VideoHelp Forum

Try DVDFab and download streaming video, copy, convert or make Blu-rays,DVDs! Download free trial !
+ Reply to Thread
Page 1 of 5
1 2 3 ... LastLast
Results 1 to 30 of 128
Thread
  1. I moved this issue from Forum Thread 395968-Capture-program-for-8mm-only-offers-mpeg2-Is-there-better-quality, since we were deviating from the original post.

    dellsam34 noticed wierd lines in my uploaded file:
    Originally Posted by dellsam34 View Post
    Something wrong with your captures, In the attached screen shot I see some weird horizontal thick white lines separated by very thin black lines, that's not how resolution lines should look like, They should look like the one shown on the leg of the chair.

    Last edited by GrouseHiker; 28th Apr 2020 at 21:46.
    Quote Quote  
  2. Originally Posted by GrouseHiker View Post
    Thanks,

    I had noticed those lines, but didn't know enough to realize they were weird. That one was a DV capture via firewire. The one I just posted "xmas 4-2-2..." is an s-video capture through the Magewell card. I didn't notice lines in the recent upload. Both were from the same camera (same 8mm tape).

    Brad had noticed the 120 Hz flicker in that DV capture you're looking at, but I guess those horizontal lines are a much higher frequency than 120 Hz.

    I was wondering if maybe TBC in the camera or DNR might produce these issues. I'll experiment with that when I get back home.
    Well, I tried shutting off the TBC and DNR at the camera (Sony DCR-TRV350), and the weird lines remained. I also captured s-Video to a Magewell Pro Capture HDMI, and the weird lines match the DV capture. I also captured a snip with interlace combing to verify the weird lines don't seem to match the interlace scan lines:
    Image
    [Attachment 52974 - Click to enlarge]


    Update: Looking closer, the banding actually does line up with the interlace combing.
    Last edited by GrouseHiker; 30th Apr 2020 at 21:41.
    Quote Quote  
  3. I think one of the heads on your player is dirty or damaged
    Quote Quote  
  4. That is likely... just ordered cleaning swabs... I'll reevaluate in a week, after cleaning.
    Quote Quote  
  5. Member dellsam34's Avatar
    Join Date
    Jan 2016
    Location
    Paris Ca, 92345 Mexico
    Search PM
    Dirty or chipped heads don't exhibit that, It would be a form of grainy noise and miss tracking bars. Your heads are as clean as they can be.
    Quote Quote  
  6. The only thing in common between the DV and S-Video captures (both with lines) is the camera and the tape (ruling out the computer). The original Olympus camera only has composite out, but may be worth a try for comparison.
    Quote Quote  
  7. In any case one head is returning an image that has different brightness and gamma than the other.
    Quote Quote  
  8. Could this "varying brightness from scanline-to-scanline" observation by Brad be related? I was thinking scan-to-scan will be way higher than 120 Hz. Maybe he was talking about scan lines every other frame?

    Originally Posted by Brad View Post
    For some reason, the luma has 120Hz flicker. This Avisynth script probably won't mean anything to you, but for other readers:
    Code:
    SeparateFields()
     AssumeFrameBased().SeparateFields()
    If look at the fields we see varying brightness from scanline-to-scanline. If we separate the 240@60 again down to 120@120, the lines in the image disappear and we see the varying brightness as flicker.
    Quote Quote  
  9. Member dellsam34's Avatar
    Join Date
    Jan 2016
    Location
    Paris Ca, 92345 Mexico
    Search PM
    The problem appears on certain areas of the frame only, so not a head problem. Try another camcorder or another capture card to isolate the problem, The best way is using a TV with S-Video input.
    Quote Quote  
  10. We're looking at the same phenmenon but I believe his interpretation is wrong. If you look at the output after a single SeparateFields() you have 240 consecutive scan lines. Those scan lines alternate between dark and light. NTSC video runs at 30000/1000*525 = 15,734 scan lines per second. That makes the light/dark flicker 15.7 KHz, not 240 Hz.
    Quote Quote  
  11. The odd scan lines are at 15.7 kHz, but they're all darker in each, complete frame - which would create the banding. It looks to me like the entire odd-numbered "field" is darker. The kids have lizard skin and the pool tile changes to a darker shade in every odd field. It seems this would produce the banding. If that's the case, it would be a 30 Hz variation - first field brighter than the second field (or vice versa) of each, interlaced frame. I wouldn't even call it flicker - more like field/scan-variation banding. I grabbed another capture via s-video that shows this and ran a single SeparateFields() as you stated jagabo. That 60-fps file is attached.

