VideoHelp Forum
+ Reply to Thread
Page 1 of 4
1 2 3 ... LastLast
Results 1 to 30 of 102
Thread
  1. Member
    Join Date
    Jul 2016
    Location
    USA
    Search Comp PM
    Hello!

    I'm currently in the process of converting some old CEDs I have to digital format. To do this I've been using a CED player that has composite output (as good as it gets on these machines - way better than RF modulated output on most).

    The output signal is running through a Videonics MX-1 to provide full frame TBC. The output looks rock solid, however I've noticed some artifacts that are present only in the time base corrected output. Namely some fuzz/noise on the border of objects. I'm wondering if this is inherent in all TBCs or if there is a way to improve it?

    On digitalfaq.com, I noticed that Datavideo TBC-100 and TBC-1000 are recommended as well as the AVToolbox AVT-8710. This got me wondering if the issue could just be the TBC. However, if I run relatively good video from my SVHS player into the MX-1 it doesn't exhibit the same kind of artifacts.

    Example:

    https://forum.videohelp.com/images/imgfiles/IVRJcr2.jpg

    Would a different TBC eliminate the noise on the edge of the face? The noise is not present if captured directly from the CED player, but the flagging is unreal.

    Any thoughts are greatly appreciated!
    Quote Quote  
  2. Do you see it on static edges too? If so, you'll need a line TBC to fix that horizontal jitter.

    https://forum.videohelp.com/threads/306272-Computer-video-capture-vs-vcr-to-dvd-combo?p...=1#post1882662
    Quote Quote  
  3. Member
    Join Date
    Jul 2016
    Location
    USA
    Search Comp PM
    Thanks for the reply! Yes, static edges exhibit it as well:

    Image
    [Attachment 37803 - Click to enlarge]


    As does the right border, but not the left. My card is capturing at 720x480, so there's 8 pixel borders on each side.

    Just for reference, this is what the raw captured video output looks like without the TBC in the loop:

    https://gfycat.com/UltimateThriftyIsopod

    Note the flagging/jiggling in parts of the video.

    For the line TBC, should that be inserted in the processing chain before or after the full-frame TBC?
    Quote Quote  
  4. Member hech54's Avatar
    Join Date
    Jul 2001
    Location
    Yank in Europe
    Search PM
    You are not going to do better than the remastered Star Trek releases.
    Quote Quote  
  5. Originally Posted by CED View Post
    For the line TBC, should that be inserted in the processing chain before or after the full-frame TBC?
    Yes. But you may not need the full frame TBC at all.
    Quote Quote  
  6. Member
    Join Date
    Jul 2016
    Location
    USA
    Search Comp PM
    Originally Posted by jagabo View Post
    Originally Posted by CED View Post
    For the line TBC, should that be inserted in the processing chain before or after the full-frame TBC?
    Yes. But you may not need the full frame TBC at all.
    The less processing the better I'll see if I can find one of those Panasonic DVD recorders you mentioned on eBay
    Quote Quote  
  7. Member
    Join Date
    Jul 2016
    Location
    USA
    Search Comp PM
    Originally Posted by jagabo View Post
    Originally Posted by CED View Post
    For the line TBC, should that be inserted in the processing chain before or after the full-frame TBC?
    Yes. But you may not need the full frame TBC at all.
    Just wanted to post back with some results. The Panasonic ES15 arrived yesterday and I did some recording tests last night. I'm impressed!

    This is an example of the original video: https://gfycat.com/UltimateThriftyIsopod

    This was the video after the MX-1: https://gfycat.com/BrownSpitefulBlackbear (ignore the poor color, it's due to another issue)

    And this is only using the TBC from ES15: https://gfycat.com/BabyishWellinformedBaleenwhale

    It's night and day. Very clean edges, and some of the most brilliant color I've captured. I'm guessing the ES15 does some color enhancement (maybe?). I didn't see anything in the manual for turning it off for Line In, but that's fine. It's a great picture. The jaggies and noise seem to be mostly gone. It's still visible to a small degree in the titles, but it's much improved.

