VideoHelp Forum
+ Reply to Thread
Page 2 of 2
FirstFirst 1 2
Results 31 to 57 of 57
Thread
  1. Originally Posted by vaporeon800 View Post
    Many thanks for the samples.
    You are welcome. I got my answers regarding the "2000E vs 460E decision" but if you would like to see more samples for your personal Hi8/D8 comparison, let me know. I will be capturing a LOT of tapes in the next few days and I'd be glad to post more clips here.

    Originally Posted by jagabo View Post
    I think the 2000e capture is better overall. It has better black/bright levels. It may have a little more chroma noise or maybe it just shows up more because the the better levels. It's a little less sharp but you can easily adjust that with a horizontal sharpening filter. And of course, it doesn't have that big color inversion problem at the right edge.
    Jagabo, thanks a lot for all your help. I have been using the 2000E instead of 460E based on your suggestions and I just re-did a couple of tapes I'd previously captured with the 460E using the 2000E.

    So, I have one last question. I have about another 100 hours of footage to digitize and I am wondering; should I even bother to see if I can find another D8 camcorder that doesn't have the issues that my 460E has or is the 2000E capture quality about as good as I can get from these tapes?
    Quote Quote  
  2. Formerly 'vaporeon800' Brad's Avatar
    Join Date
    Apr 2001
    Location
    Vancouver, Canada
    Search PM
    On the subject of green and magenta lines along the right edge, I have seen a similar artifact on a DVB-S channel. It's not just a color artifact like for your tapes though. Comparing to the DVD, it looks like the discolored content is just the right-most pixel(s) stretched out rather than real picture content.

    Image
    [Attachment 17135 - Click to enlarge]
    Image
    [Attachment 17137 - Click to enlarge]

    Image
    [Attachment 17138 - Click to enlarge]
    Image
    [Attachment 17136 - Click to enlarge]
    Last edited by Brad; 8th Apr 2013 at 01:53.
    Quote Quote  
  3. Interesting. This is just a guess, but can this be happening during the analog->digital conversion process, I wonder. (Perhaps when some analog levels are not within correct limits?) Like I said, I may be wrong though. It is just a guess.
    Quote Quote  
  4. Formerly 'vaporeon800' Brad's Avatar
    Join Date
    Apr 2001
    Location
    Vancouver, Canada
    Search PM
    Just to be clear, there was no analog involved on my end in recording the DVB-S stream. Of course I don't know what manipulations were performed before that point, but the colors, levels, and detail don't appear to deviate very far from the DVD.
    Quote Quote  
  5. Member
    Join Date
    Jun 2013
    Location
    Austria
    Search PM
    Hi,

    I'm new here, I registered because I have quite the same problem with my Sony DCR-TRV 720E.
    Mine has the green edge (if red or orange objects are there) on both left and right sides, but not as intense as the TRV-460E owned by memrah.

    First I thought these edges are normal for my old V8/Hi8 tapes, but VHS-tapes, which I sent through the S-Video input of the cam, played by a good VHS recorder, had the same issue.

    So I captured with an old Terratek MPEG-II grabber, with another Camera (Sharp Viewcam VL-H400 via S-Video) and hello - there weren't any green edges on the same V8/Hi8 tapes, so...
    ...I captured from TRV 720E via Terratek and the green edges were far smaller (but not completely gone).

    Last try, I captured (via DV again) the camera picture itself, and - there were no green edges. Even not, when recorded (on Hi8 tape) and then grabbed via DV (that scratches my following theory).

    I can only assume that a part of my 720E's A/D converter is out of adjustment (since ever ?). TBC and DNR settings have no influence on the edge effect.
    I bought this used cam, once an expensive one, for grabbing old videos, but I have now to black out the borders with Virtual Dub, so it's nearly useless for me now, apart from having about 300GB of steady (TBC), but in fact useless DV-Avi files.