    Also, dellsam34 mentioned trying a different capture card, but this banding showed up in my earlier DV "captures" - direct via firewire - before I owned a capture card.

    If y'all agree with this diagnosis, the question is - what could cause it? Are alternating fields read by different tape heads?

    I guess it's possible the old camera was defective or dirty and the information was written to tape this way... I hope this is not the case.
    Image Attached Files
    Quote Quote  
  12. I looked it up:
    http://repairfaq.cis.upenn.edu/sam/icets/vcr.htm
    "Every half revolution, each of the two heads writes one diagonal track which equals half an image. The first head writes one track, i.e., the first field (the odd numbered scanning lines). The second head writes a second track, i.e., the other half of the image (the second field: the even numbered scanning lines), which precisely fits in the first image. This corresponds to the interlacing principle..."
    Quote Quote  
  13. Originally Posted by dellsam34 View Post
    ... It would be a form of grainy noise and miss tracking bars.
    It kind of looks like grainy noise in each, odd field.
    Quote Quote  
  14. Member dellsam34's Avatar
    Join Date
    Jan 2016
    Location
    Paris Ca, 92345 Mexico
    Search PM
    No the heads read timing data as well, if you get a bad head the frames will be screwed, Don't you have a blank tape or a tape with blank space? record a clip and see for yourself.
    Quote Quote  
  15. OK, I misremembered how the data is laid down on tape. Each head reads/writes one field on each pass, not one scan line. So the alternating light/dark lines in each field are read by the same head. So it's not an issue of the two heads being different. It's something else in the electronics.
    Quote Quote  
  16. The original camera (Olympus VX-801) and the newer camera (Sony DCR-TRV350) both have 3 heads. One of the heads (leftmost I think) on the old camera looks reddish (photo) - maybe corrosion?
    Image
    [Attachment 53005 - Click to enlarge]


    This is the head at the opposite end of the trio (rightmost I believe) of the 3 heads:
    Image
    [Attachment 53006 - Click to enlarge]


    I hooked up composite video from the old camera to a TV - same banding lines visible, so I'm thinking this was laid down on the tape by the old camera...

    Is there a way in Avisynth to modify every other field?
    Quote Quote  
  17. Originally Posted by jagabo View Post
    OK, I misremembered how the data is laid down on tape. Each head reads/writes one field on each pass, not one scan line. So the alternating light/dark lines in each field are read by the same head. So it's not an issue of the two heads being different. It's something else in the electronics.
    It would seem to me the alternating fields are read by different heads, with the heads spaced exactly opposite on the head drum... as the first head deactivates (after reading field A), the opposite head activates (one scan increment later) to read field B. I may be wrong, but I gathered that concept from the link I provided above.
    Last edited by GrouseHiker; 30th Apr 2020 at 13:10.
    Quote Quote  
  18. Originally Posted by GrouseHiker View Post
    Originally Posted by jagabo View Post
    OK, I misremembered how the data is laid down on tape. Each head reads/writes one field on each pass, not one scan line. So the alternating light/dark lines in each field are read by the same head. So it's not an issue of the two heads being different. It's something else in the electronics.
    It would seem to me the alternating fields are read by different heads, with the heads spaced exactly opposite on the head drum... as the first head deactivates (after reading field A), the opposite head activates (one scan increment later) to read field B.
    You are correct. And that's what I said -- maybe it wasn't clear -- a head reads an entire field on it's pass over the tape. But the video doesn't have one dark field and one bright field. It has alternating dark and bright lines within each field.
    Last edited by jagabo; 30th Apr 2020 at 13:39.
    Quote Quote  
  19. Originally Posted by jagabo View Post
    You are correct. And that's what I said -- maybe it wasn't clear -- a head reads an entire field on it's pass over the tape. But the video doesn't have one dark field and one bright field. It has alternating dark and bright lines within each field.
    It makes total sense that as each head takes its turn across the tape, it is reading or writing a complete field.

    However, in the 60-fps video I recently uploaded, the entire field is dark and the entire next field is light... I believe this is what is producing the banding in the Frame when the fields are combined.
    Quote Quote  
  20. Member dellsam34's Avatar
    Join Date
    Jan 2016
    Location
    Paris Ca, 92345 Mexico
    Search PM
    Instead of making a new recording and test the workflow you prefer to chase the ghost.
    Quote Quote  
  21. Originally Posted by dellsam34 View Post
    Instead of making a new recording and test the workflow you prefer to chase the ghost.
    The newer camera is digital, so I don't believe a recording on that one would help. All the videos I need to capture are analog. However, have I not proven the issue is the tape by playing to the TV on a different camera?

