G85 w/12-60mm F3.5-5.6 lens

[VidThreeNorth]

The G85 Menu System and Capabilities

I found this video shortly after I got my G85, but I did not really watch it closely at the time because I felt I should go through the camera on my own first. That way, I might find a way of using the camera that might be better suited to me. At this point, I have found the camera menu system and controls, ok, but a bit unnatural and with odd limits. So I thought it was about time to see what this video showed. The video is presentation is a bit fast, so if you have this camera and want to see what he has to say, then I suggest you have a paper and pencil handy to take notes, and pause the video as needed. I think I have had enough experience now that I can write some things about it, but I will put it off till some other time.

"Panasonic G85 / G80 / G81 Video Settings Explained 4K",
posted Jul 27, 2017 by "The Hybrid Shooter", [Length 8:22]
""

 

[VidThreeNorth]

[2020-08-07 As of the beginning of August I have changed the Luminance Range of my G85 for "MP4" videos from 16-255 to "0-255" which I discuss in this post. I also upgraded the firmware in my G85 to version 1.5 and the 12-32 zoom lens, I think to 1.2? I do not anticipate the firmware changes will make a difference to my tests, but it is a possibility. The Luminance Range change might affect my Exposure and Contrast tests slightly.]

The Luminance Range Settings and File Formats:

Regarding the "Hybrid Shooter" video:
"Panasonic G85 - G80 - G81 Video Settings Explained 4K"
(@03:45):

Luminance Level 0 - 255 (or 16 - 255)

This might go back to analog television signals. I am not sure. When setting up my camera I wanted to try this in 1080 first, but my 1080 was set to AVC in order to use the highest 1080 quality. I found that the "Luminance" control was grayed out, and for a long time I could not figure out how to make this adjustment. Then it occurred to me that it might be due to AVC. So I tried it in and MP4 file format and it worked. AVC is a "consumer electronics" organization and their goal it to set standards for that industry. Their mindset is not that progressive. I expect that the "16 - 255" is a part of the AVC digital video standard. So that is probably why it cannot be set for AVC formats.

About The File Formats:

So why do I want to use different formats at all?

Well first, the G85 has no AVC file format for UHD (3840 x 2160, aka "4K"). Actually, by now I expect that AVC has a standard, but the camera makers probably do not want to use it. "UHD" and "4K" are generally used serious video makers and any limits by the consumer electronics industry would probably not be welcomed. But the result is that my two Custom Presets for UHD video are both MP4 file formats.

But I also record in Full HD (1080) formats, and two important formats for 1080 are 24p and 30p. 24p is not supported on the G85 in MP4 files, so immediately, I have no choice. If I want to use 24p then it has to be in AVC. On the other hand 30p is only directly supported in MP4 at 20 mbps. However, there is a 60i (interlaced) format in AVC which is actually a 30p image capture which is stored ("recorded") as 60i. This setting is labelled "FHD/24M/30p", probably to avoid confusion. This file format is stored as 24 mbps and so, appears to be better and the 1080p in MP4 format which is 20 mbps. So, although I could choose MP4, in this case, I prefer to use the AVC file format.

FHD (1080) 60p is the only capture format that is equally supported in AVC and MP4. In either case it is 28 mbps.

So in Full HD, 24p and 60p, and the best quality "30p" are all available as AVC formats. And so, if I am working on "a 1080p project", then I will probably chose to work in all AVC, but if I am working on a "4k project" (either 24p or 30p, along with 1080p at 60 fps for slow motion), then I will chose to work in all MP4.

About Luminance ranges:

So what differences will the "0 - 255" range make? When using the "16 - 255" range, all colour values below 16 are considered "black" and are presumed not to be displayed. Actually, I learned it as "below 20", but I learned this a long time ago, and possibly it was before there was a real standard set -- I am probably just out of date. Since there are fewer colours available to use, the "jumps" from one colour to the next are bigger, and so if you start adjusting them, it is possible to end up with "posterization" earlier than if all the values are usable. Also, the lower levels tend to be noisier, so the image might end up looking better.

On the other hand, because there are values that are hidden in the "black" zone, it is possible to recover objects from the "shadows". I have done this before and found it useful. If all the values above 0 are displayed, then there is nothing left to recover. Moreover, the lower colours in the image will probably be noisier.