    I also got access to a service manual, which is for a number of those cams together (320/420/520/525/620/720), but messing with chroma and A/D settings seems to be a very hard task - without professional Video testing equipment like an oscilloscope. Sony made the access procedures to its camera systems very microcontroller-like. I would rather begin to refresh my 68k-assembler skills than doing this.

    Now my questions -
    Did meanwhile anybody find a solution for this problem ? Is it a general problem with these cams/series ? Maybe capacitor failure ? Or simply component tolerance/aging ? Or an incompatiibilty with the windows DV driver (that could give me at least some hope) ?

    I would appreciate any help, thank you for reading !
    Quote Quote  
  6. Hello.

    I am the original poster. Personally, I haven't found a solution to the problem. I ended up using my old analog camcorder (TRV2000E) for the Hi8 playback and the 460E as the analog/digital passthrough device. Rather like how you digitized your VHS tapes. What is interesting is, you say you had the same issue when digitizing VHS via the 720E. I had no such issue when digitizing analog sources via my 460E.

    For a while, I thought about buying an early but higher-end Digital8 model like the 720E, 530E, etc. to digitize my Hi8 tapes but now I see that some of those earlier models are plagued by the same issue that my 460E is suffering from.

    So maybe these camcorders didn't always have the color issue. Perhaps, like you suggest, it is caused by aging components like dried out or leaky capacitors. I don't know. I remember reading somewhere on the web, that certain other Sony Hi-8 camcorders failed due to aging capacitors.

    If I ever find a solution, I will post it here.
    Quote Quote  
  7. Member
    Join Date
    Jun 2013
    Location
    Austria
    Search PM
    Hello memrah,
    thank you for your quick answer.

    It seems that your faulty cam and mine, put together, would result in one fine working cam/dv8/hi8-Player/converter

    It's a shame that those yesterday expensive pieces are so faulty. Besides, mine (bought on the bay) has a warranty. So I have the chance to get my money back or to get another camera. Sometimes it is better to buy from traders than private.

    The used cam cdid cost me as much as a good Firewire frame grabber (or more), but less than a used JVC HR-DVS3 VHS/DV Recorder, which has no Hi8/V8 ability, they are expensive though they are well known for DV-Drive failure.

    But I wonder, if anyone had this effect/failure with his/hers Sony Cam, nobody went to a professional Video repair service - if there are remainig ones - at least to hear that a repair is possible but too expensive (or not) ?
    And - only a Sony failure ?
    Quote Quote  
  8. I have a Sony TR810E which unfortunately doesn't seem to have a strong enough output for my TV card (black and white picture only, all other devices (MiniDV camera, VHS recorder, DSLR) connected to the TV card give a nice color image, and the video camera connected to any other device (TV, DVD Recorder, MiniDV camera) works perfectly fine too).

    So I tried to connect it to my Panasonic NV-DS38 MiniDV camera, and that via Firewire to my computer... and it has EXACTLY the same issue (plus some weird blocking artifacts once in a while).

    I think it's a larger problem affecting many digital video cameras... though I don't know what exactly the problem is. The video looks fine on the screen of the MiniDV camera...
    Quote Quote  
  9. Member
    Join Date
    Jan 2016
    Location
    France
    Search PM
    While Googling on DV chroma I stumbled on this thread and though it is indeed a bit old I guess it still is of interest for many . Now and in the future.

    That is like many I recently had to make the good choice for getting some 50 analog video8/Hi8 tapes to a PC format for edit.

    Having read here and there I finally purchased a Sony D8 camcorder DCR-TRV230E. Mint condition.

    It was not my first option as I would prefer avoiding the DV codec for lossless interlaced YUV 4:2:2 (magicYUV)

    The 422 option being favoured having chroma correction in mind when editing , nothing more ; since in the end the final output (H264) will be 420 anyway ...

    Some loudspeaking folks at 'Lerepaire' ( top French video site) said they had done all kind of tests using Pro Decks, Pro range TBC , etc.. and that in the end the D8 option IS the (their) best option. Period.

    I still doubt it but it was such an easy solution and workflow getting straight to DV.avi, so I did that.

