VideoHelp Forum

Our website is made possible by displaying online advertisements to our visitors. Consider supporting us by disable your adblocker or Try ConvertXtoDVD and convert all your movies to DVD. Free trial ! :)
+ Reply to Thread
Page 6 of 6
FirstFirst ... 4 5 6
Results 151 to 174 of 174
Thread
  1. Member
    Join Date
    Mar 2019
    Location
    NY, US
    Search Comp PM
    Originally Posted by celsoac View Post
    It is true that DV compresses more than ProRes. But I personally haven't found any difference in color, chroma, bleeding, etc.
    Have you had your eyes checked lately? (JOKING!) I am just waiting for someone to launch into why DV's 4:1:1 chroma is garbage

    However, all that matters is that YOU are happy with the results you are getting, and if DV looks good to you, then that's your solution!

    What are you using as a capture device?
    Quote Quote  
  2. Member
    Join Date
    Mar 2019
    Location
    NY, US
    Search Comp PM
    Celsoac, you also mentioned you have Final Cut Pro. I don't have it, but I find it hard to believe that a program like that doesn't give you the option to split clips or not - Have you looked into the preferences to be sure? I think you can tell it whether or not to stop capture on a dropped frame.. maybe it will do what you need. Also, I suspect if capturing via FCP you can select ProRes as your codec? No?
    Quote Quote  
  3. I use an ADVC-110 gizmo, something similar to yours. The iMovie > DV method is the only one that I've found that does not break the input into many files if there are sync problems, that's why I use it.
    Quote Quote  
  4. Originally Posted by Christina View Post
    Celsoac, you also mentioned you have Final Cut Pro. I don't have it, but I find it hard to believe that a program like that doesn't give you the option to split clips or not - Have you looked into the preferences to be sure? I think you can tell it whether or not to stop capture on a dropped frame.. maybe it will do what you need. Also, I suspect if capturing via FCP you can select ProRes as your codec? No?
    Thanks. Final Cut Pro (7) was great, then it was shitty for a few versions, then it is quite good again. It does capture directly into ProRes, I believe, I don't think you can choose the codec. But for the sake of me, nope, it does not allow you to disregard syncing problems in the input stream. When there is a lot of noise, or the old tape wiggles, FCP drops frames, so you also loose a fraction of audio. And no, I have not found how to solve this in Preferences, thanks. Perhaps there is a hack somewhere by editing some weird iOS file, but I have no idea.
    Quote Quote  
  5. Member Cornucopia's Avatar
    Join Date
    Oct 2001
    Location
    Deep in the Heart of Texas
    Search PM
    FCP was very, very good. FCX was crappy, then better, but it was never as good as FCP was. It is also no longer being updated, so don't expect improvements. However, it should allow you to select 29.97. Haven't checked recently, but QT ought to allow that also, even if it isn't the default.
    Also, look up my past posts on frame sizing in QT. v7x gave you 3 options. v10 dropped one option, but might have others, or might have dropped all but one. Apple majorly dropped the ball on that one.
    If you have dropped frames, the issue is either with your subsystem not able to keep up with the bitrate (not that likely), or your source is erratic and needs a TBC.
    And, yes, if you think ProRes looks no better than DV, you SHOULD get your eyes checked.

    Scott
    Quote Quote  
  6. Member
    Join Date
    Mar 2019
    Location
    NY, US
    Search Comp PM
    Scott, one quick thought/question on ProRes vs DV. I know FireWire does not = DV, but if you’re using some kind of device that converts analog to DV and then you send that signal to the computer but capture as ProRes rather than DV, WOULD the quality actually be any better than if you captured via DV or can it only just be as good as the DV quality is, except stored in a much larger file?
    Quote Quote  
  7. Originally Posted by Christina View Post
    Scott, one quick thought/question on ProRes vs DV. I know FireWire does not = DV, but if you’re using some kind of device that converts analog to DV and then you send that signal to the computer but capture as ProRes rather than DV, WOULD the quality actually be any better than if you captured via DV
    No.

    Originally Posted by Christina View Post
    or can it only just be as good as the DV quality is, except stored in a much larger file?
    The quality can only be worse than the DV. You have the losses of DV plus the losses of ProRes.