    Thanks very much for the suggestion! Do you think a full-frame TBC would add much? If so I'll need to acquire another one to put in the pipeline. The combination of the ES15 and MX-1 produces oversaturated (and orangish) video. I didn't notice any improvement in stability beyond what the ES15 offered though.
    Quote Quote  
  8. It might be just this shot but the black level looks a little low and slightly crushed in the ES15 cap. It might also have been caused by reencoding the video at the web site. The hue looks a little off too. I would first adjusting the capture device's video proc amp settings to see if you can bring the black level up and eliminate the crushing. You should also play with the proc amp to adjust the hue, saturation, and bright level. If you don't have proc amp settings you can always adjust the video in software later. That won't help with hard crushed blacks but you'll be able to fix most of the other stuff.

    By the way, it's much better to upload unprocessed and losslessly compressed captures directly to this web site. Video streaming sites always reencode the video and you can't be sure what problems were caused by that reencoding. This site accepts up to 500 MB files.
    Quote Quote  
  9. Member
    Join Date
    Jul 2016
    Location
    USA
    Search Comp PM
    Originally Posted by jagabo View Post
    It might be just this shot but the black level looks a little low and slightly crushed in the ES15 cap.
    I think that's coming from the ES15 actually. It seems to only has one option for Composite video output and that is "Black Level Control." The options are "Lighter" and "Darker" I chose Darker as it seems to eliminate all the haze, but you're right some detail is getting crushed out.

    Originally Posted by jagabo View Post
    If you don't have proc amp settings you can always adjust the video in software later.
    There don't seem to be any processing amp settings on the capture device, at least according to the manual. I do have Sony VEGAS and Davinci Resolve I can use to clean it up a bit.

    Originally Posted by jagabo View Post
    By the way, it's much better to upload unprocessed and losslessly compressed captures directly to this web site. Video streaming sites always reencode the video and you can't be sure what problems were caused by that reencoding. This site accepts up to 500 MB files.
    Ah, good to know! I knew about the re-compression on most video sites, but I didn't realize it was possible to upload files directly. I've recaptured the video using the "Lighter" Black level setting. I'll upload that shortly.
    Quote Quote  
  10. Member
    Join Date
    Jul 2016
    Location
    USA
    Search Comp PM
    Originally Posted by jagabo View Post
    By the way, it's much better to upload unprocessed and losslessly compressed captures directly to this web site.
    Ok, I went back and recaptured a portion of this scene in all four available modes. Lighter and Darker on the Input and Lighter and Darker on the output. I pulled the video into Movie Studio Platinum and exported the segments as MPEG Program Streams. (The capture card has hardware MPEG-2 and captures in Transport Streams). I tried to grab exactly the same frames from each sample. The samples attached are named Nils[Input][Output].mpg. So for example NilsDarkerLighter.mpg means the input was set to Darker Blacks and the output was set to Lighter Blacks on the ES15.

    Note that Darker input / Lighter output is the default.

    Which do you think would be best for capturing?

    Thanks again for looking at these
    Image Attached Files
    Quote Quote  
  11. I believe the correct settings for North American DVD recorder recording from a North American VHS deck are "darker" on the input, "lighter" on the output. Those settings control the black level of the analog signal. North American NTSC adds "7.5 IRE setup" to the analog black level -- ie, the analog black level is high so it needs to be lowered when the signal is recorded on DVD (the digital data on the DVD should not include the black level boost). Then on output the DVD player boosts the black level back up to 7.5 IRE (because your TV will reduce it again for display), hence lighter on the output. Japanese NTSC analog video doesn't add the 7.5 IRE boost so the settings would be opposite in Japan. If you're using a mixture of Japanese and North American equipment you'll need to adjust accordingly.