    And now I read this thread which points exactly the same green/pink bar artefact I get from my Sony D8 transcode !

    It seems quite sure it is an artefact that *any* D8 will produce , thus I doubt it has to do with some mechanical tape transport issue for instance.

    Moreover as mentionned in the thread the full size image playing via the D8 mini-LCD monitor is always free from green/pink bar artefact.


    All we know yet for sure is that it looks like a pure 8/Hi8 tape playback issue ( no bar from camera output in DV format)


    Otherwise I tried to guess some clue for supposedly better results from a D8 transcoding vs an Hi8 Pro player + Pro TBC, and I found no better explanation that the D8 can transcode in full range component format while Hi8 camcorder+TBC will only act in degraded Composite format ( S-Video) . My 2 cts here ...


    So it could be that the inner gain of D8 transcode is acting in component format (?) . Otherwise I see NO reason why a D8 can equal or outperform a high grade Hi8 deck (or camcorder) with an added Pro grade TBC taking care of the output signal.

    This is where I now have to disagree a bit with the marked preference (in the thread) for the TR2000E output vs 460E/D8.

    It is almost obvious at first glance that we compare video excerpts (flamingos) , or any given frame of it , that do have different gamma , different saturation , different contrast.

    It immediatly shows in the pink flamingos where the TR2000 image (as displayed ) simply fades to black all the upper part of the background vegetation. And as Jagabo pointed out the TR2000 capture also provides more color saturation . Therefore the D8 image while it does obviously show much more clearly the details of background vegetation it still looses (visually) some chroma depth due to an obvious difference in gamma range . Even whites are 'more white'. So could it be here simply a gamma matter in the first place ?

    Quite possibly a more precise comparison would have to set for a good A/B balance on gamma/saturation here and then quite possibly both images would be very close . With a possible drawback in the D8 image related to some residual effect of built-in filtering ( it looks like some subtle added sharpen/contrast) . But ALL that is quite manageable once it gets to the edit timeline.

    Assuming this , then yes TR2000 is here a possible better choice since it does not show the green/pink bar artefact ! Still I am not quite sure that a simple camcorder ( with no TBC behind) can remain as free of time base jitter and other various artefacts as a D8 , this in the long run , considering the capture of 50 hours. The D8 proved quite good at repetitive stable capture even given with its 'mini-TBC' . I have some 28 years video8 tapes which proved even free of any scrambled bottom lines ! quite a surprise

    This is where I am still left thinking in 2016 about what would be the final result of a good player deck (or even a good camcorder) feeding a pro grade TBC (?) ; be it in the limited S-Video format .

    ps : sorry for my poor (long time ago college) English and of course for this quite loooong post ......

    ps2 : (BTW I have opted for 70 Mbits (D1 resolution here) Cineform as the intermediate codec after QTGMC deinterlace since it shuffles the cards again from the (mjpeg) DV capture through wavelet in 10bit 4:2:2 ) .
    Quote Quote  
  10. To add some info, when digitizing Video8 tapes my Sony D8 DCR-TRV320E had the same edge color issue.
    Quote Quote  
  11. Originally Posted by kerryann View Post
    While Googling on DV chroma I stumbled on this thread and though it is indeed a bit old I guess it still is of interest for many . Now and in the future.

    Kerryann, thank you very much for sharing your experience, observations and thoughts. It looks like more and more people are running into the same technical issues discussed in this topic and finding themselves reading the posts here and contributing.
    Quote Quote  
  12. Member
    Join Date
    Jan 2016
    Location
    France
    Search PM
    Many thanks to you in the first place for starting this most interesting thread with very precise elements.

    As, indeed, it looks like many (if not everyone) have encountered the color edge issues using D8 transcoding.

    This had never been mentionned in 'lerepaire' site, otherwise a good French site with many knowledgeable members.
    I now find this even questionning (from these lerepaire forum 'specialists' )since --any-- D8 transcode suffers from these artefacts....

    these are quite strange , not being strictly 8 pixel wide , though quite close to.