    There's one possible exception. Some DV decoders duplicate U and V samples rather than interpolating them when upsampling to 4:2:2 or 4:4:4. If your editor duplicates U and V when upsampling, but your DV to ProRes converter interpolates, you may get better colors (less posterization) from the ProRes version.
    Last edited by jagabo; 12th Apr 2019 at 18:34.
    Quote Quote  
  8. Member
    Join Date
    Mar 2019
    Location
    NY, US
    Search Comp PM
    Exactly, so that’s why someone capturing with a Canopus ADVC-100 could say they didn’t see a difference with ProRes, and don’t need to get their eyes checked.
    Quote Quote  
  9. Originally Posted by Cornucopia View Post
    FCP was very, very good. FCX was crappy, then better, but it was never as good as FCP was. It is also no longer being updated, so don't expect improvements. However, it should allow you to select 29.97. Haven't checked recently, but QT ought to allow that also, even if it isn't the default.
    Also, look up my past posts on frame sizing in QT. v7x gave you 3 options. v10 dropped one option, but might have others, or might have dropped all but one. Apple majorly dropped the ball on that one.
    If you have dropped frames, the issue is either with your subsystem not able to keep up with the bitrate (not that likely), or your source is erratic and needs a TBC.
    And, yes, if you think ProRes looks no better than DV, you SHOULD get your eyes checked.

    Scott
    Actually, when converting 25-year old VHS tapes from bad, full of noise TV reception played by a 2nd-hand VCR, my main concern is not particularly The Best Quality, but whether the conversion annoyingly drop frames or not. And funky iMovie does the job of capturing everything in just one chunk without audio-video syncing problems. Would you like to see two, 1-minute samples of the same material (DV and ProRes) and bet with me if you could tell the difference?
    Quote Quote  
  10. Member Cornucopia's Avatar
    Join Date
    Oct 2001
    Location
    Deep in the Heart of Texas
    Search PM
    Regardless of upconversion color sampling, if the cap device encodes to DV but then the capping/recording app upconverts to ProRes, you are adding a generation of lossy encoding, compounding any issues that already exist with the DV, but also including/introducing new forms of artifacts.

    IMO, the ONLY reason to do this is if you are STUCK with a DV device as your conduit, and also intend to do multi-origin editing/compositing/FX (particularly greenscreen). In that case, it would likely become necessary to upconvert to ProRes as your digital intermediate format, for use in the workflow prior to final export.
    If one was doing plain capping + cuts/Xfade editing + titles -> final master, it make more sense to start with one format and keep it in that format to enjoy the benefits of lossless smart rendering (where the only places it would re-render would be in the fade or title portion where it is anticipated).
    Many apps on Mac are stuck with ProRes because of Apple's heavy-handed control of AVFoundation media parameters, however, I would expect properly set up systems to still allow DV as a "legacy" capture format. Certainly, a reading of FCPX manual looks like deselecting "optimize media" will bypass the import/upconvert feature and should be able to use DV natively, as it is still listed as fully supported there. I stopped regularly using FCPX years ago, so cannot confirm this, however.

    Scott
    Quote Quote  
  11. Member Cornucopia's Avatar
    Join Date
    Oct 2001
    Location
    Deep in the Heart of Texas
    Search PM
    @celsoac,
    As Christina mentioned, I'm sure if you are doing VHS->DV->iMovie *.dv files or VHS->DV->FCPX *.prores.mov files, the DV damage has already been done to both, so the difference isn't worth mentioning. But that also isn't what most of us have been harping on in this thread, so no need to sidetrack. VHS->Uncompressed or VHS->ProRes is better looking than the previous alternatives, period. With VHS it may not always be readily noticeable at first to a layperson, but after editing & tweaking is done, the differences become more prominent. Especially with repeated critical viewing.
    The argument about economics (esp. with existing owned equipment) and workflow efficiency are the only valid arguments with any favorable nod towards DV. Quality isn't one of them, if you are discerning & experienced, and intending to put in the effort.
    I've done video transfers & conversions for almost 3 decades now, so your bet goading won't get a rise of me. Not worth my time.

