OSX video workflow for a Mac?
 

Hi, i'm looking to digitize a good amount of action sports (rollerblading/skateboarding) VHS tapes in the highest quality i can achieve.

My set up includes a JVC SR-V10U --> Canopus ADVC 110 --> 2014 iMac --> Final Cut Pro X

I've found that i'm at a disadvantage using a Mac so i'm wondering what is the best way to 'enhance' a video after i've captured it. I took on this project as a noob, and am still figuring out filters, deinterlacing, encoding, etc. PC users have AviSynth, and VirtualDub but i haven't found many options for OSX users.

Is there a better option than FCPx to capture the footage? Can I 'enhance' the video using FCP?

I found an article explaining how to use AviSynth and Virutaldub with WINE (PlayonMac). Are there any problems with this approach, would this be my best best?



Thanks.

If you want "highest quality", the Canopus is your first obstacle.

Many Mac fans have not yet figured out that Mac no longer intends to support many graphics endeavors, including video work. You can get a few shots at some pretty good software, but it will cost plenty for Mac.

What do you mean by enhance? If you mean cut/join, play with contrast and colors, yes up to a point. If you mean repair frames, get rid of dropouts, dot crawl, rainbows, telecine, deinterlace/reinterlace, aliasing, then you're asking an "editor" app to do the work of more specialized apps like Avisynth and VirtualDub.

Some Windows apps work with WINE, but many don't. At the outset, it's likely you can get better quality captures with a better capture device and VirtualDub capture, which is still the affordable "best way" to capture VHS to a PC. It's old-fashioned, but it was optimized at the time for that sort of analog source.

What sanlyn says is true. Most Mac users head-in-sand (fingers-in-ears) shouting that Mac rules, and MS/Windows drools. They assume the advice from from "Mac haters" and move on tho another forum. But I own OS X Macs , and have been a Mac user since the Apple IIe days.

Sadly, it's the wrong tool. Apple shunned video during the crucial years, the early/mid 1990s until 2001, when DVD/MPEG was being developed. As such, it has no tools. What it did do in 2001 was very limited. It wasn't until around 2007 that serious tools existed, and by then it was a decade too late. Not just that, but it pretty much gave BD-R the finger, which cemented its fate.

Mac is only good for a DV workflow. But even then, not really. It's extremely limited.

As a Mac user, I wish it was different. But it's not.

I've also been a Windows user since the 80s, and am platform agnostic. To me, these are just tools. And for video, Windows is the right tool for the job.

You already know some of this, it seems, but I just wanted to confirm it for you.

An important question was just asked, and I'll echo it: What do you mean by enhance? What needs to be done? What errors do you see?

In a nutshell, yes, get a Windows XP system with an AGP ATI All-In-Wonder card, and we can tell you what software is needed. Bonus: Windows has many free tools -- tools that ONLY exist for free, and are often best for the task. Although, yes, like Mac, sometimes the best (or only, or needed) tool is commercial. We'll cross that bridge when we come to it.

Right now i have no option other than my Mac, so i would like to figure out a workflow with my current set up.

Sanlyn, you mentioned a better capture card. Would that be the blackmagic intensity shuttle?

also, i understand avisynth and virtualdub are specialized apps but aren't there any that work with osx?

When i mentioned 'enhance' i was trying to figure out of FCP had similar tools to as/vd, but sanlyn answered that perfectly.

I'm still new with this terminology so i can't really name exactly what needs fixed. But i'm able to compare with a very nice vhs rip i was given. When i use my equpiment to capture the video in FCPx it's not as sharp than the rip i was given. The guy who sent me the rip said he deinterlaced, used scripts, did a MeGUI encode then converted to mkv. I assume this is how he got the quality?

Sorry, no. Many posts in many forums advising that VHS->BlackMagic is Disaster City. I'm surprised you haven't run across those warnings but, then, the method has been around for years but I just noticed those cautions maybe a year or two ago.

Not that I ever heard about. There's heavy-duty industrial stuff for modified Macs I hear are used at Disney and other super shops, but I'm informed that this is all custom stuff that you and I can't afford.

I completely understand your terminology frustration. Been there. One tip at the outset: there is no such thing as a VHS "rip". I know you see it used all the time by the ignoracii, but using "rip" in connection with analog capture is like using the word "impact" to describe everything from meteor collisions to jock itch. You might as well say you're going to the kitchen to "rip" a cup of coffee. VHS and similar analog sources are captured or recorded, not ripped. DVD and BluRay disc video can be ripped, i.e., decrypted (if need be) and transferred as a 1-to-1 non-reencoded copy onto a computer.