    I don't have another analog camera to use to test the capture card, but I believe the capture card has been ruled out since it was not used to capture the original DV video via firewire... My first DV footage also contained the banding.

    If I'm missing something, please let me know.
    Quote Quote  
  22. Member dellsam34's Avatar
    Join Date
    Jan 2016
    Location
    Paris Ca, 92345 Mexico
    Search PM
    Did you try other tapes from different sources?
    Quote Quote  
  23. Originally Posted by dellsam34 View Post
    Did you try other tapes from different sources?
    All I have are digital tapes made in the newer DCR-TRV350 and older, analog tapes made in the old Olympus VX-801. I don't have any blank analog tapes; although I'm sure there are blank sections on all the tapes.
    Quote Quote  
  24. Originally Posted by GrouseHiker View Post
    However, in the 60-fps video I recently uploaded, the entire field is dark and the entire next field is light... I believe this is what is producing the banding in the Frame when the fields are combined.
    You already separated the fields in that video. So it's 60 fields per second. And you can see alternating light and dark lines within each field.
    Quote Quote  
  25. Originally Posted by jagabo View Post
    You already separated the fields in that video. So it's 60 fields per second. And you can see alternating light and dark lines within each field.
    You're right. I was assuming those lines in the fields were caused by separating the fields... but based on your experience, not.

    Then it seems I have two, different problems - banding being produced out of both heads as well as alternating quality in frames (from head to head). Do you agree?

    I will go back to a DV firewire capture and see if both issues show up.

    I'm polling friends to see if I can come up with another camera for testing. I don't trust the Olympus any more.
    Quote Quote  
  26. Originally Posted by GrouseHiker View Post
    Then it seems I have two, different problems - banding being produced out of both heads as well as alternating quality in frames (from head to head). Do you agree?
    Yes. One field is definitely noisier than the other.

    Both issues can probably be addressed in software. Though a cleaner capture would make things much easier.

    I did a little playing around with eliminating the line brightness problem. It's not as simple as just lightening every other line of the field (or darkening the other lines). The problem is more severe in lighter portions of the image and in more colorful portions of the image. I think it has something to do with the chroma decoding.
    Quote Quote  
  27. In this DV video brought in via firewire, the field-to-field (every other scanline) banding is present full frame. The interlace combing seems to line up with the banding. The images are screen captures from VirtualDub.
    Image
    [Attachment 53016 - Click to enlarge]


    This field-separated DV video is confusing. For some reason, the separate fields look cleaner than the s-video capture, but the fields (attached video and image below) show the horizontal lines that were evident in the s-video capture. (this image of a single field is resized to 640x480)
    Image
    [Attachment 53017 - Click to enlarge]


    Shouldn't the resized Field have blank lines where the other Field would be interlaced? Maybe the Field banding is the blank scan lines that aren't being filled by the other field?
    Image Attached Files
    Quote Quote  
  28. I've never seen a deinterlacer that replaced the missing field with blank lines (except one I made for a test). Your new SeparateFields() video looks pretty much the same as the last one.
    Quote Quote  
  29. Originally Posted by jagabo View Post
    I've never seen a deinterlacer that replaced the missing field with blank lines (except one I made for a test). Your new SeparateFields() video looks pretty much the same as the last one.
    Since the SeparateFields() output is 1/2 height, it seems there should be missing rows of pixels when the video is resized... Could the VirtualDub resizer be interpolating?

    I agree on the video being pretty much the same (with regard to banding), but the previous one was s-video captured by the Magewell card. This recent one is DV via firewire card (no capture card).
    Quote Quote  
  30. Originally Posted by GrouseHiker View Post
    Originally Posted by jagabo View Post
    I've never seen a deinterlacer that replaced the missing field with blank lines (except one I made for a test). Your new SeparateFields() video looks pretty much the same as the last one.
    Since the SeparateFields() output is 1/2 height, it seems there should be missing rows of pixels when the video is resized... Could the VirtualDub resizer be interpolating?
    VirtualDub doesn't know there are missing rows. It doesn't know it's fields. It simply sees 720x240 frames. So it just interpolates when it resizes.

    Originally Posted by GrouseHiker View Post
    I agree on the video being pretty much the same (with regard to banding), but the previous one was s-video captured by the Magewell card. This recent one is DV via firewire card (no capture card).
    So the problem is either the player or it's on the tape (the original camcorder).
    Quote Quote  



Similar Threads