Quote:
Originally Posted by Oakleaf
I was just giving the specs that said "up to 720p", not that I'm going to use that resolution. Lordsmurf was saying I needed to use a capture device other than my pass-through video camera to create an uncompressed file rather than the DV-AVI file. I'm having a hard time finding the capture hardware needed with S-Video and stereo audio inputs that will work with Windows 10, Pro 64 bit and produce an uncompressed avi file. I'd rather not have to install a PCI-e card if the USB capture device would work as well. Does anybody have experience with one of good quality?
|
Umm, not trying to be a meany here, so don't get the wrong idea. The above paragraph and previous posts persuade me that you're reading from some other forum than this one. So I'm compelled to revise a shorter post into a much longer one in order to repeat material that appears in hundreds if digitalfaq posts and dozens of guides.
There is no post or article at digitalfaq that advises capturing VHS to uncompressed AVI. We advise against it. It's kinda tough on CPU's and causes capture problems. You've heard of lossless high-speed compressors, right? If not, read on. If yes, feel free to read on anyway.
Quote:
Originally Posted by JPMedia
It is possible to run interlaced analog video through a deinterlacer. This step is highly recommended if you plan to create mp4 files for online streaming. The best software for this purpose is called QTGMC. However, to use this software, you need to learn how to use another software called AviSynth. AviSynth is a programming language for video editing. Basically, one will program QTGMC using AviSynth. Then, once an AviSynth code (called a script) has been written to instruct QTGMC, load and run your script with VirtualDub.
This process will result in a deinterlaced AVI file. You then need to use an encoder to turn your deinterlaced AVI into an mp4. |
Mmm, well....those are good tips, but I don't think Avisynth or QTGMC will be Oakleaf's cup of tea. An additional step is required, namely to resize video to square-pixel formats since anamorphic video can't be posted and won't work in internet players. From what I'm reading here, most newbies want simple icons to click and instant results with all the decisions made for them.
@Oakleaf, I've been using Adobe software and visiting their forums since 1998. They're great resources for learning advanced pro-level color correction, photo retouching, applying special effects and blend techniques, and fairly decent encoding. But when it comes to analog transfer and restoration, you won't get there with Adobe. They have no concept of restoration and their software can't be used for it. PP is inferior at resizing, deinterlacing, inverse telecine, colorspace changes, and capture. Lossless video processing is as foreign to Adobe users as it is to you, and something like QTGMC or Avisynth is out of their restricted field of view. I hung around their forums and user groups for many years, but when it came to the hard grunt of real video work from analog source, I saw more nonsense than anything else. Their approach is too destructive and ineffective. the opposite of techniques in advanced video tech forums. I'm a great fan of Adobe for cut-and-join editing, timeline, color, special effects, and I love AfterEffects. But analog transfer? Cleanup? Restoration? Adobe doesn't even make it to the ballpark.
Quote:
Originally Posted by Oakleaf
Actually I'm underthinking about the 32/64 bits, because I really don't understand anything about that subject!
|
We get it that you don't understand it. There seems to be much about basic video processing that you haven't tackled yet. But that can be fixed.
Most of the work that you do with video for capture and post-processing with restoration apps wlll use 32-bit software. For 32-bit apps you install 32-bit plugins and codecs. You can use 64-bit software if you want, in which case you install 64-bit codecs, plugins and attachments to match 64-bit software. You can install and use both 32-bit and 64-bit software and codecs at the same time in 64-bit systems. Windows keeps track of what you have installed and what you don't. Is there an advantage to using 32-bit video cleanup apps? Yes, indeed: there are many more features, filters, and free apps available in 32-bit than in 64-bit. But you can always use both if you want. So, what's to think about?
Do you know the main differences between 32-bit and 64-bit? Have you looked it up (
https://www.geeksforgeeks.org/differ...ing-systems/)?
Quote:
Originally Posted by Oakleaf
I started out trying to use Cyberlink PowerDirector to capture...
|
Okay, I'm sorry, this is getting a little hysterical. Cyberlink? I don't know where you got this advice about using Cyberlink, but you didn't get it here. Sorry to see that you were sidetracked by their advertising.
Quote:
Originally Posted by Oakleaf
You seem to be saying that using the camera for the conversion rather than a capture card is losing 50% of the VHS color.
|
We don't "seem" to be saying it. That's exactly what happens. VHS tape has twice the color resolution of consumer DV. VHS is 4:2:2 YPbPr uncompressed analog color waveforms and is normally captured as lossless 4:2:2 YUY2. Consumer DV is lossy compressed digital pixels, either 4:2:0 or 4:1:1 YUV with half the color resolution of VHS. Look it up yourself. There are also differences in the way DV interprets analog color storage systems. In the post-processing software that we recommend here, translations between colorspaces and color systems for processing are done with far more precision than Adobe can muster, which is another reason why we don't recommend Adobe for restoration or capture work.
Quote:
Originally Posted by Oakleaf
Hunt over at Adobe says that the compression of the DV-AVI Type 2, MS DV codec has only a "tiny quality loss, but it can usually not be detected, even with a side-by-side comparison."
|
What does that have to do with the quality of processing? Side by side comparison with what? Did Hunt offer examples? What can you do with DV? it disappeared years ago with the demise of Pentium III and the rise of DVD and, later, h.264. It's PC-only playback but DV isn't furnished by any OS. It's not supported by external players or by the internet. So what can you do with it? All you can do is send it through yet another stage of cumulatively lossy re-encoding into a format that modern devices can recognize.
