Sony a6400

Quite a conversation you've had with yourself the past two years!
 
"Slog in 8-bit APS-C"

Looking at the latest APS-C cameras by Sony, the a6400 is closely related to the ZV-E10, but not the latest FX30. Put simply, everything I have posted so far about the a6400 applies directly to the ZV-E10 because what I have written is related to the sensor and the 8-bit image processing chain which are still used by the ZV-E10. The FX30 has a new slightly higher resolution sensor, which Sony has never used before in a "consumer" camera, and that sensor is used with a 10-bit capable processing chain.

While the FX30 can replicate what can be done with a camera like the a6400 or the ZV-E10 (though not perfectly), it is also capable of doing more. In terms of replicating the a6400 / ZV-E10, the problem with the FX30 comes from the slightly higher resolution of the new sensor which means that the crop factors will not quite match up in some situations, and that where crop factors do match up, detail levels might be measurably different from the older tech to the newer tech, generally favoring the FX30.

However, Sony's designs and implimentations are deliberately similar, so that specifically, knowledge and experience with the Picture Profiles carry forward from 8-bit to 10-bit.


The "S-Logs"

The big leap forward is that "S-Log3" is practically usable for general use video on the FX30 whereas it was generally a waste of time on the 8-bit APS-C cameras. "S-Log2" on the other hand, was actually more than just useable on the 8-bit cameras, there were situations where the added dynamic range might even be the only way to get something done. So the "S-Logs" are the big dividing line between these 8-bit to 10-bit cameras.

I have not used a Full-Frame camera yet, but from what I have seen, the difference is still there, but the added dynamic range moves the "borerline" a bit. The "Full-Frame" sensor noise in the shadows is less, so that one might decide that even "S-Log3" is useable despite an 8-bit image chain. In the case of the later 10-bit Full Frame cameras, "S-Log3" seems to be well liked.

Anyway, I have been using the S-Logs for a while now and I will eventually post samples and analysis of "S-Log2". I might also post some samples of "S-Log3", but I might just skip it. As I wrote above, for APS-C, it is just a waste of time.
 
Last edited:
"S&Q" Time Lapse: Full Length Sunrise


This video was made using what I would call "sub-optimal" settings on my Sony a6400 with Sony's SEL20F28 (20mm F2.8) lens. I used "S&Q" at 1 fps. The a6400 S&Q does not support frame rates slower than 1 fps., so the resulting clip runs slower and longer than I would like. If you have seen my previous sunrise, it was made using another camera at 2 fps, which is a better frame rate for sunrise captures. If Sony asked me for recommendations, I would suggest adding frame rates of 1.5 and 2.0 frames per second. I believe that the 1.5 fps rate would also be useful, and it can be difficult to find a way to do this in post. The 20mm lens does "focus breath", so be sure to set it to manual for a time lapse.

Also S&Q on the a6400 only supports time lapse up to 1080p resolution, and there is no real reason for this limit. So I would suggest Sony should add UHD (4K) S&Q time-lapse support, for 1/2 second frame rates and longer.

Why is it Resized?

A long time ago when UHD support was new on YouTube, I was told that if you resize a Full HD file to UHD then more detail can be seen by the viewer even when viewing the file at 1080P. I have never confirmed that this is true, but I resized some Full HD files before uploading before in case it is true. If data cost is an issue, I would suggest that you view it at Full HD.


To cover the dynamic range for the overnight capture the camera mode was S-Log2 with exposure +1.7. I have been experimenting with S-Log2 for still pictures and so I had some idea what the result might be like for video. Recording video was the point in using S-Log2 for stills in the first place. I was aware that when being pushed to record video, the sensor might react differently, which was one of the first things I looked for. So far, I think that the a6400 reacts to S-Log2 closely enough that my tests and practice usages have been valuable.

Why is it Silent?

I did look for some background music, but I did not find something that I liked, so I left it this way. If I understand correctly, if I find a royalty free clip that YouTube provides, there is a "tool" to attach it later. If not, then I will have to work on it again and add the music in Magix or Pinnacle.

"[UHD-Silent]20221127 Sony a6400 S-Log2 Sun Rise"
posted Dec 5, 2022, [length 9:15]
""

I am including some frame captures of one of the two original clips used in this video. The file names include a rough time where the frames occur in the original clip. If I have the time, I will try to find out the offset from the start of the video to where the frames occur.
 

Attachments

  • C00_10-2022-12-01-19h10m48s402.png
    C00_10-2022-12-01-19h10m48s402.png
    877.1 KB · Views: 5
  • c01_10-2022-12-01-19h17m32s447.png
    c01_10-2022-12-01-19h17m32s447.png
    667.7 KB · Views: 6
  • c02_15-2022-12-01-19h30m16s427.png
    c02_15-2022-12-01-19h30m16s427.png
    719.9 KB · Views: 5
  • C04_00-2022-12-01-19h46m32s873.png
    C04_00-2022-12-01-19h46m32s873.png
    681 KB · Views: 6
  • c05_10-2022-12-01-20h11m49s832.png
    c05_10-2022-12-01-20h11m49s832.png
    727.9 KB · Views: 5
Last edited:
Various Settings for the previous "Sunrise":

As indicated it was "S-Log2", with "ITU709 Matrix", "Daylight" colour balance

The Magix adjustments were: Brightness 54, Contrast 56, Saturation 24
- these were supposed to be applied at 11 sec, but I think they turned out to be global adjustments
 
This is the final version of the time-lapse in the previous comment. I will probably remove the "old" version eventually. I am just leaving it right now for people curious about how much difference there is.

Magix Movie Edit Pro 16+ had a bug. I was able to remove every 2nd frame to effectively doubt the playback speed resulting in a playback time of 4:38. However, Magix failed to adjust the playback speed for the file resulting in a +9 min file with the last 4:38 being black (empty) screen). I could not find a way to correct this in my editing software, but it turned out that YouTube had the ability to make the adjustment after I uploaded it.

All music was YouTube Royalty free from Asher Fulero. The first was "Ceremonial Library" and the second was "Snowfall Butterflies".

" 221127 [UHD] Sunrise time lapse, Sony a6400 v1.0 with music",

posted Feb 23, 2023 by VidThreeNorth, [length 4:40]
""

Update: 2023-03-11 14:56

A couple of weeks ago I posted my S&Q time-lapse video to YouTube. This turned out differently from the usual. I received a message informing me that YouTube had "recommended" the video. I have never heard of this before. The view count has grown significantly from what I would expect for this type of video. No, I did not get hundreds of views. In a couple of weeks I got about 80 views. But considering the subject matter, it was a surprise. The sun comes up. Clouds blow around. Interesting if you like climatology, but not earth shaking stuff. Still, it was a nice surprise.
 
Last edited:

Most reactions

New Topics

Back
Top