    Scott
    Quote Quote  
  12. Originally Posted by Cornucopia View Post
    Regardless of upconversion color sampling, if the cap device encodes to DV but then the capping/recording app upconverts to ProRes, you are adding a generation of lossy encoding, compounding any issues that already exist with the DV, but also including/introducing new forms of artifacts.

    IMO, the ONLY reason to do this is if you are STUCK with a DV device as your conduit, and also intend to do multi-origin editing/compositing/FX (particularly greenscreen). In that case, it would likely become necessary to upconvert to ProRes as your digital intermediate format, for use in the workflow prior to final export.
    If one was doing plain capping + cuts/Xfade editing + titles -> final master, it make more sense to start with one format and keep it in that format to enjoy the benefits of lossless smart rendering (where the only places it would re-render would be in the fade or title portion where it is anticipated).
    Many apps on Mac are stuck with ProRes because of Apple's heavy-handed control of AVFoundation media parameters, however, I would expect properly set up systems to still allow DV as a "legacy" capture format. Certainly, a reading of FCPX manual looks like deselecting "optimize media" will bypass the import/upconvert feature and should be able to use DV natively, as it is still listed as fully supported there. I stopped regularly using FCPX years ago, so cannot confirm this, however.

    Scott
    Thanks for all of your advice. I am not a professional video producer/editor, I'm just a teacher who needs to prepare some materials and preserve a corpus of old recordings. I realize both DV and ProRes are lossy. When I used Windows years ago, I would encode to AVI (short segments). When I play around with MPEG2, FLV or other crappy codecs (including H264 at low bitrate), I first convert them to lossless FCP Uncompressed 10etc , then I splice, cut, trim, etc., and reconvert them. For tasks such as those, I don't even use FCP, but MPEG Streamclip, which encodes directly from DV or whatever without intermediate codecs.

    It would be wonderful to digitize directly from a VCR to an uncompressed format (it would also have been wonderful that FLAC, not MP3, would have become the standard compression format), but for what I want to do (preserve dozens of hours-long videotapes from TV) uncompressed video would exceed my means. First, the VCR should be the best quality. Second, I would need hundreds of petabytes, along with their respective backups.
    Quote Quote  
  13. Originally Posted by Cornucopia View Post
    @celsoac,
    As Christina mentioned, I'm sure if you are doing VHS->DV->iMovie *.dv files or VHS->DV->FCPX *.prores.mov files, the DV damage has already been done to both, so the difference isn't worth mentioning. But that also isn't what most of us have been harping on in this thread, so no need to sidetrack. VHS->Uncompressed or VHS->ProRes is better looking than the previous alternatives, period. With VHS it may not always be readily noticeable at first to a layperson, but after editing & tweaking is done, the differences become more prominent. Especially with repeated critical viewing.
    The argument about economics (esp. with existing owned equipment) and workflow efficiency are the only valid arguments with any favorable nod towards DV. Quality isn't one of them, if you are discerning & experienced, and intending to put in the effort.
    I've done video transfers & conversions for almost 3 decades now, so your bet goading won't get a rise of me. Not worth my time.

    Scott
    Thanks again. I would appreciate your expertise on the following: how can I digitize an old VHS to ProRes directly (not via DV) without stupidly losing frames because of sync issues which I don't care about, as all I want to have is a single continuous recording for each single VHS tape? I would really appreciate any hint on hardware and Mac software. Again, QuickTime => ProRes, which would be a good alternative, stopped on me after a few minutes (buffer issues?). Until I find that system, I'll have to continue to use the iMovie > DV method for storage.
    Quote Quote  
  14. Member Cornucopia's Avatar
    Join Date
    Oct 2001
    Location
    Deep in the Heart of Texas
    Search PM
    Lots depends on your budget, but for starters, if I were putting together a system for xfer and -HAD- to use a Mac, and wouldn't accept my aforementioned choices in hardware & software, I would try:

    1. Vcr that is recommended by others here such as lordsmurf who have done even more vhs xfer work than I (and have a backup, or 2, possibly same model, possibly alternate), with line tbc, dropout compensation and dnr options. Possibly S-vhs or D-vhs industrial models.
    2. Full frame TBC & frame sync - almost mandatory for serious work on rambunctious vhs sources - datavideo, for-a, prime image, tvone, etc. Throw in proc amp & scope to do it right.
    3. Magewell USB AIO, with custom hardware optimization tweaks via capture utility
    4. OBS, with Soundflower for Mac audio tweak
    5. Save to separate internal or Tbolt connected SSD, in ProRes.