    That is there can be 1 or 2 (or even rarely 3) edges color issues bands roughly 8 pix wide each , thus up to ..a possible 24 pixel wide band issue ! It then becomes a real problem. "luckily"most of the time it stays as 8 pixel , or so

    The real bonus of the D8 transcode seems ( if I am not wrong) to have access to a straight inner component transcode process ,
    rather that the usual composite output solution ( via S-video output of a camcorder, deck player )

    otherwise some high-end Sony camcorders and deck players also provide with built-in mini-TBC and Chroma noise filtering.

    But in 2016 it is quite more difficult to get such one 'amateur range' equipment in true mint condition , as compared to the more recent D8's.

    As for the Sony Hi8 EVO pro range of equipment , they often have been ....over used as it is just too normal for a pro purpose equipment ! so it is too much of an expensive gamble to look after...

    while sometimes we are also ready to spend much for these family-memory tapes to grasp the very last minor enhancement , anticipating to now have to watch these on 50" (and more) TV sets.

    For instance my early D8 tapes were shot on one of the best camcorder at that time (>2000€) , a Canon VME2 which featured ....a 220k mono CCD sensor !! Then if we add the very narrow color PAL standard (sub)carrier , it is quite a little miracle to me to still get some quality in the end . Luckily opticals were top notch (as an amateur camcorder indeed) which helped a lot
    Last edited by kerryann; 20th Jan 2016 at 08:49.
    Quote Quote  
  13. Formerly 'vaporeon800' Brad's Avatar
    Join Date
    Apr 2001
    Location
    Vancouver, Canada
    Search PM
    Originally Posted by kerryann View Post
    It seems quite sure it is an artefact that *any* D8 will produce
    I can't dispute this for PAL models, but there are samples on this forum from NTSC models that don't show the issue. Now that I think of it, I don't remember any NTSC user noting this problem.
    Quote Quote  
  14. Member
    Join Date
    Jan 2016
    Location
    France
    Search PM
    that's a very interesting point ! very weird then

    now I have no evidence either that any PAL D8 does produce these artefacts , just that they are present in yet quite a few D8 models.
    Does it mean that all units among these various models will produce the artefacts remains indeed more a question .

    We have to wait for a PAL D8 user that would claim for no problem , and this would turn even more tricky !

    As this PAL vs NTSC distingo is yet quite questioning ...
    Quote Quote  
  15. Hi there!

    I can testify to the general consensus that this is a problem with probably any European Sony Digital8 Camcorder.

    I bought a DCR-TRV345E, and although it's replay quality is much better than the recordings of my previous camcorder, CCD-TR45E, it creates this miscolored edge on the right side of the overscan area.
    I can also tell you that there is no such thing to be found when using that camcorder to record on a Digital8 tape and then playing it back, it only occurs on any Video8/Hi8 tape that I put in it.

    On a solution: I've tried to wrap my head around this issue for the whole day yesterday, learning back how to use Avisynth from scratch again, splitting the video output into its YUV channels for fixing, but just like jagabo previously said in this thread, it's near impossible to get a perfect result. It hurts my head so much, since I specifically bought this cam to replace my old one in terms of quality, TBC, DNR and so on.

    The "best" solution that I have come up with right now is to record the tapes twice, once in the TRV345E and once in the TR45E, and then to mask overlay the edge of the chroma channel from the good recording onto the miscolored one. That is, if both recordings are even running the same frames
    Quote Quote  
  16. Somehow I find comfort in learning from this thread that Digital8 was the best option to digitize Video8 quality wise. Because I've used a Sony PAL camcorder, too and like others assumed it was in the original recording already, just in the overscan area. With todays 16:9 TVs I am happy to crop off head switching noise at the bottom, but the sides are somewhat "precious". The idea with extrapolating the last good chroma pixels outwards was good. I'll try that as it is always better than having red cars or jackets turn green.
    Quote Quote  
  17. Member
    Join Date
    Mar 2014
    Location
    PALaland
    Search Comp PM
    Just a small addition, DCR-TRV120E PAL, used for Video 8 and Hi8 tapes transfer to PC via S-video. The green bar to the right is there.

