VideoHelp Forum

+ Reply to Thread
Page 2 of 2
FirstFirst 1 2
Results 31 to 44 of 44
Thread
  1. Originally Posted by Skiller View Post
    .... And in many cases this means it [DV] is actually the better choice.
    For PAL as well as for NTSC, you mean?
    Quote Quote  
  2. Originally Posted by Sharc View Post
    Originally Posted by Skiller View Post
    .... And in many cases this means it [DV] is actually the better choice.
    For PAL as well as for NTSC, you mean?
    Actually, there has always been a little less controversy about using DV for PAL because most people agree that the PAL 4:2:0 color model provides a little better color than the NTSC 4:1:1 approach.

    However, the main points made by both of us apply equally to NTSC and PAL.
    Quote Quote  
  3. Member Skiller's Avatar
    Join Date
    Oct 2013
    Location
    Germany
    Search PM
    Originally Posted by Sharc View Post
    Originally Posted by Skiller View Post
    .... And in many cases this means it [DV] is actually the better choice.
    For PAL as well as for NTSC, you mean?
    Yes, for both.
    While 4:1:1 is kind of annoying to deal with and not great (but not too bad for the super low resolution VHS chroma anyways), it is the lesser evil compared to the flaws of the numerous crappy capture devices the unaware user may buy. Plus the large number of possible mistakes to make in terms of software and settings.

    I'm with johnmeyer on this DV is obviously not the pinnacle, but it's a whole lot better than what most users get when they try to capture lossless without having years of experience.
    Quote Quote  
  4. Originally Posted by Skiller View Post
    Originally Posted by Sharc View Post
    Originally Posted by Skiller View Post
    .... And in many cases this means it [DV] is actually the better choice.
    For PAL as well as for NTSC, you mean?
    Yes, for both.
    While 4:1:1 is kind of annoying to deal with and not great (but not too bad for the super low resolution VHS chroma anyways), it is the lesser evil compared to the flaws of the numerous crappy capture devices the unaware user may buy. Plus the large number of possible mistakes to make in terms of software and settings.

    I'm with johnmeyer on this DV is obviously not the pinnacle, but it's a whole lot better than what most users get when they try to capture lossless without having years of experience.
    I see. The DV "robustness" advantage materializes only with an external VHS->DV conversion (e.g. videocam as digitizer in passthrough) as I understand it. When the DV codec is SW in the PC one would basically experience the same hassles as with lossless analog captures, right?
    Quote Quote  
  5. Member Skiller's Avatar
    Join Date
    Oct 2013
    Location
    Germany
    Search PM
    Yes, exactly.

    Choosing a DV codec for capturing with hardware that delivers uncompressed would be irrational in any case considering the alternatives available (Ut Video, HuffYUV, etc.). The advantages of DV only play out with a combination of native DV hardware plus an unexperienced user and/or crappy lossless-hardware.
    Quote Quote  
  6. Originally Posted by Sharc View Post
    I see. The DV "robustness" advantage materializes only with an external VHS->DV conversion (e.g. videocam as digitizer in passthrough) as I understand it. When the DV codec is SW in the PC one would basically experience the same hassles as with lossless analog captures, right?
    Yes, that is correct. If you have a choice of codecs to use with the capture card in your computer, DV should not be your first choice. The big advantage is using your DV or Digital8 camcorder in pass through mode. That capture chain is near-bulletproof. The only issue these days is finding a computer that has a Firewire/1394 card and then finding drivers that will work with modern versions of Windows. I still have lots of computers running XP, and everything purrs with that old workhorse.
    Quote Quote  
  7. @skiller and johnmeyer: thank you for clarification.
    Quote Quote  
  8. Hi,
    Sorry but I was wondering, won't it be better if after deinterlacing (I recommend QTGMC for that) then cropping the noisy borders, you should resize to 640x480 which is a true 4:3 resolution ?
    Your VHS tape has a 4:3 aspect ratio, so this may be closer to the orignal.
    I would be interested by the opinion of the others who have already replied.
    Quote Quote  
  9. Member Skiller's Avatar
    Join Date
    Oct 2013
    Location
    Germany
    Search PM
    640x480 is not recommended.

    For a h.264 based MKV or MP4 you just crop to mod4, then encode with SAR signaling. Upon playback there is one (and the only) resize to whatever player size or display you use. It is of course perfectly 4:3* then but you avoid the intermediate resize.
    If you really want to resize before encoding, go for an upscale to 1440x1080. There is no benefit in 640x480, actually it makes it slightly worse. It is legacy.

    * Edit: depends on cropping whether it still ends up as exactly 4:3 or just very close
    Last edited by Skiller; 3rd Oct 2022 at 20:05.
    Quote Quote  
  10. I for one don't even deinterlace VHS captures. I crop mod4 and filter interlaced (typically even/odd grouped fields, depending on the filter type) because I am not very happy how QTGMC processes VHS noise. I encode as interlaced with SAR signalling without any resizing. I know it's debatable but it's just my preferred workflow for my PAL VHS tapes.
    Quote Quote  
  11. Member Skiller's Avatar
    Join Date
    Oct 2013
    Location
    Germany
    Search PM
    [delete this]
    Quote Quote  
  12. Originally Posted by Sharc View Post
    I for one don't even deinterlace VHS captures. I crop mod4 and filter interlaced (typically even/odd grouped fields, depending on the filter type) because I am not very happy how QTGMC processes VHS noise. I encode as interlaced with SAR signalling without any resizing. I know it's debatable but it's just my preferred workflow for my PAL VHS tapes.
    I do the same thing. I do not understand some people's obsession with deinterlacing. You don't need to do it unless you are going to change the resolution and since up-resn'g is also a waste of time, you don't need to do either of them.

    There are plenty of other things you can do to VHS captures to dramatically improve them, like denoising, stabilization (of unstabilized hand-held camera movement), color correction, dropout removal (if you used cheap tapes), chroma correction (especially for NTSC color), etc.
    Quote Quote  
  13. Originally Posted by Hunk91 View Post
    Hi,
    Sorry but I was wondering, won't it be better if after deinterlacing (I recommend QTGMC for that) then cropping the noisy borders, you should resize to 640x480 which is a true 4:3 resolution ?
    Your VHS tape has a 4:3 aspect ratio, so this may be closer to the orignal.
    I would be interested by the opinion of the others who have already replied.
    If deinterlace and resize is involved:
    Deinterlace, crop sides 16 pixels total, resize to 4:3 and only then crop or mask.
    Keep your archives interlaced. Deinterlaced footage is for show, posts and bulletproof delivery.
    Quote Quote  
  14. Captures & Restoration lollo's Avatar
    Join Date
    Jul 2018
    Location
    Italy
    Search Comp PM
    I would be interested by the opinion of the others who have already replied.
    I never resize. I mask the bottom head switching noise, and beeing in the PAL land I crop the original 720x576 frame to 704x576, and then encode with a DAR of 4:3.
    For YouTube I upscale to 1440x1080 to allow less degradation in YouTube video processing.

    I do not understand some people's obsession with deinterlacing.
    Because, as said several times, at some point you need to deinterlace the video for watching. If the TV or the player is doing a better job than QTGMC (for example), it is ok to leave it interlaced, otherwise is better to do it in post-processing.

    In addition, many of the AviSynth filters are designed for progressive material, and using the SeparateFields().SelectXXX() is less effective than working with (lossless) deinterlaced material, because you use half of the temporal data.
    If somebody wants to keep the video interlaced, he can use a lossless deinterlacing (i.e. QTGMC(lossless=1)) prior to filtering and then interlace back at the end.
    Quote Quote  



Similar Threads