I can't answer for the party you mention, but MEGUI uses Avisynth. Hopefully he knows better than to use "rip" in the same sentence with VHS. Just as hopefully the other party knows better than to deinterlace VHS movies (which really might have been telecined, which should never be deinterlaced) and probably didn't re-interlace or restore pulldown for the encode. And hopefully the other party also mentioned that advanced hobbyists get a whole lot more involved with software than using MeGUI. Apparently, too, your mkv is progressive for PC display only, which is unnecessary nowadays with deinterlacing media players; the final mkv format would be neither DVD nor BluRay compliant, and if played via memory stick or external drive you'd find many set top players that can't handle mkv. But those are other topics. In summary, though, he apparently used advanced techniques and lossless media to clean and tweak, using filters designed for the specific problems associated with VHS sources, before re-encoding. I should also add that MeGUI uses better mpeg and h264 encoders than you will usually find with a lot of Mac video software unless you're willing to invest in pro gear.

-- merged --

I forgot to mention: there are a few basic editors and encoders around that run on Mac and Windows that i've seen recommended, although some of their feature set might not be 100% available on Macs. But they seem to be popular anyway. There's the 64-bit intel version of Handbrake (http://handbrake.fr/downloads.php) and the multi-OS version of AviDemux (http://avidemux.sourceforge.net/). Here's a site that lists a ton of software in many categories with indicators for what works on Macs. http://www.videohelp.com/tools. Happy hunting, as many of these are freebies with either very short or rather long learning curves. And of course some of it is either paid or highly paid or outrageous.

Thank you for the thorough response sanlyn.

When i explained how my party recorded the VHS, I completely butchered his process which probably set off some nerves to someone who knows what they are talking about lol. He actually reordered it to a dvd first. I don't mean to put him on the spot, but when i asked how he achieved such quality, he sent me this.

Do the tools he mentions not improve the quality? I am still very confused how I can improve my recording.

Was this VHS source, intended for TV viewing? If so, he raped the quality by deinterlacing it. :screwy:

Read the reviews for this card on this forum. Search the forum for it, Several members here have experience with it, and have posted on this in the past.

Well, butchered description aside, that method also sets off someone that prefers a quality workflow. Because that's not it. Dazzle, unknown-quality VCR, iMovie -- YUCK! The source he used was further degraded by the tools (hardware/software) used. They're pitiful. Those tools are infamous for making DVDs that look worse than the tape. It should be the other way around -- digital/DVD versions can look better than a tape!

Post-capture Avisynth likely compensated for this some, but some errors cannot be fixed in software alone. The capture card and VCR are extremely important here.

Not entirely, no. What he's describing is a method to rip an already-fubar (badly captured) video from a DVD, filter it, then try to put it back. Granted, sometimes that's all that can be done, as earlier source is gone. You're at the mercy of somebody elses' butcher job.

But from the sound of things, you're just going to toss his work, and start over with the tapes. And that's the ideal solution here.

So your workflow would be this:

VCR (JVC S-VHS SR-V10U = good)
> external TBC (the Canopus may or may not remove all errors, depending on the source; DataVideo TBC-1000 suggested)
> Canopus ADVC 100 series (DV workflow not ideal, but not many solutions exist for Mac)

Then the "raw" (DV) digital file is on the computer.
> Avisynth to process, via AvsPmod GUI **
> VirtualDub to accept the AVS script, and add further filters as needed, output to Huffyuv AVI

Here's where your workflow gets muddy...
You can:
(A) edit the footage in FCP/Premiere/etc
(B) then encode the VirtualDub-output AVI to interlaced disc format (MPEG-2), then author
(C) add QTGMC("slow") deinterlacing into the Avisynth script for streaming progressive output (MKV, MP4, Youtube, etc)

You've never really specified the output and post-capture steps.

I suggest a full archive copy on disc. And if needed, as second encode for the streaming version. That's what quality-conscious people do. We save the best for disc use, and another version for the streaming "sharing" version that is uploaded elsewhere. Why? Even the best deinterlacer leaves artifacts.

If relying on free encoders, use Avidemux for both MPEG-2 and MP4/MKV (H.264).