    But not all the time, depends on the background colour.
    Last edited by kodec; 10th Feb 2019 at 09:06.
    Quote Quote  
  18. I realise this thread is quite old now, but bought a used DCR-TRV240E camera a couple of years ago exactly for the purpose of transferring old Video8 tapes to a digital format at the best quality possible, I only got around to do that now and I'm getting exactly the same issue as memrah – green band on the right side of the image when reading Video8 video through DV, which isn't present in the camera LCD when playing. This is the only place on the entire web that I can find that documents this issue, I was totally unaware of this when I sourced the camera. Digital8 tapes are obviously unaffected.

    This is a bit frustrating as I don't even have any camera that can do Video8 anymore. From reading this thread, I imagine all PAL Sony Handycams that do Video8/Hi8 to DV out must suffer from this issue. What's the general consensus around this, should I just source a Hi8 camera and use S-Video AV passthrough to the DCR-TRV240E in order to capture the tapes?
    Quote Quote  
  19. By no means ideal but sometimes better: I used the DeLogo filter http://rationalqm.us/delogo132/delogo.html and set it to use the pixels on the left to bleed into the band at the right. I think cropped the pxiels I couldn't blend and used black bars left and right to center the picture.
    You lose info/picture but it is better in some cases than the color band.
    Quote Quote  
  20. I use an Avisynth scrip like this for it to expand from the non-broken colours out to the edge. It's not perfect but it makes things less noticeable.

    Code:
    # U channel
    function fixsonychroma_b1(clip c,int goodline)
    {
        origW = c.width()
        origH = c.height()
        badcolor=c.vtoy()
        fillcolor=badcolor.crop((origW - goodline) / 2,0,2,0).pointresize(goodline / 2,origH)
        fixcolor=stackhorizontal(badcolor.crop(0,0,-(goodline / 2),0),fillcolor)
        ytouv(c.utoy,fixcolor,c)
    }
    
    # V channel - not always needed
    function fixsonychroma_b1_v(clip c,int goodline)
    {
        origW = c.width()
        origH = c.height()
        badcolor=c.utoy()
        fillcolor=badcolor.crop((origW - goodline) / 2,0,2,0).pointresize(goodline / 2,origH)
        fixcolor=stackhorizontal(badcolor.crop(0,0,-(goodline / 2),0),fillcolor)
        ytouv(fixcolor,c.vtoy,c)
    }
    
    
    AVISource("c:\file\name.avi")
    assumetff()
    fixsonychroma_b1(last,28)
    I also have a Hitachi Digital8 camera which gives pretty decent picture quality without the bar, and doesn't blue screen like the sonys. The TBC has a tendency to jitter vertically on some tapes though and the chroma is a bit noisy compared to the Sonys.
    Quote Quote  
  21. Member
    Join Date
    Jan 2023
    Location
    Budapest
    Search Comp PM
    This may be totally irrelevant for you by now, still... if anyone is still listening:

    I have the same problem. I mean not _similar_ but totally identical to yours. Same width, same characteristics of the discolored part. See attached screenshots.
    I have used two D8 camcorders (TRV345E and TRV480E PAL cameras) and both do the same. Both cameras in almost mint condition.
    Playing the tapes on an analog device is perfect, edge to edge.
    For me this looks like a system wide error on Sony side. I wonder whether it affects pro level D8 decks when playing analog tapes (if they exist at all).