    I'd normally recommend disabling the ES15's noise reduction circuit -- it can be too aggressive and create smearing and ghosting artifacts. But since you're using a hardware MPEG 2 encoding capture device you may have to leave the noise reduction enabled in order to reduce the artifacting from MPEG encoding (more noise requires higher bitrates). At least try a capture without the noise reduction to see which looks better. I don't remember exactly where the control is on the ES15 but I remember it wasn't where you would expect it to be. Look through the menus until you find it.

    Your video samples appear to have been blend deinterlaced and reencoded by Vegas. The original transport streams would have been a better choice for uploading. You don't need to worry about exactly aligning the videos. People here know how to deal with that. Or you can trim and remux your videos as MPG with a program like Mpg2cut2. These old videos were shot on film and telecined for NTSC broadcast/recording. They should be inverse telecined to restore the original 24 fps film frames, not deinterlaced.
    Quote Quote  
  12. Member
    Join Date
    Jul 2016
    Location
    USA
    Search Comp PM
    Thanks for the explanation about NTSC black levels! The Manual for the ES15 treats them as a preference (e.g., "how would you like the video to look?") and doesn't really indicate there's a proper setting for capturing the video correctly. I'm sure it's aimed at non-technical users, but that's an important piece of information

    The default setting is as you suggest, Darker on the input and Lighter on the output.

    I did stumble across the noise filter setting at one point. I believe it's under Display. It's set for Automatic and I left it that way. I can test out both ways and see if there's much different in the MPEG output files. I've recorded this particular CED a number of times in testing and always seems to be just a hair over 3 gigs every time. Nothing in the system was doing noise filtering previously so I'd guess the automatic setting on the ES15 never turned it on.

    The capture card (WinTV HVR 1955) was chosen because I initially needed something with an NTSC tuner and there weren't many options available new that still offered them. It also has an ATSC tuner which meant it would be useful after I finish this project for recording over the air broadcasts. I've since found a good CED player with Composite output which alleviated that need. I'd be open to getting a new capture card if it would help.

    In regards to the MPG files, you're correct. Sony Movie Studio / Vegas re-sample the source video by default unless you turn that off for each clip. I forgot to do that I've attached original interlaced transport stream that this was captured in.

    What would be an appropriate tool / method to reverse telecine these back to 24 fps? I'm guessing that just setting the output target to 24 fps in Vegas would not achieve the desired result (or would it?)
    Image Attached Files
    Quote Quote  
  13. Originally Posted by CED View Post
    I'd be open to getting a new capture card if it would help.
    It depends on how much time and effort you wish to put into this. Capturing uncompressed video (and typically saving with a lossless codec) and filtering it yourself has the potential to deliver the best qulaity. But you'll need another capture device, the capture files will be very large (on the order of 30 GB/hr) and you'll have to learn a lot about video filtering. At the other extreme, recording 1 hour per single layer DVD with a DVD recorder may get you 80 percent of that quality and is effortless.

    Originally Posted by CED View Post
    What would be an appropriate tool / method to reverse telecine these back to 24 fps?
    The gold standard would be AviSynth (a text script based video processing system). VirtualDub (free, GUI based) can usually do a decent job. I don't know Vegas' capabilities for this. Several of the free converters have some ability to deal with it too (Handbrake, Vidcoder, AutoGK, etc.).

    After seeing your TS file I'm pretty sure the noise reduction feature is kicking in. I can see some ghosting and smearing. My guess is that's not on the disc itself.
    Quote Quote  
  14. Member
    Join Date
    Jul 2016
    Location
    USA
    Search Comp PM
    Originally Posted by jagabo View Post
    It depends on how much time and effort you wish to put into this. Capturing uncompressed video (and typically saving with a lossless codec) and filtering it yourself has the potential to deliver the best qulaity. But you'll need another capture device, the capture files will be very large (on the order of 30 GB/hr) and you'll have to learn a lot about video filtering. At the other extreme, recording 1 hour per single layer DVD with a DVD recorder may get you 80 percent of that quality and is effortless.
    I'm willing to put in some time on this. I've been working on this project for about two months now and learning quite a bit along the way. I've upgraded a several components to improve quality, so that's fine. It's all part of the learning process. I wish I'd found this forum before I began. I had to rely on a little bit of experience I had from capturing some old home videos on a much older WinTV card about 15 years ago.