I have just started using "0 - 255" as of Aug 1, 2020, so for now, it will be in all my 4K files and probably all my Full HD 60 fps files. I do not know how this is going to work out. Not only that, but I will have to see whether there is a problem mixing "0 - 255" files with "16 - 255" files in the final outputs. There should not be, but I won't know until I try it.

 

[VidThreeNorth]

The Hybrid Shooter's Custom Flat Profile:

This is another post by "The Hybrid Shooter" that I found a while back. I have been intending to try out this custom profile, but I have delayed it until I had more experience with the G85 and the "CineLikeD" profile. It is based on the "Natural" profile, and since I have just recently started to try the "Natural" profile, I expect that I will be testing out this custom profile fairly soon.

"Panasonic G85 / G7 Custom Flat Picture Profile Inspired by Sony Cine4",
posted Nov 23, 2017 by "The Hybrid Shooter", [Length 6:43]
""

 

[VidThreeNorth]

"Natural" Contrast Tests:

Results where EV = +0:

+0 121.7 (taken from Comment #5 & 6 above, Apr 26 & May 13, 2020)
-1 105.3
-2 101.9
-3 94.2
-4 95.9
-5 94.1


The first thing you will notice "Contrast +0" = 121.7 but for "Contrast -1" = 105.3 which is a big jump. And then it back-tracks from "Contrast -3" to a higher value at "Contrast -4". This is what happens in real-world testing situations. Seemingly random factors can affect results.

In this case, I think I understand what happened, and it has to do with the angle of the sun changing, and clouds passing over. Ideally, I would go out and re-do the tests, but I don't have the time. And besides that, the tests showed me what I wanted to know.

What I learned is that the changes in contrast are not huge between settings. If you look at the pictures, the differences between two adjacent settings (like -3 and -4) are hard to tell unless you look at them consecutively on a screen, flipping back and forth. It would be hard even if they were side-by-side.


NOTE: All the video clips recorded for this set of tests were accidentally focused on the background instead of the card. Since I am not reading pixel data near the edges of the colour samples, that should not affect the test results. In fact, it might even make the readings more accurate.

"00080.MTS" 1920 x 1080
- focused on background

Windows:
Size 208,386,048 bytes
Created May 31, 2020, 10:02:29
Video
Length 1:11
Frame width 1920
Frame height 1080
Data rate 23249 kbps
Total bitrate 23441 kbps
Audio
Bit rate 192 kbps
Channels 2 (stereo)
Audio sample rate 48 kHz

Natural
Contrast -1, Exposure +0

"g85-01-080-Natural-ConMns1-EV0-16h26m43s354-C1.jpg"

"Black" Luma: 93.8
"Black + 1" Luma: 119.1
"Black + 2" Luma: 160.4
"White - 2" Luma: 197.2
"White - 1" Luma: 224.4
"White" Luma: 248.6

Contrast: 224.4 - 119.1 = 105.3


"00082.MTS"
- focused on background

Windows:
Size 233,035,776 bytes
Created May 31, 2020, 10:05:10
Video
Length 1:19
Frame width 1920
Frame height 1080
Data rate 23239 kbps
Total bitrate 23431 kbps
Audio
Bit rate 192 kbps
Channels 2 (stereo)
Audio sample rate 48 kHz

Natural (ID for this clip looks RIGHT)
Contrast -2, Exposure +0

"g85-082-Natural-ConMns2-EV0-22h12m32s557-C1.jpg"

"Black" Luma: 95.6
"Black + 1" Luma: 119.1
"Black + 2" Luma: 158.1
"White - 2" Luma: 193.4
"White - 1" Luma: 221.0
"White" Luma: 245.4

Contrast: 221.0 - 119.1 = 101.9


"00085.MTS"
- focused on background

Windows:
Size 186,421,248 bytes
Created May 31, 2020, 10:09:24
Video
Length 1:04
Frame width 1920
Frame height 1080
Data rate 23078 kbps
Total bitrate 23270 kbps
Audio
Bit rate 192 kbps
Channels 2 (stereo)
Audio sample rate 48 kHz

Natural
Contrast -3, Exposure +0

"g85-03-085-Natural-ConMns3-EV0-13h51m25s381-C1.jpg"