** Note that I'll be posting some Avisynth samples later this month and in coming months, as I finally get all my Avisynth restoration guides online. This has taken me several months, but I think I finally have most of my Avisynth knowledge committed to an organized set of documents! (And I've never seen anything like this anywhere else online.)

lordsmurf had some good recommendations, so I have little to add to the workflow except that the Mac will be a big set of limitations. Myself, I'm far from a devotee of AvspMod. Many use it. If you do, you should learn first to run Avisynth on its own and use VirtualDub to monitor the output. Otherwise newcomers to AvsPmod can have a rough time trying to figure out what's going on.

lordsmurf emphasized the quality of the playback machine. Your JVC is probably pretty good, but it's not ideal if you have 6-hour tapes. Most of us have more than one VCR anyway, as some tapes just refuse to track on one machine but will run OK on another. Recording VHS directly to DVD is hardly the best way. It's frought with problems unless the source tape is in fairly pristine shape on a good player. Capture to DV is a step up (but you'll have some color resolution loss, and tape artifacts will be somewhat more complicated to clean up). Capturing with a decent card to lossless AVI with huffyuv or Lagarith lossless compression is better yet. You would use one of those two compressors for all of your intermediate work files. Encoding to MPEG or mkv or whatever would be the very last step.

Many assume that just because a VHS tape is recorded to DVD, making it "digital" automatically cleans up everything. It doesn't. Not at all. If the tape has defects or problems, as most of them do, it just makes the workflow more difficult. I don't really want to put your friend on the spot myself, either -- after all, he's trying to be helpful. But I think lordsmurf's description of the results in probably pretty accurate. To explain that a common script and unvarying process is used to clean up everything because "everything looks alike", well....it's not true. They might look similar, but not alike. There's no way that any two VHS tapes will "look alike".

But this, after all, is how everyone learns this stuff. You can read about it from now 'til the cows come home, but until you actually try it the whole thing will just be a great big mystery.

Much of this learning process is enhanced by posting samples of videos and having members check them out. Yeah, they always find something wrong no matter how "good" it is. But it helps to identify problems, understand their causes, and get a handle on what can be done about them. You can also learn a ton by looking over restoration and repair threads that explain what was done to improve the results.

As lordsmurf says, the first step is learning to get a workable capture. All you can do there is do the best you can. Then go from there.

Please do not buy the BlackMagic. I have had nothing but problems and have gone through three, each with its own quirks.

You have two primary options worth considering. Before I get to those, you either need a copy of Windows via Bootcamp or Parallels. Mac software is limited for VHS filtering and correction, and while I generally create my final files via FCP X or Compressor because I like the encoders, all work should generally be done in Windows one way or another.

Hardware-wise, these I have found to be the only one really worth considering that you can buy new for MAC ONLY:

Elgato Game Capture HD - this is not the Video Capture Device sold for $80. This device is intended for HD game capture, but you can capture VHS with it using composite with its included adapter or S-Video with the $10 additional retro gaming adapter (which I recommend). At its highest bitrate, its quality is very good. I will not argue that it is better than a 4:2:2 uncompressed capture, but for what is Mac compatible this would be my first suggestion. I have done plenty of correction with captured MP4s, in VirtualDub for Windows-only filters and FCP X color correction/noise reduction. They are better for clean sources or at least brighter sources, but its software lets you adjust color/brightness. I picked one up at Best Buy for $140 and once you are done capturing your tapes, it will not be made obsolete for a while if you use it for other purposes. Best Buy has a good return policy, so if you don't like it you can bring it back.

If you end up running Windows, you have other devices you can consider. I would recommend the VC500 as the second option because its one of the few capture devices that isn't too expensive, works well with VirtualDub, and you can buy new.

Beyond that, you run into driver issues and finding good equipment. I own both of these and the newest Mac Retina and they both work well.

lordsmurf, when you say the 'raw dv file', how do i achieve that? I am capturing with FCPx, so would i save it from there in the DV format?

As far as the output, i'm looking for a streaming / download concious file at the end. For the master file or maybe raw file would be a better term, I would rather just save it to an external harddrive. I don't need or really want to convert to DVD right away.

Post capture steps, i'm still trying to figure out the best way to capture the video onto my computer which i assume is FCPx. Then how to actually use AviSynth/VirtualDub to correct the video. I obviously have research to do on the different tools.

here are some screenshots, per sanlyn's suggestion.