    Right now my goal is better quality. The DVD player's TBC has made a big difference and the WinTV is doing a good job, but I'm open to improving it.

    What would you recommend as a good capture card for capturing raw?

    Originally Posted by jagabo View Post
    The gold standard would be AviSynth (a text script based video processing system). VirtualDub (free, GUI based) can usually do a decent job. I don't know Vegas' capabilities for this. Several of the free converters have some ability to deal with it too (Handbrake, Vidcoder, AutoGK, etc.).
    VirtualDub is the only one of these I'm really familiar with. I've used it for video editing and transcoding in the past. Is there a tutorial/walkthrough for any of these to set up a reverse telecine? I'm happy to learn

    Originally Posted by jagabo View Post
    After seeing your TS file I'm pretty sure the noise reduction feature is kicking in. I can see some ghosting and smearing. My guess is that's not on the disc itself.
    I found the noise filter setting again. It's under Display->Video. I've set it to Off instead of Automatic. I'll recapture that bit again with the setting forced off and see if that makes any difference.
    Quote Quote  
  15. Member
    Join Date
    Jul 2016
    Location
    USA
    Search Comp PM
    I decided to capture the CED directly to the DVD player, just to see how it would work. I ripped the MPEG data from the DVD using MakeMKV, so no re-encoding was done. Then trimmed a section out with a Free Video Dub (it doesn't re-encode). The result is attached. Unfortunately the quality seems poor even thought the ES15 was set to record in the highest recording mode (XP). Perhaps the attached has less ghosting though (?)

    My main reason for doing this was to see if the WinTV HVR might be the cause of the ghosting. I played some video back through the system and I think I see what you're talking about. Unfortunately the ES15's noise filter was turned off. I'll capture some of that as well.
    Image Attached Files
    Quote Quote  
  16. Originally Posted by CED View Post
    I decided to capture the CED directly to the DVD player, just to see how it would work.
    This video is much better than the earlier one. Yes, it has more noise (probably why you think it looks worse) but it also has more fine detail and less ghosting. Open the earlier TS video in VirtualDub and step through frame by frame. At 3.97 seconds into the video, frame 119, it cuts from the guy in the yellow shirt to Kirk and Spock. But you can still see the guy in the background -- for the next 4 or 5 frames. In the new video it's only 1, maybe 2 frames. I suspect the latter is noise reduction done by the studio when producing the CED. At other places you can see more file details, wrinkles in the people's brows, the weave of the cloth, more strands of hair, etc. You can always apply your own noise filtering if you want to obliterate all those details!

    Here's a quick cleanup where I restored the original 23.976 fps film frames, removed the green cast, reduced oversharpening halos, stabilized the camera shake, cleaned up edges, removed spots and a little noise, and sharpened the luma and chroma. On the left is the original video with just the IVTC and on the right with all the above.

    I should have removed a few rows/columns at the edges of the frame too -- so you don't see the black borders from the desharker moving around.
    Image Attached Files
    Quote Quote  
  17. Member
    Join Date
    Jul 2016
    Location
    USA
    Search Comp PM
    After reviewing the video recorded on DVD again this morning, what I was perceiving as "lower quality" is probably just the noise. Comparing the direct DVD recording with older recordings I've made, the older recordings seem to be smoothed and have some detail loss.