I disagree with Mr. Hunt. DV has its own signature "look". Anybody can see DV compression noise and its missing inner detail, the subtle loss of fine textures, and the levels blowout, even when the source itself is DV. DV has a very distinctive "plastic", mildly eviscerated appearance and noise profile, and it has its own peculiar form of mosquito noise, especially during motion. Nothing else looks quite like it. Take a look at the image when DV is re-encoded to DVD or h.264, and it gets worse.
Quote:
Originally Posted by Oakleaf
Back then, I was happy with the DVDs I made which looked about as good as the pitiful old VHS tapes.
|
No, I can't agree that the results "looked about as good" as the tapes. It's not possible. For one thing, VHS noise gets translated into added digital compression artifacts. Digital artifacts don't exist in analog source, which has no digits, no pixels, and no compression. You also had unmanaged color resolution loss and no input signal level control, so the color and gamma configuration couldn't possibly look like the original.
VHS or other analog tape captured to any kind of digital media in a computer, no matter how it's done, will look nothing like DVD. it will look like tape, but at least slightly worse, with all of tape's simmering grain, dropouts, spots, ripples, chroma discoloration, head-switching noise, uneven side borders, possibly sloppy interace that has to be smoothed out, illegal signal levels, edge halos, DCT ringing, and all of tape's other pimples and scars. A good VCR can help minimize that, but it'll still be visible.
Now, would you like to add more grunge and dirt to the results? You can do that by taking all those imperfections and capturing them as lossy DV, which adds mosquito noise and other digital compression artifacts that don't exist in analog sources, along with poor input level controls resulted in blown-out highlights and crushed shadow detail. Add buzzy, sawtooth edges from the field phase differences and DV compression algorithms, and add color translation problems between VHS and DV color storage systems (DV colors will look more raw and "baked"). Then factor in 50% chroma loss, and remove a visible amount of fine detail that gets discarded by lossy DV compression. I assume that you know what lossy compression means. It means "you lose". The data that is discarded is gone forever. it doesn't come back. Do you want to filter that grunge and noise? Think again -- it's a mess to clean up (you can't do it with Adobe anyway), and you'll need special filters that will take time and cost more detail. You've already lost a lot of data and detail anyway, and you'll lose more when you go into your final lossy encodes for your output formats. For a final format you'll use another lossy encode whether you want to or not.
If you have no intention of cleaning up VHS or DV defects (and there are plenty of them, even with a good VCR), you're wasting your time capturing to a computer. Just find a DVD recorder and a smart-rendering MPEG editor and be done with it. The idea behind lossless capture is to allow lossless cleanup, lossless edits, lossless restoration and repair, and final re-encoding without intermediate damage buildup.
Quote:
Originally Posted by Oakleaf
I’ve read there are now some good USB capture devices with 4:2:2 video sampling at up to 720p.
|
I don't understand what you mean by 4:2:2 video up to 720p. The cards will transmit 4:2:2 to any frame size. Why 720p? You mean you really want small low-resolution frames blown up into big low-resolution fuzzy ones? It's is a dead giveaway that shouts "newbie". YouTube is flooded with examples. Don't fall for that. High defintion is based on high resolution, not on big frames. Anyway, your TV and media players will upscale better than you can, so why are you paying for their superior performance and then feeding them lower quality by working so hard to do it yourself? Software deinterlacing is a destructive process, not an improvement. Like DVD and Sd-BluRay, VHS is interlaced and plays that way, and so is HDTV. Capture cards that deinterlace and upscale during capture are notoriously horrible. If you want to deinterlace for the web or streaming services, use QTGMC. Don't use Adobe for it.
Quote:
Originally Posted by Oakleaf
If a capture card is not too expensive, I’d be willing to do my own comparison before going to the time and trouble of converting 40-50 old tapes. The only available slot I have is a PCI-e. Please suggest some good capture cards, and they would have to have an S-Video port and audio ports, right?
|
What will be the basis for your comparison?
There are three USB capture devices that this site and others have been recommending for years as optimized for analog sources and usually run under $100. They are: lordsmurf's favorite, ATI's 600 USB dongle. They're hard to find, but you can find them sold in our marketplace. A little less difficult to find is the Hauppauge USB-Live2. A choice that's competitive with those two and easier to find, with updated drivers for XP to W10, is the Diamond Multimedia VC500 USB. It's been a favorite for years worldwide. It costs less than $50, brand new (it used to cost more, when I first bought it years ago and added it to my All in Wonder capture setup). Ignore its generic free software and use free
VirtualDub or AmarecTV for capture. The USB devices will accept s-video and composite, and RCA audio inputs.
VHS is captured to 720x480 NTSC (or 720x576 PAL), interlaced, with appropriate input level controls (we show you how to do that), to lossless 4:2:2 YUY2 AVI media using
Huffyuv or Lagarith lossless compression.
Huffyuv is faster for capture and less CPU intensive. Lagarith makes slightly smaller working files and also works with YV12, YUY2, and RGB, so it's usually used for lossless intermediate working files. Adobe PP also has problems with some popular lossless codecs, including huffYUV, which is inexcusable for a product with "Pro" in its moniker. Intermediate working files are generally not saved after a project ends. The initial capture is either archived as-is or is saved as high-bitrate compression with MPEG or h.264.
If you want advice on your captures or processing, post short unaltered sample edits directly to this forum. Please don't post samples on YouTube -- no one here will deal with YouTube's ugly processing, which won't help you. If you don't know how to make and post video samples here, just ask.