    Right now, my only choice seems to use the D8 camera as a digitizer and use an external analog player as signal source. And re-capture hundreds of hours to get rid if 15 pixels worth wrong colors. I'm not sure I want this.
    Image Attached Thumbnails Click image for larger version

Name:	vlc_2wjPI1N5NG.jpg
Views:	19
Size:	128.8 KB
ID:	68486  

    Click image for larger version

Name:	vlc_3otYOAwAMq.png
Views:	26
Size:	1.82 MB
ID:	68487  

    Click image for larger version

Name:	vlc_jZUsLEE2iz.jpg
Views:	22
Size:	163.1 KB
ID:	68488  

    Quote Quote  
  22. Capturing Memories dellsam34's Avatar
    Join Date
    Jan 2016
    Location
    Member Since 2005, Re-joined in 2016
    Search PM
    I don't know what causes this, I've seen it numerous times. Some say bad capacitors in the Y/C board due to age, not sure. But it's clearly a shift in the chroma burst signal, The only solution as far as I know is get another camcorder or apply a special script to the digitized files to shift the chroma back to its position, See old posts, Neither way is convenient.
    Quote Quote  
  23. Originally Posted by shepherd View Post
    Right now, my only choice seems to use the D8 camera as a digitizer and use an external analog player as signal source. And re-capture hundreds of hours to get rid if 15 pixels worth wrong colors. I'm not sure I want this.
    In the analog times you would not see 3-5% of the image anyway. After you get rid of extra pixels on the sides, you will still have some black on the sides, and you will have the head switching area to get rid of, and in digital 625/50 for every 9 rows you need to remove 11 pixels on the sides. So... I would just crop it all around and call it done (it is not easy to honor 11:9 ratio and to ensure that the frame dimensions are divisible to 4 or 8 or 16, so will need to fudge aspect ratio a bit). But this is me.
    Quote Quote  
  24. Originally Posted by dellsam34 View Post
    I don't know what causes this, I've seen it numerous times. Some say bad capacitors in the Y/C board due to age, not sure. But it's clearly a shift in the chroma burst signal, The only solution as far as I know is get another camcorder or apply a special script to the digitized files to shift the chroma back to its position, See old posts, Neither way is convenient.
    It seems to be some sort of design flaw or something as noted in the newer (from around 97/98 ish) PAL Sony 8mm camcorders, it happens on all of them. Don't think it's anything to do with bad capacitors. It doesn't happen on the older Sony camcorders that used a less sofisticated/analog video process on playback or the Hitachi camcorders from I've seen (those come with their own tradeoffs though). It also doesn't happen on NTSC playback. The EV-C2000 8mm vcr I have here also does something similar and I've seen the same symptom as that on some other posts about the fancier hi8 vcrs so not sure it's limited to camcorders either.

    It could be possible that it's tied to some function like noise reduction or something that could be adjusted in the service settings (where you can also adjust head switch point etc) but those are only accessible with a special LANC adjustment remote which are quite rare. (it's supposedly possible to emulate them with a computer and a simple circuit that connects to parallel or serial ports but I haven't managed to get that to work so far.)
    Quote Quote  
  25. I see similar edge coloration on some 1970s Umatic captures (not mine, saw them online).
    Quote Quote  
  26. Originally Posted by shepherd View Post
    This may be totally irrelevant for you by now, still... if anyone is still listening:

    I have the same problem. I mean not _similar_ but totally identical to yours. Same width, same characteristics of the discolored part. See attached screenshots.
    I have used two D8 camcorders (TRV345E and TRV480E PAL cameras) and both do the same. Both cameras in almost mint condition.
    This is unfortunately still relevant to me, as, like you, I ended up capturing about 30 Video8 tapes via DV out on a DCR-TRV240E I got just for this purpose and haven't gotten around to process them all yet. It's the same conclusion that everyone else reached: this seems like a design flaw on all Sony Digital8 camcorders that can play Video8/Hi8. It most likely got past their quality control since it's PAL only, not NTSC, and the issue is invisible on the viewfinder / TVs of the time.