    Unfortunately this is present in the oldest recordings I've made, which makes me think this is an inherent noise filter in the WinTV HVR 1955 and not something introduced by the ES15.
    Quote Quote  
  18. Originally Posted by CED View Post
    Unfortunately this is present in the oldest recordings I've made, which makes me think this is an inherent noise filter in the WinTV HVR 1955 and not something introduced by the ES15.
    Some of it may indeed be in the HVR 1955. If Hauppauge's software doesn't give access to the proc amp there may still be some hope. You can try using GraphEdit or GraphStudio. Select Graph -> Insert Filter. Exactly where the proc amp controls are varies from device to device. Try adding any capture, crossbar, or other filters related to your capture device. Back at the main window right click on a filter and select Properties. Look for proc amp controls there. You may find sharpness, noise reduction, brightness, contrast, hue, saturation, etc. These can often be adjust "live" while you preview the video in your capture software.
    Quote Quote  
  19. Member
    Join Date
    Jul 2016
    Location
    USA
    Search Comp PM
    Originally Posted by jagabo View Post
    Some of it may indeed be in the HVR 1955. If Hauppauge's software doesn't give access to the proc amp there may still be some hope. You can try using GraphEdit or GraphStudio. Select Graph -> Insert Filter. Exactly where the proc amp controls are varies from device to device.
    GraphStudioNext was able to get a hold of some of the filter controls. The DNR seems to be part of the MPEG encoding filter. By default the the temporal filter is set to level 8 and the spatial is set to 0. I turned the spatial all the way down to 0 thinking this would turn it all the way off. It seems some filtering is still applied.



    The video captured while those settings were in effect does have more noise, but there I believe there may still be more detail in the DVD recording. Particularly, looking at Kirk's shirt it seems some of the color variation in the fabric is visible in the DVD capture. It seems less so in the WinTV capture. I'll attach a clip. Note that while the attached file is an mp4 container, the included MPEG-2 stream is the raw capture.

    Do you think any other settings on the filters would help? Perhaps the thresholds?
    Image Attached Files
    Quote Quote  
  20. Yes, then new cap with the Hauppauge has more detail and noise than the earlier cap, but not as much as the ES15 cap. I don't remember -- does the ES15 have a sharpness control? Maybe that was turned up and accentuated the detail and noise (especially the latter) in its caps?
    Quote Quote  
  21. Member
    Join Date
    Jul 2016
    Location
    USA
    Search Comp PM
    Originally Posted by jagabo View Post
    Yes, it has more noise (probably why you think it looks worse) but it also has more fine detail and less ghosting. Open the earlier TS video in VirtualDub and step through frame by frame. At 3.97 seconds into the video, frame 119, it cuts from the guy in the yellow shirt to Kirk and Spock. But you can still see the guy in the background -- for the next 4 or 5 frames. In the new video it's only 1, maybe 2 frames. I suspect the latter is noise reduction done by the studio when producing the CED. At other places you can see more file details, wrinkles in the people's brows, the weave of the cloth, more strands of hair, etc. You can always apply your own noise filtering if you want to obliterate all those details!
    I totally missed this post earlier! I think I'd already hit reply and was writing up a response while looking at some of the video around the same time. I see what you mean about the earlier TS capture. The ghosting is more prominent in another CED I was capturing. It has a lot more motion and it's even more obvious that there are ghost trails left behind when people move suddenly.

    It seems that despite my efforts to put together a capture system with my PC, the best solution I have so far is a $60 DVD recorder from eBay, lol

    I definitely want to keep as much detail as possible, but reducing the noise a bit would be good.

    Originally Posted by jagabo View Post
    Here's a quick cleanup where I restored the original 23.976 fps film frames, removed the green cast, reduced oversharpening halos, stabilized the camera shake, cleaned up edges, removed spots and a little noise, and sharpened the luma and chroma. On the left is the original video with just the IVTC and on the right with all the above.