    Not cheap, but the you chose the Mac, and by your own admission you're still working with lots of wiley vhs.

    Or if not needing to do many after all, use a known good service.

    Scott
    Quote Quote  
  15. Originally Posted by Cornucopia View Post
    Lots depends on your budget, but for starters, if I were putting together a system for xfer and -HAD- to use a Mac, and wouldn't accept my aforementioned choices in hardware & software, I would try:

    1. Vcr that is recommended by others here such as lordsmurf who have done even more vhs xfer work than I (and have a backup, or 2, possibly same model, possibly alternate), with line tbc, dropout compensation and dnr options. Possibly S-vhs or D-vhs industrial models.
    2. Full frame TBC & frame sync - almost mandatory for serious work on rambunctious vhs sources - datavideo, for-a, prime image, tvone, etc. Throw in proc amp & scope to do it right.
    3. Magewell USB AIO, with custom hardware optimization tweaks via capture utility
    4. OBS, with Soundflower for Mac audio tweak
    5. Save to separate internal or Tbolt connected SSD, in ProRes.

    Not cheap, but the you chose the Mac, and by your own admission you're still working with lots of wiley vhs.

    Or if not needing to do many after all, use a known good service.

    Scott
    Thank you again. I don't understand half of your abbreviations in points 2-4, but I'll try to learn. I haven't read the entire thread with all of your suggestions, but I'll have a look at them, too. Again, I am not a professional. But, since by now I do believe you that it's better to stay away from storing in DV, perhaps I would at least look into a gizmo that outputs ProRes directly (not DV => ProRes, which I now understand is a trick), or better yet lossless (that Magewell thing?).
    Quote Quote  
  16. Video Restorer lordsmurf's Avatar
    Join Date
    Jun 2003
    Location
    dFAQ.us/lordsmurf
    Search Comp PM
    Originally Posted by Christina View Post
    Also I should add before someone pounces on me - I am not the expert here, but I've also learned from hanging around these forums that supposedly Aja and Blackmagic are not great options for capture UNLESS you have a superb VCR and TBC between your tape and your computer. So, there's that.
    Even with TBCs, the Blackmagic cards have problems. And that's the problem.
    Aja is fine.
    Want my help? Ask here! (not via PM!)
    FAQs: Best Blank DiscsBest TBCsBest VCRs for captureRestore VHS
    Quote Quote  
  17. Originally Posted by Christina View Post
    Originally Posted by jagabo View Post
    Your capture resolution should be...
    Thank you for such a detailed explanation. I’m finding this topic a little challenging to wrap my head around but that definitely helped.

    My final destination for my files is an external hard drive to be played on a tv (either directly through USB if the TV has that or through some type of media player if not). This (playing on tv) is where I first noticed that one of my test videos was wider on the screen than the other. Prior to all of this, I didn’t know what anamorphic even meant. I didn’t know a pixel could be anything but square. I didn’t realize a video could be stored at one aspect ratio and played back as another.

    [head explodes]
    Hi there, Christina. So, I'm back here after having tested quite a few things, as I think we have common purposes and similar limitations (Canopus hardware). Let me explain what I'm doing:

