comparing the moire from 18sec to 20sec (Red scarlet 4k)

to the GH4 4k 43sec (stairs)

which one would look better?

@Aria @Vitaliy_Kiselev Defenetly agree. the thing is that I believe that the work Vitaliy have been doing with the GH1, GH2 and the rest of the Hackable Lumix cameras must have had some impact on the rout Pany has taken. Showing the way to the more advanced users and to the pros with his hacks and the big interest they have got from many pros and others (me);) Thanks Vitaly!

Just as a point of interest for those who are comparing bit rates, We deliver finished HD masters to the broadcast networks in mxf files at a MAXIMUM bit rate of 50mb/s 50i. The GH4 bit rates seem to me to be a good compromise between quality and efficiency.

I do post production work on a lot of Sony XDCAM HD originated footage which is long GOP 50mb/s and it looks ok. The GH4 will do 200 mb/s in HD. I think we may see some incredible work once this camera is in the hands of talented creative people. I have seen amazing quality shot on hacked GH2 and GH3 cameras. I have also been grading Canon log footage from C300 and C100 cameras and getting very bad results! Why? Because the camera operator was sloppy and did not set up the camera to deliver well exposed and proper white balanced footage. The quality is in our skill, not in the equipment.

@amateur Comparing aliasing and moire paters in those two videos is pointless because severity of the issue really depends strictly on the pattern/object and specific camera combination. The angle of the lines, how they are spaced, how far they are from the cameras, etc. combined with the camera's resolution make all the difference. After all it is an interference so as such the camera's resolution and the type of the pattern have to be considered as frequencies. Even putting the cameras side by side and filming the same target does not give an absolute answer about how they handle moire and aliasing, only about how they handle it in the given shot.

I loaded that Yucatan video from YT and watched it with potplayer in 200% enlargement (100% pixels in screen). In my 50 inch plasmaTV it looked OK and I didnt see moire and aliasing almost at all. Some were in parrots feathers. Resolution in static shots was quite good but not perfect. Compression was very heavy in moving parts due to YT. Bitrate was generally 8-15 Mbs peaking 20.

@caveport

"The quality is in our skill, not in the equipment."

+1 !

Are there any news on how the gh4 handles low light, higher iso settings. I know it comes with a new sensor, but still at the same size of the gh3.

But seriously.. stop these discussions / comparisons of bitrate because they seem absolutely pointless! Bitrate is one thing but the encoder parameters play the most important part. Just ask Driftwood or any other person in this forum which is skilled in the craft of developing encoder settings. And if you have a decent implementation of parameters it´s up to the shooter to make best use of that given situation.

Without a thorough side by side comparison you are just throwing numbers back and forth. Even then you have problems in that some encoder parameters might handle certain types of scenarios better than others.

And then we have not even touched upon other differences between cameras, which influences what kind of material you will get from it.

I am personally looking forward to trying out the gh4, and with an Odyssey 7Q in the picture before long (mainly for the fs700), I don´t worry about 4k acquisition. IF I was to make a setup for fast recording on the gh4, I´d consider the cheapest (reliable) recorders to go with it to get prores 422 out of the box as that is great for both clients and fast editing. 4k is largely superflous for most applications so far. But first of all, I want to try out the cam and see how I like the image and the overall handling.

Its funny how Canon has a codec that can squeeze 1080p 50mbits 4:2:2 on the C300 but the GH4 needs up to 200mbits on 4:2:0

Maybe it will have more details on the GH4 but the color res can be more important. a Canon MXF MPEG2 codec at 200mbits would look beautiful and plays back very easily on a lower end system. Editors love the codec and it is approved by the BBC at 50mbits. Would love to have a similar codec on the GH line.

The scaling and output could be so good that it will be very good quality color even with 4:2:0 from the initial sampling.

@jakepowell that's what I was reading only, but it seems that it does true 10bit 422, and it also does record 4K footage.

Nicely put @bwhitz

@jamesgh2 The Canon C300 records 8 bit 4:2:2. When it comes to grading, the banding is a bit depth issue not a colour sampling issue. I'd be interested to see what the 4k 4:2:0 recordings look like when converted to HD 10 bit files before grading. Everything is only speculation at this point & it won't be until we see the actual camera in use that we will REALLY be able to criticise it.

Since Panny didn't specifically state that the GH4 complies with the new Rec. 2020 color gamut for UHD, I'm going to assume that even the 4k footage is limited to the very old yet still current HD Rec. 709 standard.

I'd love to see the first company that actually delivers an HDMI field recorder that can capture 4k at at least 10 bit 4:2:2 using a superior codec that isn't budget-busting expensive. Indie filmmakers will need it. It's not that 100 Mbps can't look decent for 4k, it's the codec that's doing the compression that is the issue. The GH4 uses MPEG-4 and 8 bit 4:2:0 internally. That doesn't exactly scream "professional-grade" image capturing to me. We all know the problems with 8 bit consumer video.

The cost to do 4k, let alone 10 bit 1080p, is going to be relatively expensive for the foreseeable future, unfortunately.