    I should have removed a few rows/columns at the edges of the frame too -- so you don't see the black borders from the desharker moving around.
    This looks amazing! May I ask what tools you used to clean up the video and any plugins? If I could get it to this level, I think this would be amazing. This episode has the shakes pretty bad. There's a scene later where it looks like the characters are jumping around a few inches. It's present even in the HD masters. The deshaker you used makes it look rock solid though!

    editing this in:
    Originally Posted by jagabo View Post
    Yes, then new cap with the Hauppauge has more detail and noise than the earlier cap, but not as much as the ES15 cap. I don't remember -- does the ES15 have a sharpness control? Maybe that was turned up and accentuated the detail and noise (especially the latter) in its caps?
    It doesn't seem to have a direct Sharpness control. There is a "Picture" setting which has the following options according to the manual. It seems to affect contrast and brightness as well when I tried it.

    Picture

    Select the picture mode.
    • Normal: Default setting
    • Soft: Soft picture with fewer video artifacts
    • Fine: Details are sharper
    • Cinema: Mellows movies, enhances detail in dark scenes

    I've had this set to Normal every time I've captured with the ES15 in the loop.
    Last edited by CED; 24th Jul 2016 at 20:18.
    Quote Quote  
  22. I did all the processing in AviSynth and encoded with the x264 command line encoder. The script I used was:

    Code:
    Mpeg2Source("title00_cut_track1_eng.d2v", CPU2="ooooxx", Info=3) # CPU settings = reduce DCT ringing artifacts
    Crop(6,0,-10,-0) # remove some of the black borders at the left and right, leave a 704x480 ITU frame
    TFM(d2v="D:\Downloads\title00_cut_track1_eng.d2v") # field matching -- remove comb artifacts
    TDecimate()  # remove the one extra frame of every 5 left by TFM, restores 24 fps film
    src=last
    
    ConvertToRGB().RGBAdjust(g=0.90, b=1.0).ConvertToYV12() # remove green cast
    Dehalo_alpha(rx=3.5, ry=1) # reduce oversharpening halos
    Spline36Resize(400, height) # reduce frame size, reduces dot crawl artifacts and other noise
    Stab() # remove small frame bounce
    Santiag() # a little antialiasing of edges
    RemoveDirtMC(25, false) # remove spots and a little noise
    Sharpen(0.3, 0.0) # Sharpen a little, only horizontally
    MergeChroma(last, aWarpSharp(depth=10).Sharpen(0.5)) # sharpen the chroma
    nnedi3_rpow2(2, cshift="Spline36Resize", fwidth=src.width, fheight=src.height) # high quality upscale back to 704x480
    Sharpen(0.3, 0.0) # sharpen a little more
    Many of the filters are from third parties and have to be downloaded and installed (usually just copying the dll to AviSynth's plugins folder). Most of the filters have lots of settings you can play with to fine tune them to your liking.

    Regarding the picture settings on the ES15 -- you might try them just to see how they effect the captures.

    Moving to a capture device that captures raw YUV video will get you a little more quality but the capture software may be more finicky. I don't keep track of what devices are available but I know Hauppuage has a few that can capture analog video as YUY2 (or other YUV 4:2:2 format). There's the EzCap from EzCap.tv. Beware of the cheap EasyCap (and variations on that name) that sell for $5 to $10. Most of those are crap.
    Quote Quote  
  23. Member
    Join Date
    Jul 2016
    Location
    USA
    Search Comp PM
    Originally Posted by jagabo View Post
    I did all the processing in AviSynth and encoded with the x264 command line encoder. The script I used was:

    Code:
    Mpeg2Source("title00_cut_track1_eng.d2v", CPU2="ooooxx", Info=3) # CPU settings = reduce DCT ringing artifacts
    Crop(6,0,-10,-0) # remove some of the black borders at the left and right, leave a 704x480 ITU frame
    TFM(d2v="D:\Downloads\title00_cut_track1_eng.d2v") # field matching -- remove comb artifacts
    TDecimate()  # remove the one extra frame of every 5 left by TFM, restores 24 fps film
    src=last
    