    - First, I digitize VHS with and old version of iMovie that does not break the recording due to bad syncing etc. The result is, yes, limited DV, but that's what I've got. It makes no sense to digitize directly into ProRes through a DV device.
    - Now, this interlaced DV file is tricky -- interlacing in general is, unless the source is 24fps film. When the source is film, it is easy to restore a progressive frame made up of the two fields (top + bottom). But with an interlaced TV signal, actually each field is recorded at a different time, so blending them may yield blurry frames. And another problem is, of course, "teeth" or edges when blending the two fields. BUT,
    - I've found out that a very good option is deinterlacing the DV file by blending the fields into a progressive-frame format, specifically with JES Deinterlacer, free, very flexible. I've tried other deinterlacing/blending methods, including with Final Cut Pro, Handbrake (Bob), etc., and so far JES Deinterlacer is the best by far. The options I use are these: Blend fields, Adaptive (to movement), Local.
    - The output for this may be ProRes, or uncompressed 2vuy (lossless in regards to the DV source, which is lossy, of course). Since I'm going to fiddle around more with the file, I often choose uncompressed.
    - Since my video source is VHS PAL, the 576 horizontal lines don't fit nicely into any of the standard sizes for upload to, f.ex., YouTube, which are 640x480 or 1280x720 (YouTube is one of the final destinations, so I also need to take this into account). NTSC is 480 lines, but it has 720 real (coded, sampled) vertical pixels, so in my mind downscaling a NTSC file to 640x480 means losing real pixels, right? Wouldn't it be easier to upscale the file (by changing pixel ratio) to 720x540? If so, you've also got a non-standard resolution. So,
    - In order to preserve the best quality possible from a VHS source, in my humble opinion a good solution is to insert the 768x576 or 720x540 frame into a 1280x720 frame, with black edges around. MPEGStreamclip can do this: source 768x576, destination 1280x720 with negative cropping. So, the picture inside doesn't take the entire screen, but when played at 100% size (for example, in the computer) it maintains the original size/quality.
    - For the final output for playback I choose MP4 at high bitrate. Info can be edited with Subler

    In short, my tools are iMovie, JES Deinterlacer, MPEGStreamclip, Subler.

    End!

    Image
    [Attachment 48741 - Click to enlarge]
    Quote Quote  
  18. Video Restorer lordsmurf's Avatar
    Join Date
    Jun 2003
    Location
    dFAQ.us/lordsmurf
    Search Comp PM
    Originally Posted by celsoac View Post
    - First, I digitize VHS with and old version of iMovie that does not break the recording due to bad syncing etc. The result is, yes, limited DV, but that's what I've got. It makes no sense to digitize directly into ProRes through a DV device.
    Correct.

    - Now, this interlaced DV file is tricky
    Why do you find interlaced difficult? (It's not.)

    -- interlacing in general is, unless the source is 24fps film.
    When the source is film, it is easy to restore a progressive frame made up of the two fields (top + bottom).
    Film/progressive source is not interlaced, it's telecined.

    - I've found out that a very good option is deinterlacing the DV file by blending the fields into a progressive-frame format
    That's terrible advice. May as well not even convert the video, it'll look like crap.

    - Since my video source is VHS PAL, the 576 horizontal lines don't fit nicely into any of the standard sizes for upload to, f.ex., YouTube, which are 640x480 or 1280x720 (YouTube is one of the final destinations, so I also need to take this into account). NTSC is 480 lines, but it has 720 real (coded, sampled) vertical pixels, so in my mind downscaling a NTSC file to 640x480 means losing real pixels, right? Wouldn't it be easier to upscale the file (by changing pixel ratio) to 720x540? If so, you've also got a non-standard resolution.
    You're also confused about AR, discussed earlier in this thread. Read those posts.

    - In order to preserve the best quality possible from a VHS source,
    Your method doesn't even preserve 25% of the original quality, between the 50% DV loss, and the 50% deinterlace butchery.
    Want my help? Ask here! (not via PM!)
    FAQs: Best Blank DiscsBest TBCsBest VCRs for captureRestore VHS
    Quote Quote  
  19. Originally Posted by celsoac View Post
    - Since my video source is VHS PAL, the 576 horizontal lines don't fit nicely into any of the standard sizes for upload to, f.ex., YouTube, which are 640x480 or 1280x720 (YouTube is one of the final destinations, so I also need to take this into account). NTSC is 480 lines, but it has 720 real (coded, sampled) vertical pixels, so in my mind downscaling a NTSC file to 640x480 means losing real pixels, right?
    Like lordsmurf, I found this whole post full of nonsense. But I'd like to address one of the subjects covered, keeping in mind that one output will be for upload to YouTube. I, myself, when working with tapes output to two different formats, MP4 for upload to YouTube and DVD for other purposes. It doesn't have to be one or the other. I do the captures and intermediate work losslessly, and only resize for the final output formats. Or, when I captured as DV before, made all the intermediate encodes as lossless AVI.

