Personal View site logo
Make sure to join PV on Telegram or Facebook! Perfect to keep up with community on your smartphone.
Official Panasonic GH3 topic, series 2
  • 1024 Replies sorted by
  • Looks like the issues we have been seeing are related to the 12-35mm lens. Something that a firmware update could correct.

    It is funny that a general electronics site comes up with one of the better GH3 videos.

  • I have to agree with you Hoodlum on the video quality. Interestingly, their review hits most of the relevant features and concerns of the GH3. This isn't always the case with a lot of review sites who miss the really important details (at least the ones important to many of us). They seem competent enough to be "trusted" as advertised.

    http://www.trustedreviews.com/panasonic-lumix-gh3_Digital-Camera_review

  • @Manu4Vendetta

    do you have a vimeo account to please ask him if it is ALL-I and the iso setting, and let us download it...

  • hey guys, does the trusted review video demonstrate aliasing has been effectively addressed? Watching the video over youtube, I'm seeing all sorts of stair-casing on the diagonal or curved lines of animals' hairs, feathers, insects legs, etc. Is it my bandwidth issues or youtube compression? Do other people have better results? Totally want to see aliasing resolved before I order the camera (else GH2!). Thx!

  • I see no aliasing on the 1080p version of the youtube video.

  • I can't wait to have a GH3 in hand to really experiment with shooting and to grade. It looks good overall but still hard to say just how good it can look under the best circumstances. I'm looking forward to more videos and hopefully one done by someone with a real creative flair and understanding of how to use the camera. Of course more importantly a production ready camera. So far it looks good in some videos and less so in others. I think the assumptions are right about there being some kind of issue with the 12-35mm lens and the GH3 video.

  • I think the assumptions are right about there being some kind of issue with the 12-35mm lens and the GH3 video.

    May be it has quite strong corrections, as hey had sharpness as priority.

  • All-I 1080p looks blurry (fake), but 720p is perfect, no way to make 1080p look clean with some higher bits (HACK) or is the small sensor ?

  • GH3 testclips are getting bit better, I see blown highlights everywhere. :/

  • @blackspot, from everything i've seen the GH3 has better DR than the GH2 and I think that the future tests will show that it has better range, but perhaps skewed more towards the low end rather than better highlight handling. I think eventually users will go with lower ISO's and use the GH3's improved low light capability to use the full DR of the camera retain more of the highlights and they can just lift the shadows in post.

  • Most test samples probably have pretty high ISO. With faster lenses and lower ISO, dynamic range will probably improve. It can't be worse than the Olympus E-M5.

  • rx1 vs gh3

    Links to video removed.

  • @RRRR

    "1) the shots are indeed straight from the camera"

    So, they uploaded the 50 Mbps (or 72 Mbps "all-i") directly to Vimeo and Vimeo delivered the 50 Mbps files to my browser? And, the edits to the various tests were not done in post?

    Respectfully, I maintain what I wrote. If these guys had mediocre export settings -- and I expect they did (photography reviewers and not pro photographers nor pro video guys) -- then that could explain both less than impressive images and aliasing.

    "2) what do you mean by: "factor in the additional light loss for MFT"?"

    There are more issues to the crop sensor than just the crop factor. f/2.8 on a full frame means something different than f/2.8 on an MFT sensor. They're not the same.

    "3) 50mbps in mov container seems to yield the best video results so far from the gh3."

    I expect this is correct. 72 Mbps is low for intraframe, and will very likely exhibit more compression artifacts than 50 Mbps long-gop.

    What I don't know enough about at this point is the 32bit issue people are talking about with the "mov" container.

  • @amateur

    This guy has posted the same video for every single new camera that was released in the past few months. I seriously doubt every company sent him their new cameras all at once. He also has disabled comments on all of his videos. This is so that people can't post that these videos are faked.

    I doubt he has any of the cameras that he has said he has. Wait for the real reviews once the final firmware is complete.

  • @mpgxsvcd

    Regarding that guy with the lighter video -- I suspect the same thing.

  • @manu4vendtta that video looks really good to me. It has what I consider the "Film Look". With the GH2, I am kind of disappointed the majority of the time because I feel it looks too video like in its motion.

  • With the GH2, I am kind of disappointed the majority of the time because I feel it looks too video like in its motion.

    Look into the Intra patches.

  • Ok, is there a chance, that the GH3 will arrive in Europe before CHristmas? I mean, they change Hardware-Buttons 6-8 weeks before shipping, the Firmware is not final yet. And what do you think of image-QUality? From what i saw, it looks really good at ISO 3'200, i look forward to test the Cam in low light! I will test it on the first day in ISO 25'600. And then i will not buy a new Cam for at least 3-5 Years, because we should make Pictures, and don't read Rumors in Internet :)

  • I moved all flame into http://www.personal-view.com/talks/discussion/4879/f-stop-flame

    Any attempt to continue it in this topic will lead to instant ban for one month.

  • I find this whole question of "video" motion curious every time it is brought up.

    I was going through clips recorded over my last 4 months of GH2 experimentation and evaluation and my own observation is that MJPEG 30.00p clips are easily the best looking as far as motion is concerned. When viewed back on an nice IPS monitor they look proper and it is not surprising because computer monitors are 60.00p which means no drop frames on playback.

    The other framerates of 23.976, 29.97, 29.97i, 25p, 25i that you get with AVCHD will have the playback engine of mediaplayer drop frames to sync to the display refresh of 60.00p. Is this drop frame is noticeable to you?

    I don't notice drop frames but consistantly I am going through my clips in a playlist and when I notice motion standing out as being nice I check the recorded format and it is 30.00 MJPEG.

    Obviously my preference is to have smooth video and this is my preference and not something other people may like.

    @Aria Good comment. I am most interested in seeing footage of the GH3 paired with the fastest lenses to see if it can produce a more distinctive look. If the DR performance is biased to low luminance it is a good thing because badly encoded shadows can be problem with GH2 (if you want to grade-not as bad if you take the footage as it is).

    We don't know yet what will be worth hacking in a GH3. I think maybe the Wifi and remote control features could offer the most interesting possibilities. Electronic follow focus like this http://www.okii.net/product_p/fc1.htm would make automatic Lumix lenses a great choice for rigs.

  • Zacuto link removed, as all the video is just advertising of their products :-)

  • @vics you are right ! 23.976, 29.97, 29.97i, 25p, 25i <=> a mess, midle age.

  • @onionbrain (cut-pasted from another post into the right topic)

    I take it you didn´t download the source files? (They were straight from the cam). Please do! IMO the .mov looked better than any of the avchd´s. 50mb seemed quite a bit better than any of the other (the grille which was completely moire ridden in the other was a lot closer to what it was supposed to be - not without moire, but much, much better). In any case, it looks like an approximate 15% loss of resolution at a glance. (this is my own estimate, feel free to think otherwise).

  • image

    Seems like thermo modelling (from Panasonic shop page).

  • grate image, seems to be a good disipation design, from the front side

This topic is closed.
← All Discussions