview review:
"Art Filters can help produce interesting results from dull subjects"
man, wtf
Alex, this is one of the most customizable cameras i know. The OM-D is one of the most difficult cameras to understand at the start. But in 2-3 days you can do anything with everything the camera offers. One sugestion, the fn2 button should respond to multifunction option, when you press the button it does exactly what the GH2 does with the push button. My tests with the OMD differ from yours, but i can't upload anything yet. Maybe when i have time, i'll do a quick review from a photographers point of view and another in video's. Be sure to watch and read steve huff review he's quite honest in all articles. Regards. Still here's a good examplo of IBIS on the OMD
P.S Read the OM-D advance manual, most of your doubts are only that, doubts.
João
It looks like you're new here. If you want to get involved, click one of these buttons!