    Yes, the 576i or p that PAL people often work with can present a problem for YouTube uploads. Make it progressive and then downscale to 480p or upscale to 720p. A good case can be made, I think, for making a 720p MP4 for upload to YouTube. But I have no idea why you think that has to be 1280x720. If you have a 1.33:1 aspect ratio (as 640x480 is), the 720p equivalent is 960x720. There's no reason in the world to add black bars on the left and right to make it 1280x720. Especially these days now that YouTube has the adaptable player. Any added black bars will just look amateurish.

    ...so in my mind downscaling a NTSC file to 640x480 means losing real pixels, right?
    Wrong.

    And, like lordsmurf, I think blend deinterlacing is about the worst possible choice. Too bad you're on a Mac as AviSynth has the best deinterlacers, if your sources are really interlaced and not telecined.
    Quote Quote  
  20. Originally Posted by lordsmurf View Post
    Originally Posted by celsoac View Post
    - First, I digitize VHS with and old version of iMovie that does not break the recording due to bad syncing etc. The result is, yes, limited DV, but that's what I've got. It makes no sense to digitize directly into ProRes through a DV device.
    Correct.

    - Now, this interlaced DV file is tricky
    Why do you find interlaced difficult? (It's not.)

    -- interlacing in general is, unless the source is 24fps film.
    When the source is film, it is easy to restore a progressive frame made up of the two fields (top + bottom).
    Film/progressive source is not interlaced, it's telecined.

    - I've found out that a very good option is deinterlacing the DV file by blending the fields into a progressive-frame format
    That's terrible advice. May as well not even convert the video, it'll look like crap.

    - Since my video source is VHS PAL, the 576 horizontal lines don't fit nicely into any of the standard sizes for upload to, f.ex., YouTube, which are 640x480 or 1280x720 (YouTube is one of the final destinations, so I also need to take this into account). NTSC is 480 lines, but it has 720 real (coded, sampled) vertical pixels, so in my mind downscaling a NTSC file to 640x480 means losing real pixels, right? Wouldn't it be easier to upscale the file (by changing pixel ratio) to 720x540? If so, you've also got a non-standard resolution.
    You're also confused about AR, discussed earlier in this thread. Read those posts.

    - In order to preserve the best quality possible from a VHS source,
    Your method doesn't even preserve 25% of the original quality, between the 50% DV loss, and the 50% deinterlace butchery.
    You don't even know why I do what I do with my videos, so throw your condescending tone right in the garbage.

    Here you can compare a short segment in interlaced DV and deinterlaced (both fields blended) with JES Deinterlacer, in uncompressed 2yuv and ProRes, from a 2002 video recorded from TV:
    https://www.dropbox.com/sh/wot8vslxwxtxth7/AACm13FMH2m1EM226ipHMQBoa?dl=0
    (of course, you have to download the files, not watch them online)

    The best way to preserve a DV digitalization is of course keep it unconverted. But it turns out that DV needs to be converted to MP4 to be shown on-line. That's the purpose of deinterlacing (turning it progressive). Of course I know that MP4 means a loss of quality.

    Of course I know what telecinema is. I've done inversed telecinema with shitty NTSC recordings from TV (the worst system ever created on earth) which were originally 24fps film. I may have expressed it wrongly because English is fortunately my third language.

