@adventsam the points you mentioned are also my concern. But as I said I used standard setting -4 -3 -3 -3 so that may be one of the reason for soft and mushy... Also I learned by these tests to NOT use 72mbps at low iso (i have to compare with more shots at high iso).
Most people said were happy the green tone went away with GH3, but as you said i think now we have a more magenta-like dominance. The tighter crop may be a problem or an avantage depending on the situation but I would say that x1.8 was already enough...
I really tried to do some run&gun shooting without having mastered the camera so I wasnt looking for the "pretty picture". It seems that some people could achieve that already in some videos posted in this topic. This week I'll try using the camera to get the most beautiful image possible (by my standard of course) and see how it goes.
Sorry for a longish first post, this is a great and informative forum!
@FilmingArt, the trouble with the stock picture profiles on many if not most of these cameras is that they are overly saturated with too much contrast, sharpening and noise reduction "baked in". Therefore the dynamic range is somewhat compressed and there's a loss of information especially in the shadows. The goal of the Canon picture profiles, for example, that are used on the 7D and 5D, like Cinestyle or Marvel, is to get the most neutral image possible and retain as much tonal and detail information as possible. The value of "flat" or "log" is that it gives much greater latitude for grading in post. Even with a 4.2.0 codec. I'm hoping that someone here will be able to develop a series of "cinestyle" profiles for this camera that will give a truly neutral image (without the magenta push).
I have now had the GH3 for 3 days and have begun to play around with it. My initial impressions are positive BUT I am finding that there is a lot of strobe effect and aliasing on slow pans. I've tried to get slow, smooth "strobe free" pans in MOV-50mps & 72mps-24p @ both 30 and 50 shutter speeds and MOV-50mps & 72mps-30p @ 30 and 60 shutter speeds, but I am finding there is a lot of staccato motion, kind of a repetitive delay in the panning motion. I am using the 12-35 lens. I've been seeing this on a lot of the Vimeo and Youtube clips as well but I thought it was just compression problems. I'm seeing it in the raw footage. Seems much more pronounced than on the Canons. Has anyone else noticed this?
The other issue I'm finding is that there is a brightness shift when zooming with the 12-35mm. Its almost as if the aperture pumps slightly mid-zoom even if its a fixed aperture lens.
I really want to hang on to this camera but these issues have me doubting that I will. Could others please let me know if you have these issues? It may just be a problem with this particular specimen.
thanks
@ conundrum
ive encountered problems in the past panning with the newer panny lenses optimized for micro 4/3. try using old nikkor, olympus or canon glass with an adapter. the results often look smoother as these lenses resolve less detail (lp/mm).
http://www.luminous-landscape.com/tutorials/resolution.shtml
@Tenjin, good luck, I am interested in results, but currently I am disappointed with GH3 iq? if Panasonic have moved to Sony for the sake of 1ev dr, its a big mistake.
Just read that article, thanks but I'm not sure what it has to do with my issues? This may be my first MFT camera but I've used a Canon 7Ds & 5Ds on productions with the 17-55mm EF-S and some very sharp primes and the pans at the same frame rates and shutter speeds had a smooth film-like quality. This staccato, stuttering motion on my GH3 is weird and totally disappointing. I do not profess to be an expert, I've simply followed this rule and normally had great success: http://tylerginter.com/post/11480534977/180-degree-shutter-learn-it-live-it-love-it.
Just tried an old Canon lens and it is the same if not worse. If anyone can give me some other way of testing to see if its just a problem with my particular camera, I'd be very grateful. The way it is now, this is unacceptable.
slowmo test with FCP X 10.03
So what is a footage converted to 300 mb/s worth? Transcoded from 72 to 300. Any difference other than the latter is heavier?
@windcharger: I see a lot of micro stuttering of moving objects - e.g. the adult slide at 0:18 and the child sliding to the left at 0:24. Is this from the camera or from the pcpx?
Great forum!!
Are anyone using the i.dynamic and i.resolution options in any way? Will they give you a flater profile or whats the deal with them?
I´m going to film some documentary work on Wednesday and I´m still trying to figure out the settings for the camera. The project I´m filming will continue for at least one or two years and It would be nice to start filming in the most correct way possible =)
Are everybody saying that the 50mbit mov codec is the best one? Better then the 72 all i? How do you grade your material? Sharpness, noise reduction, contrast etc - What do you ad and how?
If anyone takes that Will Crockett guy seriously, I need to punch myself in the dick. If you're going to promote yourself as some sort of camera expert and blog guy, AT LEAST USE A F*$KING BOOM MIC OR SOMETHING!!!!!! Good god. I heard 5 seconds and off it went. And as for the frame rate stuff he said, yeah, that should tell ya enough!
@filmiwilliam i.resolution mode whilst recording video automatically switches to LOW setting. Still doing the tests to see if there is any gain.
“Cinematography is more than a camera, whether that camera is a Red an Alexa or a Bolex. There is a little more to it that resolution, colour depth, latitude, grain structure, lens aberration etc. etc. etc. The lenses use for ‘Citizen Kane’ were in no way as good as a Primo or a Master Prime and the grain structure in that film is, frankly, all over the place. But the cinematography? Well, you tell me.”
Make more of such videos :-)
:) It was indeed delicious.
Reminds me of my ex. She is dead to me, but the pilaf lives on...
Seen it in every GH3 sample video so far : stuttering video in pans and object movements... Is this due to Youtube/Vimeo compression or can you guys see this in the source material too ?
Thanks !
I would expect that the 50Mb long GOP video should look very good on the GH3 - at least as good as GOP12 at 66Mb on the GH2 - probably even better.
The 72Mb all I frame video probably won't look very good. At 72Mb GOP1 streams will have small I frames - about 375 kilobytes each. Even at 24Mb a GOP12 stream can have I frames up to around 1 megabyte. I frame size pretty much limits image quality, and 375 kilobytes just isn't enough. In order to maintain reasonable individual frame quality, Nick Driftwood had to push data rates to over 150Mb for some of his all I frame (GOP1) settings.
Video captured at 24 FPS will always exhibit an uneven motion cadence when displayed on a monitor refreshing at 60 Hz because 60 isn't divisible by 24. If you can increase the refresh rate to 72 Hz on your monitor things should look better for 24 FPS footage (assuming you haven't rendered it to 60 FPS). However, then 30 and 60 FPS footage will end up looking odd because 72 isn't divisible by 30 or 60. Ideally you would want to change the monitor refresh rate according to framerate.
At 25P you would need to have a monitor refresh rate of 50, 75, or 100Hz to avoid the cadence problem. Does the footage show the same stuttering, etc... when displayed on-camera? Maybe there's something about your computer setup than doesn't like the GH3 video files.
@ cbrandin
Yes same problem when displayed on camera. It is visible as soon as I do "rec". I never had this problem with GH2 in 24P or 25P. Ordinateur : macbook Pro retina.
I think the problem is coming from the 25P.
@jjay, @cbrandin I am bothered by the jittery moves too. To figure out if this a GH3 flaw or my monitor/vimeo combination, I looked at other shots made with different, more expensive hardware. No such problems there. See for example http://benjamineckstein.com/2012/06/silencio/ - shot with SONY F3.
What is the best format in 24p for wedding videos - .*mov or AVCHD? AVCHD is easier to work with on FCP though.
Is the 25p actually 25p, or is it 25p wrapped in 50i? Wrapped 25p isn't encoded in the same way as true 25p - rather it is encoded as half intra and half inter interlaced, even though both halves are derived from a single image. Unwrapping it will not produce exactly the same result as a true 25p encoding.
It looks like you're new here. If you want to get involved, click one of these buttons!