"Black" Luma: 106.8
"Black + 1" Luma: 128.5
"Black + 2" Luma: 164.4
"White - 2" Luma: 194.2
"White - 1" Luma: 222.6
"White" Luma: 247.2

Contrast: 222.6 - 128.5 = 94.2


"00083.MTS"
- focused on background

Windows:
Size 184,737,792 bytes
Created May 31, 2020, 10:06:50
Video
Length 1:03
Frame width 1920
Frame height 1080
Data rate 23074 kbps
Total bitrate 23266 kbps
Audio
Bit rate 192 kbps
Channels 2 (stereo)
Audio sample rate 48 kHz

Natural
Contrast -4, Exposure +0

"g85-04-083-Natural-ConMns4-EV0-17h30m37s021-C1.jpg"

"Black" Luma: 106.7
"Black + 1" Luma: 129.5
"Black + 2" Luma: 163.8
"White - 2" Luma: 196.4
"White - 1" Luma: 225.4
"White" Luma: 249.0

Contrast: 225.4 - 129.5 = 95.9


---
"00084.MTS"
- focused on background

Windows:
Size 193,665,024 bytes
Created May 31, 2020, 10:08:05
Video
Length 1:06
Frame width 1920
Frame height 1080
Data rate 23252 kbps
Total bitrate 23444 kbps
Audio
Bit rate 192 kbps
Channels 2 (stereo)
Audio sample rate 48 kHz

Natural
Contrast -5, Exposure +0

"g85-05-084-Natural-ConMns5-EV0-09h38m36s054-C1.jpg"

"Black" Luma: 107.2
"Black + 1" Luma: 128.5
"Black + 2" Luma: 162.5
"White - 2" Luma: 194.2
"White - 1" Luma: 222.6
"White" Luma: 247.6

Contrast: 222.6 - 128.5 = 94.1

 

[VidThreeNorth]

G85 Contrast Control

I don't know how many things I will have to say about issues I have with the G85. Some things are broader than the G85. Some things might be Panasonic issues. Some things might be much broader than that.

I think this issue might be "industry wide". I don't have enough experience with other brands to know that, but I know where the designs of these cameras came from, so I have some idea of what is likely.

This might well be a problem common to all camera designs, or at least many:

Setting contrast (where available at all) is treated as much less important than setting exposure compensation, and thus is often more difficult. It should not be. Contrast with current sensor technology is only a little less important than setting an exposure compensation and should be about as easily done.

In order to set contrast on the G85 I have to go to the control menu for the profile (see the video posted by "The Hybrid Shooter" linked in Comment #18). At that point I can open a sub-menu to adjust Contrast, Sharpening, Noise Reduction or Saturation. However, there are no aids like Histograms, Zebras or Wave Forms. to help to set the contrast. When I am setting contrast, the Histogram seems like the most natural tool to use. Instead, I have to get from my recording mode to the sub menu, then make a guess about how much of an adjustment I need, then get back to the recording screen and see what the histogram looks like.

What I would like might be, either a method of adjusting contrast while on the main recording screen, similar to adjusting the contrast, or to have the histogram available when I am adjusting the contrast while in the sub-menu.

 

[VidThreeNorth]

Offloading Settings?

Some high end cameras have the ability to save settings to files and offload them from the camera. This means that it is possible to restore settings later, or move them to another camera of the same type. I have not seen any such ability on the G85. If there is such an ability, I hope someone will tell me. If not, then I hope it can be added. It might seem late, but it would add to the brand's reputation.

Also, it should be reasonably possible to make settings files that can be moved between models. This is a characteristic of XML files. Programs that read XML files can skip parts of the files that do not apply. For example, setting a profile to one that does not exist would simply be ignored and left to a default setting. For cameras like the G85 and G95, I would expect many of the settings could be moved in both directions, which could speed up use quite a bit.

 

[VidThreeNorth]

Disappearing Screen Info

I set the screen to show info, and then when I am recording a video, after 1 min., the screen info disappears. I can get the screen info back by hitting the screen info button on the back of the camera. I would rather not have the screen info disappear at all. I need to monitor what is going on whenever I have a moment to do so. Remember that I am going to be busy watching the composition and what is happening as it is being recorded, so I do not always know in advance when something will happen that might cause a change that I will need to check. Also, when the camera is on a tripod, I am often trying to record without touching the camera. When using a long focal length, this is particularly important, so I do not want to touch it at all. I know that some people might want to seen the "clean" display, but at the least, this display switching should be optional and not forced on me.