    Same thing about aspect ratio. I may not express it correctly, but I know how to correct aspect distortion in a MOV or MP4 without reencoding the file, which is what matters to me.
    Last edited by celsoac; 20th Apr 2019 at 22:47.
    Quote Quote  
  21. Originally Posted by manono View Post
    Originally Posted by celsoac View Post
    - Since my video source is VHS PAL, the 576 horizontal lines don't fit nicely into any of the standard sizes for upload to, f.ex., YouTube, which are 640x480 or 1280x720 (YouTube is one of the final destinations, so I also need to take this into account). NTSC is 480 lines, but it has 720 real (coded, sampled) vertical pixels, so in my mind downscaling a NTSC file to 640x480 means losing real pixels, right?
    Like lordsmurf, I found this whole post full of nonsense. But I'd like to address one of the subjects covered, keeping in mind that one output will be for upload to YouTube. I, myself, when working with tapes output to two different formats, MP4 for upload to YouTube and DVD for other purposes. It doesn't have to be one or the other. I do the captures and intermediate work losslessly, and only resize for the final output formats. Or, when I captured as DV before, made all the intermediate encodes as lossless AVI.

    Yes, the 576i or p that PAL people often work with can present a problem for YouTube uploads. Make it progressive and then downscale to 480p or upscale to 720p. A good case can be made, I think, for making a 720p MP4 for upload to YouTube. But I have no idea why you think that has to be 1280x720. If you have a 1.33:1 aspect ratio (as 640x480 is), the 720p equivalent is 960x720. There's no reason in the world to add black bars on the left and right to make it 1280x720. Especially these days now that YouTube has the adaptable player. Any added black bars will just look amateurish.

    ...so in my mind downscaling a NTSC file to 640x480 means losing real pixels, right?
    Wrong.

    And, like lordsmurf, I think blend deinterlacing is about the worst possible choice. Too bad you're on a Mac as AviSynth has the best deinterlacers, if your sources are really interlaced and not telecined.
    1) I may have explained it wrongly, but it seems you got my point anyway.
    2) Yes, one of the output formats is MP4 for YouTube. I deinterlace the DV before uploading it because I don't like the outcome of YouTube turning interlaced input into progressive.
    3) Yes, I use 960 x 720. I do add black borders on the sides and top in order to preserve 1 PAL pixel = 1 720p pixel. If you upload a 768x576p file, YouTube downscales it to 640x480. And converting the 768x576 to 960x720p before uploading it loses quality, as the 720/576 factor is 1.25 , not an integral number (2, 3...). If you download the 720p with black borders and you remove them you get the original 768x576 file pixel by pixel, if you want that for whatever reason.
    Quote Quote  
  22. Video Restorer lordsmurf's Avatar
    Join Date
    Jun 2003
    Location
    dFAQ.us/lordsmurf
    Search Comp PM
    Originally Posted by celsoac View Post
    You don't even know why I do what I do with my videos,
    Based on your posts, your butchering your videos with a double-lossy conversion. What else is there to know?

    But it turns out that DV needs to be converted to MP4 to be shown on-line. That's the purpose of deinterlacing (turning it progressive). Of course I know that MP4 means a loss of quality.
    There is no MP4 format. There's the MPEG-4 wrapper (.mp4 extension), but the usual format is H.264. That format can carry both interlaced and progressive footage, though viewing device dictates what's needed. It also depends what "on-line" means. You can stream MPEG1/2 on a LAN to many devices, not just MPEG4.

    If your footage is destined for websites (Youtube, etc), the deinterlace will be required. But again, blended is butchery, low quality. Ideally use QTGMC, at very worst Yadif or Yadifmod (which is available in Avidemux Mac/Linux/Windows).

    Of course I know what telecinema is. I've done inversed telecinema with shitty NTSC recordings from TV (the worst system ever created on earth) which were originally 24fps film.
    Both PAL and NTSC have benefits and flaws. They both suck in their own way.

    I know how to correct aspect distortion in a MOV or MP4 without reencoding the file,
    There's really no way to do that, aside from setting flags. But that's often not valid, depending on usage. For example, you can't just re-flag the video for DVD authoring. Nor Youtube submission. Re-encode will happen, whether you realize it or not, often a hidden function of the software.

    What you see as "condescending" is actually someone trying to tell you how to not f-up your videos.

    Read it, do better, don't argue.
    Want my help? Ask here! (not via PM!)
    FAQs: Best Blank DiscsBest TBCsBest VCRs for captureRestore VHS
    Quote Quote  
  23. Originally Posted by lordsmurf View Post
    Originally Posted by celsoac View Post
    You don't even know why I do what I do with my videos,
    Based on your posts, your butchering your videos with a double-lossy conversion. What else is there to know?

