@bdegazio: "2) irregular decoding in playback (i.e sometime a frame is decoded a field or two late) 3) interference in playback with 60 fps monitors"
I agree about these kinds of playback issues; I find my footage looks different on my various different computer systems, but once I finally edit it to an HD clip for playback on a widescreen TV via a TViX, it becomes consistent and OK. One test I did for myself - and the one that convinced me that my raw GH2 footage was OK - was to transcode the .mts file to a format with intraframe compression (e.g. Cineform) and add manual 3:2 pulldown as I did so. My computer systems all play 24P packaged within a pure 60P intraframe footage very consistently.
i don't see anything so offensive about it that dalefpf is saying, definitely not what his friend sees "as if the image paused every 1/2 second or so". having looked at the frame grabs of every single frame, it literally looks like a perfectly linear translation from right to left with constant movement from frame to frame across the entire sequence. there is no frame that is doubled up or shows more or less movement or more or less motion blur than any other. it literally looks like the scene was copied and pasted x number of pixels to the left from frame to frame. this is exactly what i would expect from a motion picture recording (or even an animation) of an object moving linearly from right to left.
@bdegazio you can download the raw MTS files from vimeo
i'm sorry if i sound argumentative, but i sincerely don't get it. if the GH2 is recording motion incorrectly in 24H mode, surely it would be evident in frame by frame analysis where the motion is non-linear from one frame to the next? a "pause every 1/2 second" should show up as less (or no) movement every 12 frames, but i can't see anything at all like that.
edit: i removed a paragraph with my math which was totally wrong :)
@Mark_the_Harp As a result of the various tests and discussion in this thread, I've come to the conclusion that the strobing is a combination of:
1) common 24 fps strobing 2) irregular decoding in playback (i.e sometime a frame is decoded a field or two late) 3) interference in playback with 60 fps monitors 4) possibly some exaggeration due to highly detailed GH2 image or absence of motion-blur
Could be. I couldn't see it (particularly) on stepping through either. Could it be that something in the GH2 is making these effects more noticeable on certain monitors (shutter speed not being what it says, for example)?
Downloaded the mp4 file, single-stepped though it and it looks OK. Just the normal superimposition you would expect from a certain amount of motion blur. The effect must be in your monitor, or maybe it's a perceptual thing like those "figure/ground" reversal tricks.
@bdegazio Just looked on another machine and it looks fine - the effect isn't there at all. So maybe if anyone else is seeing judder, have a look at my clip above and if you see the 'black hole effect' it might be a monitor and / or video card thing.
@bdegazio Have a look at the funny three-sided hole above the strings (it's the same shape as the one bottom of frame), as the hand crosses it from time to time during the clip. Do you see it apparently sitting "on top of" the hand as it sweeps past? If not maybe it's my monitor! That might explain why some of us are seeing these things and some aren't.
Watching the vids above, I have downloaded the middle (fastest / worst) one and stepped through it in VLC and can't see frame-by-frame what's wrong. But the fast motion does look a bit unpleasant and it reminded me of something I've seen recently, connected with motion and dark / light transitions.
On a shoot I did that there is a REALLY strange effect where there's a "persistence" to black objects which looks odd - this video shows you an upload of part of the offending shot (otherwise untouched and only shortened and transcoded to mp4 for Vimeo as the original is 5 mins long). Recorded originally in 1080 50i but I'm wondering if there's some weird GH2 processing thing going on and that it's related to strobing in some way? Never seen this in my other cameras:
@Elenion Yes, I can see strobing in all three versions, but by far the worst in version #2. Although some strobing is present Version #3, it's is quite acceptable, depending on your presenetation medium. Internet would be OK, large screen projection I would have to check.
@bezerk55 If I can get your original files I can check them on a 40' screen with a Christie 2K projector later this week. I'd like to see for myself.
BTW, I like the police siren effect as the black box "speeds by" in #3 ;-)
i've uploaded three test videos to my vimeo account (please excuse the salsa videos...). hopefully the content is what is required to most clearly show any problems.
these are all raw MTS files which you can download from vimeo.
filmed at 24H mode standard film mode 0 0 0 0 i.resolution off i.dynamic off shutter 1/50 iso 320 WB incandescent lens is voigtlander 25mm f.95 at F2.8. focus is on the tape holding up the middle black object. video head is a manfrotto 701hdv I live in PAL land so lighting is 50hz. firmware is hacked to 32000000 24H mode (don't believe this will make any difference as bitrate doesn't get anywhere near 20Mbps in any of the videos).
first video is a pan with the left edge of the middle black object on the right edge of frame, taking 7 seconds to get to the left edge of frame.
second video is about twice as fast
third video is much slower
i have used ffdshow tryouts rev3914 (with no post-processing) to frame grab 50 frames (so about 2 seconds) of the first video to PNG format which you can download from here http://www.mediafire.com/?mseg7gma6dd0avv if you want to look at every single frame pixel by pixel as raw as it can get
Hi there, one of the samples posted from above is from me (Gunther). Regarding strobing, while I haven't noticed anything wrong with the 1080p24 mode from GH2 (also compared to GH13) as can be seen in my comparison video above, what I have noticed is something strange: I always use the main trigger (release button) to start filming (the one used for photos normally). When I use the dedicated video button, the GH2 seems to switch and use the automatic 1080i50 mode (I have a PAL model). Could this be one of the main reasons for the observed problems, i.e. people not using the main trigger in creative movie mode and then getting shutter speeds, etc. messed up?
For what it's worth, we haven't had this problem on our GH2 any more then we do on our non MX'd RED. Progressive cameras create this effect, and the sharper the image, the worse it gets, or so had been our experience.