 

[VidThreeNorth]

Yongnuo 42.5mm F1.7 Lens

I recently bought the first generation M43 Yongnuo 42.5mm F1.7 lens. These might still be available. There is a version 2 of this lens also which costs more. According to the website, the improvements in second-generation lenses compared to the first generation are:

"Appearance: New family style, simple and bright.
Drive Chip: The chip has been upgraded and muted (aperture and video) program optimization: using a new drive algorithm, better focus accuracy, faster focus speed, and effective resolution of the noise of the focus motor during recording"

It looks like the description says that the 2nd generation lens is 9 elements in 8 groups and minimum focus is 0.30m (max magnification 0.15 times). I believe this much also applies to the 1st generation.

[2021-03-14 18:03 Almost complete re-write of this paragraph.]
This lens failed to operate on the Yi-M1 (no focusing and no aperture control at all). In the next comment I linked a test that shows it also fails on the Olympus OM-D E-M1 Mkiii (complete failure for autofocus in Continuous mode). As such, I would suggest avoiding buying it. Even if you do not have a body with a problem right now, it appears that the risk of some future body failing is too high. In particular, I would want to know that it worked on a Blackmagic Pocket Cinema 4K.

Images:

"1670475.jpg" [Not Posted]
Windows 8.1:
Size: 6,619,626 bytes
Created March 11, 2021, 15:26:16
Partial EXIF
Program name Ver. 1.5
Dimensions: 4592 x 3448
Bit depth 24
Resolution unit 2
Color representation Uncalibrated (Adobe RGB)
Compressed bits/pixel 2
F-stop f/10
Exposure 1/25 sec.
ISO speed ISO-200
Exposure bias -1 step
Focal length 43 mm [sic]
Max aperture 1.53125
Metering mode Pattern
Flash mode Flash, compulsory
35mm focal length 86
Contrast Soft
Exposure program Shutter Priority
Saturation Normal
Sharpness Soft
White balance Manual
EXIF version 0230
Size 6.31 MB

Paintshop Pro:
White point 0.31 0.33
Primary chromaticites 0.64 0.33 0.21 0.71 0.15 0.06
Scene type Direct capture
Light source Reserved (9)
Image sensor tpe One-chip color areaa sensor
Image source Digital camera
White balance Manual (Daylight)
Gain control Low gain up

Because of the weather and advice to restrict travel I decided to take pictures of what I had on-hand. I had some nice strawberries, so that became the subject. I put them in a bowl and lit them roughly. The results were ok, but I ended up working with a small (f10) aperture to achieve depth of field. So despite good focus and flash, the image might be a bit soft from defraction. Also, I did not notice a hair on the far strawberry, which detracts from the composition. Since I washed and ate the strawberries before I realized the hair was in the picture, I could not re-do it, so the best that I could do was to avoided it in the "detail crop" version. At best that proves I am not a food photographer. So far the lens performance appears quite nice, but I will try again some other day -- especially, I need to look at the corners.

"01_1670475a-rsz1920-C1.jpg"
- whole frame resized

"01_1670475b-Crop1920-C1.jpg"
- detail crop

[2021-04-11 17:26
- comparison pictures:
"Yi Technology -- Yi-M1"]

 

[VidThreeNorth]

I found a test of the Yongnuo lens which gives enough detail to come to some conclusions. This first version lens is definitely to be avoided. It had focus problems on an Olympus OM-D E-M1 Mk iii completely failing in Continuous auto-focus mode. Again, the fact that your current camera might work with it might be enough, but I would suggest avoiding a lens that has such a significant chance of not working on bodies that you might want to buy in the future. Optical performance does not appear to be consistent. It is good stopped down but tends to fall apart a bit more than I would like when wide open. This first version operates slowly, which appears to be addressed in the 2nd version lens. But the basic glass is not worth it. There are plenty of used Olympus 45mm lenses around. They cost more, but will outperform this lens, and will be usable on all M43 bodies.
[2021-03-14 18:13 Correction]

"Yongnuo 42.5mm f1.7 Lens Review vs Olympus 45mm f1.8 ep.286",
posted Dec 22, 2020 by "Rob Trek", [length 20:05]
""