    ConvertToRGB().RGBAdjust(g=0.90, b=1.0).ConvertToYV12() # remove green cast
    Dehalo_alpha(rx=3.5, ry=1) # reduce oversharpening halos
    Spline36Resize(400, height) # reduce frame size, reduces dot crawl artifacts and other noise
    Stab() # remove small frame bounce
    Santiag() # a little antialiasing of edges
    RemoveDirtMC(25, false) # remove spots and a little noise
    Sharpen(0.3, 0.0) # Sharpen a little, only horizontally
    MergeChroma(last, aWarpSharp(depth=10).Sharpen(0.5)) # sharpen the chroma
    nnedi3_rpow2(2, cshift="Spline36Resize", fwidth=src.width, fheight=src.height) # high quality upscale back to 704x480
    Sharpen(0.3, 0.0) # sharpen a little more
    Many of the filters are from third parties and have to be downloaded and installed (usually just copying the dll to AviSynth's plugins folder). Most of the filters have lots of settings you can play with to fine tune them to your liking.
    It took a bit to hunt down on all the plugins, but I believe I got a working install of Avisynth and was able to re-run your script in VirtualDub. Output is attached

    Thanks so much for the pointers. I'm going to try to learn more about Avisynth and use that script as a base to do some clean up, with some tailoring for each CED. I'm also going to have to figure out how to get the audio back in there Is there an automated way to have the video move along with the audio? If I understood correction the Mpeg2Source call splits out the audio.

    Originally Posted by jagabo View Post
    Regarding the picture settings on the ES15 -- you might try them just to see how they effect the captures.

    Moving to a capture device that captures raw YUV video will get you a little more quality but the capture software may be more finicky. I don't keep track of what devices are available but I know Hauppuage has a few that can capture analog video as YUY2 (or other YUV 4:2:2 format). There's the EzCap from EzCap.tv. Beware of the cheap EasyCap (and variations on that name) that sell for $5 to $10. Most of those are crap.
    It's amazing that a cheap 10 year old DVD recorder is producing better output than a dedicated capture card. Hindsight being 20/20, I could have saved the cost of the WinTV, the MX-1, and potentially a better CED player (old one only has RF Modulated output). I learned quite a bit along the way though, so not all bad.

    I may try inverting the Thresholds on the DNR settings for the WinTV. I believe they're set to the highest tolerance if I understand them properly. Perhaps reversing those will exclude the noise reduction entirely.

    One thing I did note in looking through all the filter settings for the WinTV is that it's set for 8MB/s target rate, but can go up to 12MB/s. I believe that's beyond what the DVD recorder is capable of, and might provide some higher quality assuming the DNR can be turned off. I also take a look at a better capture card, but I went ahead and ordered a spindle of DVD+RWs.

    Thanks again for all the help! I'm sure I'll have more questions, but now I have something that I can work with and gives decent output.
    Quote Quote  
  24. Member
    Join Date
    Jul 2016
    Location
    USA
    Search Comp PM
    Whoops! Almost forgot the example avisynth output!
    Image Attached Files
    Quote Quote  
  25. Congrats on getting AviSynth and all the filters set up and working. Your video looks pretty good. One thing you can do is crop the video down a little more to get rid of all the black borders, then use the "mirror" option of Stab() to fill the borders with mirrored data rather than black, Stab(mirror=15). If you need the original frame size back you can use AddBorders(left, top, right, bottom).