Left: My raw capture in FCPx. Right: 3rd party source

On my capture during the title screens I noticed horizontal lines in the lettering when it faded into the next title screen. Also the words weren't perfectly still.


http://www.digitalfaq.com/forum/imag...Pt0Zppng-1.jpg

http://www.digitalfaq.com/forum/imag...0hK49png-1.jpg

http://www.digitalfaq.com/forum/imag...NjUMvpng-1.jpg

If you keep all your media in your project files, when you import from the Canopus, it will save an original DV copy inside the project file. If you right click it and click show package contents, you can find it there.

Your left side is interlaced, the right is de-interlaced.

I don't think that's correct. I'd want to see a sample first. From the look of it, the LEFT side has obvious deinterlace "jaggies" from a basic drop-frame method. And the RIGHT is simply showing an interlaced motionless area.

To be honest, they both look deinterlaced. I think a preview deinterlacer is involved here. Meaning screen caps will be useless.

AviSynth does not run on Mac at all. VirtualDub runs in Wine, but with caveats and errors.

Parallels is what you want -- not Bootcamp.

Put Parallels into "Coherence" mode. :wink2:

There have been times when I ran a full Mac workflow, from capture to encode to output. I've not captured on one in years, but I still frequently use it for all post-capture work.

The still shots are a start, at least. But I think you can see how difficult it is to work from still captures. For instance you say the title letters keep moving, but as you can see in the screen cap they don't move at all (!). Meaning, what does "move" mean? Rapid horizontal jitter of the letters, the entire image hopping up/down, etc., etc. ? There's a big difference between a single frame and frames in motion.

Something else to keep in mind: you can't analyze motion problems from a still frame, but you can get a ton of info by watching frames move in sequence, especially at 1 frame every 1 or 2 seconds, or slower. You can also enlarge frames for viewing -- you can do that with images, too, but how much has changed between the original, unprocessed frames and the jpg compression in the images posted? Each image itself is second-generation (at least) and is lossy re-compresseed.

All the images look deinterlaced to me.

premiumcapture, i found a .mov file for my raw capture, is that correct?

Does AviSynth run with Parallels? I'm going to download the free trial instead of using the Wine method. Although, the article i mentioned earlier stated Wine could run AviSynth and VirtualDub. I guess it's very buggy as lordsmurf stated?

I will try to be less broad with my description of the 'moving letters'. I've been rewatching it and it seems like the color isn't 'solid' or 'still' inside the lettering. it's sort of fuzzy inside, but a very minimal amount. Also, here is a better example of the horizontal lines i was referring to during the fade out.

http://www.digitalfaq.com/forum/imag.../bagY6l1-1.png

Those lines and other noise come from improper interlace/resize, and the denoising in the other images looks rather destructive without accomplishing all that much (but as I said, it's really tough to tell from resized jpg stills).

BTW, that "VG3" image is 732x401. Did you resize it from the original? How was the image made?

I wanted to bring that statement to premiumcapture's attention, as he appears to be more familiar with Macs. I don't know what to make of it, exactly. ".mov" is a container, not a format. The file ending itself doesn't tell us what codec was used.

AviSynth runs in Windows, Parallels just lets you run Windows while you run your Mac. Very useful all around, though I would only recommend XP - 7.

The .mov should be roughly 14 GB/hr, the bitrate of DV. FCP X just rewraps everything, but that should be it as long as its as big as DV. I am not totally familiar with Wine, but you have to keep in mind that filters are coming from all over the place doing all sorts of things in different ways. Wine might let you run VirtualDub, but for a full workflow I would stick with Windows unless impossible.

For the stills, i just did a screenshot (cmd-shift-3), cropped them, and uploaded to imgur. Is there a better way? Usually the easiest is not the best..

I believe that is the raw DV as I only have 2 minutes captured and it is 465 MB.

Sadly, I have a copy of Windows 8. I know it's very different from xp and 7, but will it affect me? I will using running parallels strictly for this project.

Re-upload but don't crop, it can distort the image and give the impression that there are other problems. The easiest way is to open the video and take a screenshot and upload it directly to the forum. The file will be in PNG which is better than JPEG.

That sounds like DV to me, but you can verify in VLC by clicking Codec Details under View. Those are the files that you want to work with. They have the best quality you will get with your current equipment.

Windows 8 should be fine, I don't thing you will run into too many issues, but I would stick to Lagarith over Huffyuv for editing, only because getting the drivers to work in Windows 8 will be a PITA.

did as you described premium. i did not resize the quicktime window, i wasn't sure if it would effect the results.

