Yi Technology -- Yi-M1

Discussion in 'Mirrorless Cameras' started by VidThreeNorth, Aug 25, 2017.

  1. VidThreeNorth

    VidThreeNorth No longer a newbie, moving up!

    Joined:
    Oct 21, 2016
    Messages:
    386
    Likes Received:
    44
    Can others edit my Photos:
    Photos NOT OK to edit
    I have avoided posting "UHD" videos to YouTube this long because my limited (25 GB per month) Internet account does not allow me to waste such activity. I have noted a few times that I do not actually view videos in "UHD". Most of what I watch is less than 720p, and I watch 720p much more often than 1080p.

    This is the second "UHD" video I have posted that was made using the Yi-M1. I used three lenses, none of which was from Yi Technology. This is not because I am dissatisfied with the Yi Tech lenses -- actually I like them both, but I needed to use the Panasonic 14-42mm F3.5-5.6 zoom because it is "new". This "kit zoom" has "plastic-on-plastic" parts, just like the Yi Technology zoom lens, and many others. Such parts tend to be "sticky" and need to be "worn in". So I used that lens just so I could "wear it in".

    I used the SLR Magic 8mm F4 lens because it was the widest I have, and though I have been critical of its quality, I recognize that it can be adequate as long as editing is simple (changing exposure is ok, but cropping is out of the question).

    Finally, I used my 50mm F2.0 SMC Pentax lens because I was planning to use a focus pull. In fact, I recorded the focus pull, but the tripod was not steady enough, so in the end I cut the focus pull and just used the clip after the focus pull was done.

    "[UHD] 20190515 Toronto High Park Cherry Blossoms"



     
  2. VidThreeNorth

    VidThreeNorth No longer a newbie, moving up!

    Joined:
    Oct 21, 2016
    Messages:
    386
    Likes Received:
    44
    Can others edit my Photos:
    Photos NOT OK to edit


    About the "2019 Toronto, High Park, Cherry Blossoms" Video:

    The camera work was rushed for a few reasons. First, I only allocated three days to record, edit and upload this project. Also, for the recordings, I knew that I had to work quickly to avoid the crowds of visitors to the park. And then there was the weather. The cloud cover on the first day turned out perfectly while I was recording, but it cleared by the time I ran down the battery on the Yi-M1, leaving harsh contrast and deep black shadows. Two days later when I returned, it was similar.


    Part 1 (file 042)

    Frame Cap to PNG:
    "01-2019-05-26-09h34m04s508.png" [not uploaded]
    Size: 12,770,082 Bytes

    Convert to JPEG:
    "01-2019-05-26-09h34m04s508a-C1.jpg" [not uploaded]
    Size: 7,104,546 Bytes

    Resize:
    "Part01-2019-05-26-09h34m04s508b-rsz1640-C1.jpg"

    Detail Crop:
    Crop Start:
    3000,800
    Size 1640 x 1230

    "Part01-2019-05-26-09h34m04s508c-Crop01-C1.jpg"

    This "establish" clip is unusual both in composition and in the fact that I actually do not know very much about it. First, I do not know the focal length, nor the aperture, nor the exposure time. I picked the location moving the tripod and camera around trying to get the crest of the hill in the picture, and the foreground trees, all forming a window-like appearance. Then the focal length was adjust to crop the scene, and I did not record it. The aperture and shutter speed were automatic. I can guess that the ISO was 200, because that is where it stays for typical daylight exposure.

    I also do not know where it is focussed. In a situation like this, I have found that the Yi-M1 usually focuses on the background, but on occasion, it picks something near and throws the background out of focus. When I setup the composition and looked at the focus, it looked sharp all through on the screen, and I accepted it without manually picking a focus spot. I thought that I would be able to find the focus later, but I have not been able to do so.

    I think that it might have picked the crest of the hill in the foreground. The bark on the tree to the far right in the detail crop looks particularly good. Looking at the branches in that tree, some of them also look very good. The aperture must have been on the small end because even the cherry blossom clusters in the background seem to have some distinct shaping, yet there is noticeable softening past the foreground trees. Part of the "softness" is caused by the Y'CbCr 4:2:0 chroma sub-sampling and is used in typical video, and that disguises the focus too.

    Was there something about this Panasonic lens that added to the preference to focus in the foreground? I do not think so, but I cannot say for certain.

    ---
    Part 2
    Capture to PNG:
    "01-2019-05-26-10h24m45s856.png" [not uploaded]
    Size: 11,150,185 Bytes

    Convert to JPEG:
    "01-2019-05-26-10h24m45s856a-C1.jpg" [not uploaded]
    Size: 6,256,175 Bytes

    Resize:
    "Part02-2019-05-26-10h24m45s856b-rsz1640-C1.jpg"

    Crop:
    "Part02-2019-05-26-10h24m45s856cCrop01-C1.jpg"

    The recording situation for this clip was similar to the first, though I adjusted the tripod location based on the start and end of the pan. Unfortunately, I did not have a good tripod for this project, so it did not turn out as smooth as I would have liked. I wanted to have as many cherry trees as i could get in the composition at the end of the pan.