Have people done an AVCHD vs MJPEG at 1080 comparison with this?
OK, here are the tech notes, since those dont show up if you watch the embedded version of the Vimeo clip: A Panasonic GH2 and a HMC40 were mounted together on a pan-head tripod. Both were set to shoot 1080p23.976 at 1/50 shutter. The resulting MTS files were converted to Cineform avi 1080P23.976. These clips were overlayed in a 23.976 timeline in Premiere CS5.5 and exported to .MP4 for upload to Vimeo using H.264 720p23.976. Watch with HD on.
@Elenion:"I think it is quite incredible that no one here can see the effect ..... just take a medium-wide lens, go outside in the garden..."
OK, I did just that. I strapped an HMC40 onto the tripod along with my GH2, and went outside in the garden. What I see from both cameras is simply what I would expect to see from 24P from any camera. But judge for yourself:
Question for those having the strobe problem: What filters are you using? I was shooting a bunch of panning b-roll on my 701 head today and had zero strobe, even in similar situations to above, even in hdmi playback. I almost always have a nd filter on, and often a black frost. Could that be softening or diffusing the strobe perhaps? Have people still experienced strobe with a strong nd filter?
While not disagreeing with lpowell about the pixel-level stuff he's commenting on, let's not lose sight of the bigger issue being highlighted by those two side-by-side screengrabs (which I posted way back when on DVXUser). The original GH2_Strobe_issue.MTS was supposed to have been shot at ~1/48 (i.e. "180 degree shutter" for a 24P clip). But if that were so it should have had motion blur similar to that seen in the right-hand clip, which was a screengrab from a very similar pan I did on my GH2 camera . So clearly GH2_Strobe_issue.MTS was shot at a much higer shutter speed, for whatever reason, and so yes it strobed badly, as would be expected.
I suggest to use high bitrate MJPEG for testing of any such issues. As it is simple intra encoder, so all motion artefacts can be caused by sensor only.
lpowell explained this issue some time ago where high contrast vertical subjects were strobing in video because the GH2/GH1 is optimized as a stills camera not a video camera.
posted by lpowell
Judging from the high-contrast vertical edges in the GH2_Strobe_issue.MTS frame shot, I think it's likely this strobing phenomenom is the same as the AF100 issue discussed in recent thread
I analyzed a sample video with a similar type of pan-induced flickering, contributed by one of the posters in that thread:
Quote Originally Posted by nugat View Post What we need to see is is really slow movement of an object across the frame for more than 7 seconds...
I downloaded the MP4 file from Vimeo and imported it into a Premiere Pro CS5 25p sequence. At 100% magnification, I saw a prominent flicker on the edges of the potted plant as it scrolls across the frame, both at 25p and single-stepping through the frames. I've seen this distracting effect numerous times on high-contrast scrolling shots I've taken in daylight, particularly when shooting at high shutter speeds.
As an experiment, I displayed the video at 200% magnification in Premiere Pro (with Full Playback Resolution). This dramatically reduced the flickering effect, though I could still detect small traces of flickering on the edges of the plant stalks. It's obvious to my eyes that Premiere uses a smoothing filter at 200% magnification that reduces the contrast on sharp edges.
I believe this flicker is a side-effect of the sharpening filter used in the demosaicing algorithm the camera uses to convert the RAW Bayer Filter output into RGB pixels. When an object with a high-contrast edge moves across the frame, its alignment with the image sensor RGBG grid varies slightly with each frame. As a result, the calculated RGB values of the pixels on the edges can vary noticeably from frame to frame. In the early days of low-resolution 3D computer animation, we used to call these edge effects "farkle" to distinguish them from the flicker due to low frame rates.
If you single-step through the video, you'll see that this edge flickering is only apparent in the instant that one frame is replaced by the next - it does not persist for the entire duration of each frame. In addition to the sharpening filter in the camera, the response time of the video monitor will have an effect on the flicker. Unfortunately, hi-res displays with faster response times (such as LCD's designed for gaming) will make the flicker appear more prominent. Slower displays will tend to smooth the high-contrast edges and minimize the flickering.
There are two approaches that may help suppress this flickering effect. When shooting high-contrast moving subjects, using slow shutter speeds and minimum in-camera sharpening should help smooth out the sharp edges. In addition, post-production smoothing filters can be used to selectively blend problematic edges into the background. I'm sure the gross improvement I saw using Premiere's 200% magnification could be done with far more skill and subtlety.
I think it is quite incredible that no one here can see the effect. Obviously in the clips posted by Ian_T this effect is not evident because movements are minimal, most with tele, blurred with low dof, vintage lenses with few definition. And they are in most cases clips that want to be beautiful, then bad shoots are out. If you want to check the effect just take a medium-wide lens, go outside in the garden, close the diaphram and move the camera quite fast around the garden. You can see 24p strobe effect that in gh2 is way more evident than elsewhere. If you can't make this test i'll make a clip tomorrow or the day after. But, again, i think that the effect is obvious and i can't believe that you never noticed it.
@DouglasHorn Hmmm.... I'm afraid I'm not going to be a terrible help on that -- unfortunately don't know much about final cut workflow -- I'm just importing directly into the bin at dnxhd 175 x. I know before avid got the avchd import bugs fixed, a lot of people were converting with external converters before bringing it in not sure if that would help.
Good to know that you are having fcp7 problems though -- I've just moved over to a final cut shop and was actually going to get them to pull in some test files because we're going to be trying to match (as close as possible) a gh2 with f800 so I'll be fiddling with the cameras settings. But I'll look out for strobe now too. How are you monitoring the fcp7 footage?