    I don't have anything that can play Video 8 natively anymore (I did find copies of all the Video8 tapes in MiniDV tapes afterwards but as I'm unsure of how the transfer was done – I'd have to check whether it's worth using those over the original Video8 tapes), so I ended up going the post-processing route. I can share the vapoursynth script I wrote to process these captures:
    Code:
    import vapoursynth as vs
    import havsfunc as haf
    import adjust
    
    core = vs.get_core()
    core.std.LoadPlugin(path="/usr/local/lib/libmvtools.dylib")
    
    clip = core.ffms2.Source(source=file)
    
    ## a) Crop extra pixels included in DV format
    # https://web.archive.org/web/20191026141308/https://bjohas.de/wiki/Tutorials/Video/Pixel_Aspect_Ratio
    # https://web.archive.org/web/20050404092907/http://www.bbc.co.uk/commissioning/branding/picturesize.shtml
    clip = core.std.Crop(clip, 6, 12, 0, 0)
    
    ## b) Correct the broken V channel at the edge coming from the D8 camera in the Video8 -> DV conversion
    # https://forum.videohelp.com/threads/354425-Hi8-capture-using-Digital8-camcorder-Edge-color-issues
    badPixels = 20/2
    
    y = core.std.ShufflePlanes(clips=clip, planes=0 , colorfamily=vs.GRAY)
    u = core.std.ShufflePlanes(clips=clip, planes=1 , colorfamily=vs.GRAY)
    vorig = core.std.ShufflePlanes(clips=clip, planes=2 ,colorfamily=vs.GRAY)
    v = vorig
    v = core.std.Crop(v, 0, badPixels, 0, 0)
    
    # Method 1 - inverting V and magnifying - didn't manage to get this to work
    #extra = core.std.CropAbs(vorig, badPixels, vorig.height, vorig.width-badPixels).std.Invert()
    #extra = adjust.Tweak(extra, cont=1)
    #extra = core.std.Levels(extra, min_in=0, max_in=200, min_out=0, max_out=255)
    
    # Method 2 - Reflecting V channel to the right
    #extra = core.std.FlipHorizontal(v).std.CropAbs(badPixels, v.height)
    
    # Method 3 - Stretch V channel to the right
    extra= core.std.CropAbs(v, 2, v.height, v.width-2).resize.Point(badPixels, v.height)
    
    v = core.std.StackHorizontal([v, extra])
    clip = core.std.ShufflePlanes(clips=[y, u, v], planes=[0, 0, 0], colorfamily=vs.YUV)
    
    ## c) Deinterlace
    clip = haf.QTGMC(clip, Preset='Slower', TFF=True)
    
    ## d) Resize to square pixels
    clip = core.resize.Bicubic(clip, 768, clip.height)
    
    ## e) Mask head switching - disabled
    # https://bavc.github.io/avaa/artifacts/head_switching_noise.html
    #bottomPixels = 10
    #
    #clip = core.std.Crop(clip, 0, 0, 0, bottomPixels).std.AddBorders(0, 0, 0, bottomPixels, color=[0, 128, 128])
    
    
    clip.set_output()
    And the bash script to output H.264/AAC MP4s suitable for playing on an average device from the input above, just for completion's sake.
    Code:
    #!/bin/bash
    ARG="$1"
    FILENAME="${ARG##*/}"
    NEW_VIDEO="${FILENAME%.*}.mp4"
    
    /usr/local/bin/vspipe -a "file=$ARG" --y4m -p ~/transcode_video8.vpy - | ffmpeg -i - -i "$ARG" -map 0:v -map 1:a -c:v libx264 -preset slow -crf 18 -c:a libfdk_aac -vbr 5 "$NEW_VIDEO"

    It does a bit more, namely deinterlacing and square pixel resampling. Relevant part is under b), but the values there assume you previously cropped the extra black pixels from step a) – adjust accordingly.

    I attempted three methods as you can see: I wasn't able to get jagabo's suggestion of playing with inverting the V channel working (suggestions welcome!). Reflecting the V channel to the right was too distracting to the eye. I found stretching the V channel the least bad – it's only really noticeable if you get something like the edge of a red object poking through the right.

    Any feedback or suggestions are welcome.
    Quote Quote  
  27. I had this discoloration from day 1 when my Hi-8 was brand new....
    Quote Quote  



Similar Threads

Visit our sponsor! Try DVDFab and backup Blu-rays!