Also, capture it seems you are very familiar with what I am trying to do with OSX. Do you have a particular workflow, or maybe can you point me somewhere that can give me more information on this subject?

Download and install VLC for the Mac, Quicktime is going to work against what you are trying to do and you don't want to be taking screen grabs. VLC will take the image from the video itself. The window will affect the final image, I meant don't crop any black bars :P

Everything I know has come from hours of looking at forum posts here and videohelp. I will get back to you on that later today.

the correct screenshots



Attachment 3967

Attachment 3968

Attachment 3969

Fortunately, one only has to learn most of this stuff once, LOL!

As far as I know, Avisynth 32-bit and VirtualDub's newest 32-bit edition work in Windows 8. Others might have more info about that. There are 64-bit versions of those but, wouldn't you know it, there are barely a handful of 64-bit filters and all the best of 300+ filters are 32-bit. As for using them in Parallel, folks say they do it. I haven't used a Mac since 1998, so others will have to chime in.

ED: new images, post #21:
Let's see if I can understand something. Thanks to premiumcapture who sez the caps are DV wrapped in .mov container. No problem there, although DV does take a toll in some ways with NTSC VHS source. But if that's what we have, that's what we have.

The images are 720x528 (were they captured at that size??) and they look like improper deinterlace. Are these all from the same tape? The top letters look cleaner, but only a short sample of moving video can tell us more. There's plenty of the usual VHS noise in the shots, and looks as if the JVC is softening the noise. As for that three-letter logo -- these logos usually look pretty bad on most tapes anyway, and even on otherwise decent tapes. It can be cleaned up, but you'd need Avisynth to do it. One would need temporal filtering to do it or to demonstrate how, and something different for the clumpy fade-outs (which would actually look better if not deinterlaced, or if deinterlaced properly). Unfortunately temporal filters don't work on a single frame.

These shots came from the same tape. For those last screenshots, i opened in VLC (didn't resize window) and used the vlc screenshot shortcut.

i uploaded the 2 minute capture here.. https://mega.co.nz/#!gBcwALoA!BRiJmw...VNez7mvakPlv8w

For a capture card, would i be able to use something like an ATI 600 USB now that I am running Parallels?

Well, that's a prreeettttyyy big sample, LOL! Thanks for that. Definitely VHS, too, no mistake. Some of it actually looks OK, but tell us something about that tape. This is a retail job, or something recorded off tv, or what?

Give me a little time to look it over:

Meanwhile there's a dandy free tool for Windows called MediaInfo. Here's what it says about your sample:

What did you want to do with these? DVDs? Files? You may be better off just capturing and keeping the DV unfiltered depending on what your goals are. The best way to figure out a workflow is figuring out what you're starting with and knowing when you will be finished. After that everything else figures itself out.

I am working with retail tapes.

edit for premiumcapture..

My main goal is to end up with a final copy ideal for youtube and a copy for downloads. I assume this can be accomplished with one file, but two different files would be ok too.

And I guess a master or raw file to save on an external hard drive that would allow me to put on DVD later on.

Your final export will take place in FCP X or Compressor to YouTube. Your final file will end up being in H.264, so you may still want to consider getting the Elgato I mentioned.

With H.264, the codec is built on efficiency for clean digital video. More noise means a higher bitrate is necessary to make it look good. You will definitely need noise reduction along the way to make it look decent on YouTube.

How many tapes do you have in all. Did they ever get a digital release?

No digital release for these and everything on youtube / vimeo is horrible quality. I'm slowly building an archive for this niche, and am venturing into VHS.

I have about 30 tapes right now and plan on doing more in the future.

I'm confused because uploading straight to Youtube defeats everything i'm trying to do right now. If i upload to YT from FCPx, then i'll be stuck with this same poor quality, right? The 3rd party source i have is noticeably better, and i don't see a reason to spend all this time if i'm stuck with what it looks like right now.

also, i recently purchased the canopus. i don't plan to upgrade the capture equipment unless it's total lossless.

If you do a straight upload to youtube it kills it. Working in FCP X or Compressor will optimize for YouTube. With correct settings, it will not convert again. Many tape copies on youtube were not sourced from lossless copies, and video without noise reduction will waste the small bit rate you are allotted. For sharing on youtube, that step is the most critical.