    But it turns out that DV needs to be converted to MP4 to be shown on-line. That's the purpose of deinterlacing (turning it progressive). Of course I know that MP4 means a loss of quality.
    There is no MP4 format. There's the MPEG-4 wrapper (.mp4 extension), but the usual format is H.264. That format can carry both interlaced and progressive footage, though viewing device dictates what's needed. It also depends what "on-line" means. You can stream MPEG1/2 on a LAN to many devices, not just MPEG4.

    If your footage is destined for websites (Youtube, etc), the deinterlace will be required. But again, blended is butchery, low quality. Ideally use QTGMC, at very worst Yadif or Yadifmod (which is available in Avidemux Mac/Linux/Windows).

    Of course I know what telecinema is. I've done inversed telecinema with shitty NTSC recordings from TV (the worst system ever created on earth) which were originally 24fps film.
    Both PAL and NTSC have benefits and flaws. They both suck in their own way.

    I know how to correct aspect distortion in a MOV or MP4 without reencoding the file,
    There's really no way to do that, aside from setting flags. But that's often not valid, depending on usage. For example, you can't just re-flag the video for DVD authoring. Nor Youtube submission. Re-encode will happen, whether you realize it or not, often a hidden function of the software.

    What you see as "condescending" is actually someone trying to tell you how to not f-up your videos.

    Read it, do better, don't argue.
    Man, do I have to re-comment ALL of your snappy comments for your unsatiable display-of-knowledge?

    1) I KNOW MP4 is just a wrapper, an envelope. But, as you say, usually the format is H.264 "online", which, if you have read, in my case means YouTube.
    2) I KNOW many types of formats/envelopes can be streamed locally. I do that with movies in my home network.
    3) Have you watched the "butched" segment of a deinterlaced video whose link I sent? I've tried MANY deinterlacing methods with lots of software. Have you tried the one I'm talking about, JES Deinterlacer? It surprised me.
    4) "There's really no way to do that, aside from setting flags. But that's often not valid, depending on usage. For example, you can't just re-flag the video for DVD authoring. Nor Youtube submission. Re-encode will happen, whether you realize it or not, often a hidden function of the software."

    Have you ever used Subler. Well, here is the screen where I change the display appearance of a video that looks distorted (it's not real 4x3, or it looks 16x9 but the source was 4x3, etc.), by playing around with scaled size and aspect ratio. No reencoding takes place.
    Image
    [Attachment 48760 - Click to enlarge]
    Quote Quote  
  24. Video Restorer lordsmurf's Avatar
    Join Date
    Jun 2003
    Location
    dFAQ.us/lordsmurf
    Search Comp PM
    Originally Posted by celsoac View Post
    3) Have you watched the "butched" segment of a deinterlaced video whose link I sent? I've tried MANY deinterlacing methods with lots of software. Have you tried the one I'm talking about, JES Deinterlacer? It surprised me.
    Blended is blended, and looks terrible, software used makes zero difference.

    Have you ever used Subler. Well, here is the screen where I change the display appearance of a video that looks distorted (it's not real 4x3, or it looks 16x9 but the source was 4x3, etc.), by playing around with scaled size and aspect ratio. No reencoding takes place.
    Again, flagging is flagging, software choice makes zero difference. (And unless something has changed, Youtube ingest is 1x1 SAR as 16x9/etc DAR, or it re-encodes.)

    I sometimes re-flag videos for varying reasons, using MKVToolNix. But not all devices will cooperate with the re-flagging. My Samsung Blu-ray player will reject some resolutions, and distort the video. The only foolproof way to correct bad AR is via re-encoding. And again, some strict uses will entirely reject non-compliant flags, forcing re-encode.

    Video has rules, and you must abide by them. Video has constants and variables, and you must understand them. When you do whatever you want, problems happen, quality loss happens.
    Last edited by lordsmurf; 21st Apr 2019 at 00:45.
    Want my help? Ask here! (not via PM!)
    FAQs: Best Blank DiscsBest TBCsBest VCRs for captureRestore VHS
    Quote Quote  



Similar Threads