Attila Bakos

I'm not sure it's a good idea to judge motion cadence by looking at videos that were most likely shot with very high shutter speeds.

Yes I tried it on this footage but avisynth fixed aliasing much better. For noise it's king though.

Yes I tried it on this footage but avisynth fixed aliasing much better. For noise it's king though.

My cat disagrees.

I recently finished this project. If you like the results, download the LUTs for free on colorizer.net.
 

4K 16bit uncompressed RAW at 25 fps needs 422 MB/s. That's megabytes, not megabits. Based on the information we've been given so far the body can handle that, but you can't write that to the SD card, the fastest UHS-II cards can write up to 250-260MB/s continuosly. So you either compress RAW, or lower the number of bits.

Long but informative review:
 

Great lateral thinking by the guy who came up with the workaround but I'm now wondering if the best fix for these issues is to set my own calendar into the future and buy the mark IV version of it.

Long but informative review:
 

Long but informative review:
 

Yeah, the A7III is a steal :D

Yeah, the A7III is a steal :D

Yep. Works particularly well with teh SLR Magic Rangefinder, or so I hear.

Yeah, the A7III is a steal :D

So in order to have working AF, don't use DR400 (or 200), don't use F-Log, and don't use a fixed ISO. And even then the transition is not smooth at all.
 

Oh that's not my video, I don't have an X-H1 (yet?).

Yes, even at DR100, it looks herky-jerky. Is that somehow usable? There's a difference between saying that it's less bad at DR100, and saying that CAF for video is actually usable.
Also, I find it disturbing that folks seem to be almost preemptively anticipating problems as inevitable; that they'll later be fixed in firmware as part of Fuji's world-class Kaizen.
My understanding is that Kaizen (the Deming cycle, Toyota Production System, call-it-what-you-will) was meant to discover and prevent flaws like this from being pushed into market in the first place; not to fix errors that were sloppily discovered only after the release.
A more accurate term for what Blackmagic, Sony, and now Fuji and others are doing is the well-known strategy of "Release it Now; Fix it Later."
For $1800, I expected more. (And a headphone jack.)

@Attila Bakos I will do. Don‘t know if I find enough spare time today or tomorrow, since I first have to get my X-T2, but weekend at the lastest I will do

Can anyone with the X-H1 confirm this?
 

I do. just tried it. Inconsistent results with DR400, and very usable with DR100. 

Can anyone with the X-H1 confirm this?
 

The same music video shot on the GH5 and the X-H1. The GH5 had some manual lens and it's either soft or the focusing is not perfect, so try to focus on color only.
Which one do you prefer? I like the colors of the X-H1 much better.

The cleanest 8 bit Log footage I've seen so far comes from Canon (C200 e.g), comes close to 10 Bit V-Log on the GH5 which is amazing.

Yeah I used my X-T2 to create the LUTs, I just tried them on the X-H1 too and they seemed to work. I sold my X-T2 but I'm not sure if it was a good idea. I wanted to get the X-H1, but it has serious AF issues, lots of moiré/aliasing in not just 120p, but 60p as well, and I've seen jerky movement with IBIS in more videos. In many ways the A7III looks better, but I can't leave Fuji just yet, I love the colors and their lenses. With my method I could create Fuji LUTs for the Sony though, but I will wait and see if Fuji can fix the mentioned issues.
@deezid is unfortunately right, when you shoot F-log, there will be blocky noise which is not, or not entirely related to compression, and I see it especially in the red channel. At first I thought they fixed this in the X-H1, but now I see it in Ryan Carlson's footage. I had this even with externally recorded ProRes on the X-T2.
This is the red channel after nothing but a color space conversion and an s-curve:

Once you start pushing the footage, you will need heavy noise reduction. I'm not sure if this is normal.
This is the red channel of ProRes from the X-T2 + Video Assist 4K, again, only color space transformation and s-curve:

 
You can't see if from these still images but this noise is all over the face, especially problematic in the darker parts.

Nice video Attila! God I can't believe how good the Fuji stuff looks. It's not ALWAYS right, but when it is - god damn!!
Really making me re-consider the X-H1 over the A7 III. 
Edit: Oh wait, this footage is actually from the XT2. Damn! Hah. Who needs FLOG when the standard profiles already have as much dynamic range as necessary.