I am not sure how much filtering (if any) this person performed, but he uploaded VHS recorded on a VCR with an ADVC 300. I think, for what it is and what he used, it came out ok.

https://www.youtube.com/watch?v=xZad1oXQ_iU

Well now that I have parallels, I'm not limited to FCPx anymore. I captured into VirtualDub and was playing around with the filter packs provided from this site. I obviously have no idea what i'm doing right now, but I really could't produce any real improvements.

With the screens, and samples i've provided, is it possible to get to quality that my 3rd party achieved?

Also, my capture into virtualdub was 1 gig for about 1 minute?!

The FCP capture is fine, the VirtualDub capture without any codecs should run maybe 70GB per hour. Change the color depth to 4:1:1 for DV. You might save yourself some trouble by importing through FCP and dragging them into VirtualDub, I dont know enough about DV to point you one way or another.

The most useful filters are not packaged with VirtualDub by default.

I need to see a 250mb max sample clip.
Either Dropbox or FTP, not Mega. See also http://www.digitalfaq.com/forum/news...mbers-how.html

Uncompress YUY2 = ~75gb/hour
Lagarith lossless = ~40-50gb/hour
Huffyuv lossless = ~35-45gb/hour
DV = 13gb/hour

Use Huffyuv or DV when using Mac. You'll also need Perian on the Mac, for Huffyuv access.

Here is a 100mb sample via dropbox. I captured into FCPx and uploaded the raw DV.

https://www.dropbox.com/s/e7o9gmvt3pod3f8/VGsamp.mov


Also, is it possible to run something similar to a ATI 600 USB using parallels?

That's not a good sample to filter with. It's too quick, scenes to brief. It's easy to overdo and underdo filters.
And I've never been clear on this -- does this need to be interlaced (view on TV) or deinterlaced/progressive (view online).

No, hardware is OS native, so no ATI 600 card in Parallels. You could use Bootcamp, however, maybe.

The Diamond VC500 (PC edition) works with VirtualDub in Parallels. I don't know how it stacks up compared to the ATI but its like $40 and I didn't run into any compatibility issues.

These will all be for online viewing.

Here is another example from that same tape. I started to get some audio buzz and dropping in the beginning, so is it possible the tape is going bad?

https://www.dropbox.com/s/8sjih5q9v9vt8gu/vgsamp2.mov

and here is a capture from another (but similiar) tape.

https://www.dropbox.com/s/ap1wfiifx8qx0fb/VG13samp.mov

you need a LTBC, the top of your video is wobbly

vgsamp2.mov -- this an an UGLY deinterlaced video, full of artifacts.
VG13samp.mov -- I see some momentary tracking jitter (layman), but it the video looks fine.

Buzzing is always tracking related. Sometimes the HiFi signal is screwed (bad tracking on it only), and you must use the linear one. Linear may be mono, and can sometimes be missing entirely.

The biggest issue is the video is a "clip show", with multiple sources cameras. Everything is all over the place, in terms of errors and values. It was embedded in an edit, as nothing was cleaned up in advance. To completely clean it up, you'd need to manually re-edit the whole thing PLUS restore each segment separately. Yuck.

Lordsmurf beat me to it (again). The VG13samp clip is at least workable, although I can't say why it looks badly dinterlaced and then reinterlaced. One or two segments look like telecine video that was deintelaced, as there are duped frames periodically in some sections. Or as LS says, maybe mistracking.

Not only would vgsamp2 have to worked in Avisynth one shot at a time, but those shots are really unfixable IMO. It's deinterlaced (incorrectly) then encoded as interlaced, and a lot of it is roughly duped multi-generation copies. Maybe you can get some of it to stop looking so purple, but because detail and color resolution are so meager to begin with it would simply vaporize with denoising. It's a shame. Not the O.P.'s fault, though. Get out the Colt .45 and start hunting down the makers.

Maybe the capture is uncompressed RGB32 ?

mm, sorry. I can't agree. The linked video is badly discolored and seriously noisy beyond repair. This shouldn't be a matter of opinion -- the damage is basic and the mistakes are obvious. True, most of the macroblocks are due to UTube's endless creativity in finding ways to wreck video, but the rest appears to be the owner's fault. Disheartening to see how the likes of uTube have trained folks to accept just about anything.

OK, maybe after breakfast I won't be such a grinch. Is it really that difficult to do a better job with video, even VHS? Or do people just let the video and the gear and software control them instead of the other way around?