alexO

​To SleepyWill. Embarrassing to write? Really? No, I think it is just you been ignorant idiot, that's all. I don't earn my income through photography or videography. I just use my camera for leisure to take photos and make short clips of my family, but happened to enjoy good quality in both compared to majority of the crowd for whom cellphone quality is enough. I want to have a camera which my wife could use in full auto mode as well. Now you are proposing if she want to take video clip after couple of still shots she needs to switch to some manual Samyang lens just to get a reasonable field of view? NX500 is not suppose to target pro market, Samsung has NX1 and NX30(NX50 in the future) for that.

Not really. For some (me included, who is not proffessional filmmaker) it made it unusable. I believe Panasonic found a good compromise between GH4 and LX100(keep in mind a much better lens is supplied with LX100). If they couldn't fully readout and process NX1's sensor by NX500 processor they should have put a sensor with lower MP count.

With 1.7X 4K crop you could forget about wide shots. So I guess it's only good news for NX1 owners, since having crop 4K functionality in NX500 makes it easy for samsung to add it as an aditional option to NX1 in the next FW update. If it's done due to lower processing power of NX500 (which I suspect is the case), I doubt we would expect a decent 1080p either. Since even NX1's 1080p is kind of luckluster compared to downscaled 4K in post. Oh well, I guess I am pulling out my preorder.

That's a big turn off for me for NX500. I was expecting full readout->rescale for low light conditions to reduce the noise. Since Samsung decided to change how they produce 4K compared to NX1, they should have kept both options.

Has anyone tried log gamma yet ? How does dynamic range look with it?

I wonder if Samsung would push the similar firmware update with S-Log to NX500. So far NX500 seems to have specs of NX1 with the original firmware. Speaking of S-Log, I remember some people were not happy with sub-par dynamic range of NX1 even compared to GH4. Has it dramatically improved by filming in log in NX1?

Andrew, you wrote it certainly no match to A7s. But how does it compares to original A7? Is there any improvement in video quality beside adding S-log and X-AVCS codec?

Since LX100 uses the same sensor as GX7, I was able to convert LX100 raw after running exiftool (download from http://www.sno.phy.queensu.ca/~phil/exiftool/) on raw files tricking it thinking those raws comes from GX7: exiftool -model="DMC-GX7" <raw_file_name> Then you could use raw converter of your choice which supports GX7(I think all do).

I thought DFD doesn't work in video. Am I wrong, or it only applies to LX100?

Thanks, I’ve seen those comparisons Inazuma, and also followed threads: '?do=embed' frameborder='0' data-embedContent>> '?do=embed' frameborder='0' data-embedContent>> I believe the difference in details/sharpness between GX7 and a6000 is not very significant compared to what you get from real 4K downsampled to 1080p(I use ffmpeg with lanczos resampling). Even if you compare 1080p from LX100 with 4K which you would downsample yourself the difference is quite significant. I believe at this point only A7S does a proper downsampling to razor sharp 1080p. Others (GH4, LX100) despite doing a full readout are not producing that sharp image straight from the camera. I read that GX7 is doing pixel mixing, which is some sort of pixel binning. In addition, in your examples I felt comparison of Panasonic’s superb 12-35mm f/2.8 to either pancake style 16-50mm or 16mm is not fair even at 1080p. In fact, I found your snapshot of A6000 with Konika 40mm to be sharper than GX7’s.

There are plenty of discussions about a6000, but there are only few touching a5100. Is there a big difference between both in terms of video quality beside an XAVC-S 50mbit in a5100? I am kind of confused if a5100 has an identical video engine/readout as a6000. In the initial a5100 press release Sony had the following statement: "Additionally, with the power of the BIONZ X processor, the camera is able to read, process and output data from all of the sensor’s pixels during video recording, ensuring that it produces the highest quality video possible by eliminating aliasing, moiré and false color artifacts." Which was later revised and removed. Not sure if it was a copy/paste error from A7s or there was some improvement made in the readout(potentially pixel binning of the whole sensor without any line skipping). But since the wording was ambiguous and kind of implied the direct pixel readout they decided just to get rid of this statement altogether. I’ve seen some footages/test results/reports showing a significant improvement compared to older NEX’es and some showing very little. I suspect inconsistency could be attributed to different binning methods used for 24/25/30p and 50/60p modes. In the later mode engine might do line skipping to be able to keep up. Not sure how plausible is this theory. In addition, I found this test which I am taking with a grain of salt since I couldn’t verify who did it. It shows the noise level of a5100 to be significantly lower at ISO 6400 compared to both A5000 and 5R. If this improvement is real, it should indicate that a5100 does in fact have less line skipping. And while pixel binning wouldn’t completely get rid of moire and wouldn’t produce same image as direct pixel readout, I will take it anytime over line skipping. P.S. At this point I am basically deliberating if I should upgrade from my Nex 5R to either A5100 or Panasonic LX100, which would be better suited for non-professional work in moderate light conditions. In the case of LX100 I would probably film in 4K with intention to downsample it to get a razor sharp 1080p.