    Again, I do not know the focal length, aperture, shutter speed, nor the ISO. I made the adjustments and then did not take the time to note them. I think that when this frame was taken, the camera was probably focussed on the main cherry tree that fills a large part of the right side of the image. If you check the detail crop, the garbage bins in the bottom right are suspiciously sharp, but the grass around the base of the tree is detailed enough to indicate that the tree might have been the real target. But even the signs past the tree show that the depth of field was allowing a lot of leeway.

    The clouds were clearing quickly and I was lucky to get a set that I felt was perfect. It is ironic that this 2 min. short is particularly calming and relaxing, but the actual recording was tightly scheduled and rushed.

    ---
    Part 3

    "01-2019-05-30-10h11m52s075.png" [not uploaded]
    Size: 11,889,336 Bytes

    Convert to JPEG:
    "01-2019-05-30-10h11m52s075a-C1.jpg" [not uploaded]
    Size: 6,191,561 Bytes

    Resize 1904 x 1080:
    "Part03-2019-05-30-10h11m52s075b1-rsz1904-C1.jpg"

    Crop01 from top right
    1640 x 1230
    "Part03-2019-05-30-10h11m52s075c1-Crop01-C1.jpg"

    This part was recorded on May 17, which was the due date. The original clip I recorded was borderline. It was taken with the Panasonic lens and the autofocus could be seen "seeking" a bit. The focus was overall a good compromise "hyper focus", but the closest branches sometimes looked a bit soft. The clip was good enough, but I wanted to see if I could do better. So I went back with my Pentax 50mm F2.0. I tried a "focus pull" but again, my tripod was inadequate and merely touching the lens caused visible shaking, so I used the extra time at the end of the clip instead.

    Since the "focus pull" ended with the focus near the tips of the branches along the bottom, quite a bit of the background branches are out of focus. The detail crop of the upper right of the frame is actually just barely in the depth of field, but it is easier to see the branches because there are fewer of them.

    Do Pentax lenses have a "Pentax look"? Some people say that it is true. In this clip, the colours seem softer and a bit creamier, but that could just be the sunlight on that day with the camera pointed in that direction. I can't tell.

    ---
    Part 4

    Frame capture:
    "Part04-2019-06-04-13h45m34s168.png" [not uploaded]
    13,923,116 bytes

    Convert to JPEG:
    "Part04-2019-06-04-13h45m34s168a-C1.jpg" [not uploaded]
    Size: 8,261,654 Bytes

    Resize:

    "Part04-2019-06-04-13h45m34s168b-rsz1600-C1.jpg"


    Detail crop:

    Start 2180,650
    1240 x 930

    "Part04-2019-06-04-13h45m34s168c-Crop01-C1"


    Of the clips that I recorded on the 15th, this was the only one I put out of order in the final short. I did not want to end with it. It is also the only clip I recorded with an exposure adjustment. I have no record of how far off the EV was but I think it might have been EV = +1.3. I just lightened it until the blossoms looked good overall. I do not remember if there were any clouds near the sun. The blue sky retained in the corners was probably just the result of reduced optical transmission ("vignetting").

    The SLR Magic 8mm F4 lens was set at F8, which I feel is the only setting it can be used at, and the camera automatic ISO was probably ISO-200. I do not think I bothered to re-focus after the clip in part 5, so this was probably just "depth of field". The sun flare is not bad.

    ---
    Part 5 [clip 58]
    - road by Grenadier Pond

    Frame capture:
    "01-2019-05-22-19h06m37s098.png" [not uploaded]
    Size 12,714,522 Bytes

    Convert to JPEG:
    "01-2019-05-22-19h06m37s098a-C1.jpg" [not uploaded]
    Size: 7,193,245 Bytes [PaintShop Pro X9]

    Resize:

    "Part05-2019-05-22-19h06m37s098a-rsz1640-C1.jpg"

    Test Gamma Correction:


    Curves: [3 points added]
    in 5 out 14, in 28 out 63, in 125 out 184

    SmartFix
    Brightness
    Overall 0 [5 Recommended]
    Highlights -5
    Saturation 5
    Focus 30
    White balance [yes]
    Black 14
    White 2

    "Part05-2019-05-22-19h06m37s098b-rsz1640-C2.jpg"


    Detail Crop:

    Start 0,520
    Size: 1240 x 930

    "Part05-2019-05-22-19h06m37s098c-Crop01-C1.jpg"

    This clip was used without readjustments, and it actually was better than it looks. I tried adjusting the gamma in the captured frame and shadow detail is quite usable. That attempt went a bit further than I like and I think I would have preferred a gamma somewhere between the adjustment and the original. But that is the real reason I did not make it in the first place. It could take me hours fiddling with it to get it right, and it was already late into the night of the 17th -- my deadline. I started the upload to YouTube at 23:50 on the 17th and it was not completed until around 00:15 on the 18th, which was more work than I allocated for it.

    On the other hand, I am glad I did not alter this clip because I ran into the problem of having people's faces in it. If people are identifiable, then I need releases, either on paper or at least verbal. I would rather just not have faces in my videos than take chances. In this case it was borderline. If there are few pixels in the faces then they become "generic". In this case, it was becoming arguable. But the shadows helped obscure them. So even if I had the time to adjust the gamma in the clip I might have decided not to bother.
     

Share This Page

Search tags for this page

azs

,

best cameras

,
news
,

yi m1 forum

,
yi m1 jpeg processor
,
yim1 forum
,
yim1 hdr 2018