    I usually mux the audio along with the video myself with MkvToolnix GUI. But it can be automated within a batch file. If you're using VirtualDub you can add the AC3 audio exported by DgIndex with Audio -> Audio From Another File.
    Quote Quote  
  26. Originally Posted by jagabo View Post
    Congrats on getting AviSynth and all the filters set up and working.
    I agree. Congratulations! I've been following this thread for a few days now. This must be a record time for getting up and running with a fairly complex AviSynth script. It just goes to show what one can accomplish with the help of a 'mentor' when willing to do the homework and put in the hours to learn a better way. And once you see what can be done, there's no going back. Don't hesitate to return if you get stuck on something in the future. Just wait until you try and get going with some really complex filters such as QTGMC or MCTemporalDenoise. The good filters often have their own pages on the AviSynth website, such as the links I provided for those two filters just mentioned.
    Quote Quote  
  27. Member
    Join Date
    Jul 2016
    Location
    USA
    Search Comp PM
    Originally Posted by jagabo View Post
    Congrats on getting AviSynth and all the filters set up and working. Your video looks pretty good.
    Thanks! They definitely don't make it easy I wouldn't have been able to do it without having your script as a guide for what to look for.

    Originally Posted by jagabo View Post
    If you're using VirtualDub you can add the AC3 audio exported by DgIndex with Audio -> Audio From Another File.
    I was able to get the audio muxed in with MkvToolnix. That worked perfectly. Do you recommend another tool over VirtualDub?

    I do have one quick question. In comparing the output from the Avisynth script to the original, I noticed there seems to be a size change. It appears "squashed" in the vertical direction though I'm not sure why as they both appear to have 480 horizontal display lines.



    Any idea why this might be? I've verified that VLC is displaying them in their original aspect ratio at a 1:1 zoom with no effects applied.
    Quote Quote  
  28. Originally Posted by CED View Post
    I do have one quick question. In comparing the output from the Avisynth script to the original, I noticed there seems to be a size change. It appears "squashed" in the vertical direction though I'm not sure why as they both appear to have 480 horizontal display lines.
    The original MPEG stream has aspect ratio flags that tell the player the video should be played with a 4:3 aspect ratio. That information is lost in AviSynth processing. All you have to do is tell the encoder the proper display aspect ratio (DAR, 4:3) or sampling aspect ratio (SAR, 10:11).
    Quote Quote  
  29. The picture on the left is from the DVD which, while it stores the video at 720x480, resizes it at playback to a 1.33:1 (4:3) ratio such as 640x480. AviSynth has nothing to do with aspect ratios and shows you the video without any aspect resizing. As a result, 4:3 DVDs when displayed without the resize show everything as slightly wider or fatter. When you encode for your final format, the encoder you use should be told what aspect ratio you want for playback. If it's to be for another DVD, you set your encoder for a DAR of 4:3.

    Or you could resize it yourself to 640x480 or whatever (not for DVD, though).

    Edit: Oh, too late.
    Quote Quote  
  30. Member
    Join Date
    Jul 2016
    Location
    USA
    Search Comp PM
    Originally Posted by manono View Post
    Originally Posted by jagabo View Post
    Congrats on getting AviSynth and all the filters set up and working.
    I agree. Congratulations! I've been following this thread for a few days now. This must be a record time for getting up and running with a fairly complex AviSynth script. It just goes to show what one can accomplish with the help of a 'mentor' when willing to do the homework and put in the hours to learn a better way. And once you see what can be done, there's no going back. Don't hesitate to return if you get stuck on something in the future. Just wait until you try and get going with some really complex filters such as QTGMC or MCTemporalDenoise. The good filters often have their own pages on the AviSynth website, such as the links I provided for those two filters just mentioned.
    This site is definitely a great resource! I've learned a lot over the last week. I definitely appreciate jagabo taking the time to walk me through a lot of this and answer my questions. I'm glad there are people willing to help

    I'm sure I'll have a lot of questions as I start digging into AviSynth some more. At this point I've only managed to replicate an example. I'll play with it a bit and see what I can get it to do My spindle of DVD+RWs arrived today so I'll be able to start pulling in more CEDs once I get a process nailed down.
    Quote Quote  



Similar Threads

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