The Question of TBC's

[anachronon]

I am not trying to get another cat fight going. Still, I feel that the question needs to be put to bed, once and for all.

Yes, a TBC in the digitization workflow will definitely improve quality. But, how many home users have $2K to drop on such a unit? So, what is the best option for home users? There has been much "discussion" over which VCR's or camcorders have a built-in TBC. Or, more appropriately, "pseudo-TBC". If one clearly stands-out (either VHS or Hi8) please let everyone know. This reminds me of the old argument over "comb filters" in video processing (a big marketing claim, back in the 90's). Sure, they improved the signal quality, by reducing high-end noise and aliasing. But, the image quality was still limited by the response of the CRT.

Though there have been countless opinions, it really seems to come down to one thing: Which capture device works best with which unit, and which software? It's like the old question of which video card works best with which motherboard and/or OS. Actually, some of the best advice that I have read here, has nothing to do with TBC's---like, "Rip your video to a second hard drive, and not the one that is running the OS." That suggestion alone will cut down on a great many dropped/inserted frames and audio sync problems---the number one question that I have encountered here. In my case, my difficulties seem to stem from a truly crappy capture device. I will try one that has been recommended, and see if that improves the situation.

As for my experience, I have worked repairing electronics since the late 70's. I was a leading repair tech at the major warranty service center in the SF Bay Area, from the late 80's to the mid 90's. My specialty there, was TV's, VCR's and camcorders. I would go through about 10 units in a typical day. I saw every kind of repair, problem, customer complaint, use/abuse, etc... I did so many, that I still see the mechanisms in my sleep. I learned which units could handle what. Just recently, I received flack on this site, because I said that I had to clean the heads of my Hi8 camcorder twice, over the course of 60 tapes. Actually, in Sony's service bulletins of the time, they recommended professional cleaning/servicing after every ten tapes. Sony 8mm and Hi8 tapes use a thicker magnetic coating, that is more abrasive than standard VHS. Cleaning was also more difficult, as any excess fluid could damage the head-drum bearings---unlike a standard VCR.

In more recent years, I have worked with building and repairing computers (more as a hobby). I am just new to the realm of video A-to-D conversion. But, I learn quickly---provided that I don't have to fight a lot of attitudes and contradictory information. And, "flaming" me isn't going to help---even if it does make you feel better. Our discussions are supposed to be about sharing useful information, not "vindication".

So, could we please, keep this civil. A lot of non-members browse this site, looking for information. They do not want to see us behaving like a bunch of second-graders. Commercial video experience is definitely valuable. However, 99% of the folks coming here, are looking for a way to do this at home, on a home budget. Please, let's keep that in mind.

[lollo]

Yes, a TBC in the digitization workflow will definitely improve quality.

I am not an expert, but what I experienced is that a line TBC is always needed, but in rare cases the frame TBC is not necessary.
I think imho that I am lucky to be in one of this few cases and I will not introduce an additional element in the capture chain if not needed,
I am not even sure it is 100% trasparent. The price and the availability is also a problem for somebody, as you said.

About the attitude the members should have, I agree with you

[dellsam34]

External frame TBC's don't improve picture quality and they shouldn't change any value of it, They are made for compensating lost VBI signal, In most cases a good capture card/device (Not what's being made now) and a normal tape that still has the integrity of the signal will be just fine for capturing, assuming the computer CPU is adequate to process a lossless stream, The capture's card frame sychronizer will rebuild the VBI signal anyway.

As I always say capture one hour of continuous video and inspect the file, If you see nothing wrong with it you're good to go, I've captured tons of V8/Hi8 tapes with Pinnacle 500-USB and a decent camcorder with AmarecTV and never once needed a TBC, As the matter fact I try to stay away from it as I possibly can, It's two additional conversion steps that I don't need in my workflow (analog to digital then digital to analog) and ends up in most cases with changes to chroma and luma values.

Yes not all people have $2000 in the bank especially now, For that much money just send the tapes to someone who you can trust to capture them for you at a fraction of that cost.

Since you worked in the field of camcorders and VCR's do you remember this tape that use to come with Sony camcorders? It came with a customer's tapes and I though It's a good idea to put it on youtube, it may bring back someone's memories.

[lingyi]

First, read and thoroughly digest this: http://www.digitalfaq.com/forum/video-restore/2251-tbc-time-base.html

Bottom line: According to lordsmurf, in my mind and probably others, THE authority on TBCs hows and whys, an external TBC is a most for quality captures.

Also, recommend posting this at digitalfaq.com as lordsmurf is more likely to respond there or PM him to invite him to this this and your other thread.

[anachronon]

Yes, I did read that page by Lord Smurf. I studied it before I composed my post, to refresh my memory. I am sure that Lord Smurf will pop-in on this discussion any time now. I am interested in reading what he has to say.

This discussion has reminded me of an interesting story. Back when I worked at the shop, we had a bin of junkers (insurance write-offs) from a local department store. They were shelf stock that had been water-damaged by a burst pipe, during the Quake of '89. Now, they were awaiting a trip to the dump. Basically, it was a "help yourself" bin. So, I snagged one of the Sony 8mm camcorders, to see if it was salvageable. I could use it as a backup, for places that I didn't want to bring my good camera---like the beach, the bar, rowdy parties, etc... Surprisingly, it cleaned-up well, and worked great---for the first couple tapes. By the fourth tape though, the head-drum bearings began to go out (due to the water damage), and the head drum eventually seized completely. What could I expect for "free"?

The point of this story, lies in the results of that tape's playback (on a good Hi8 unit). During the last few minutes of the junker's recording, the head drum had begun to slow (due to drag from the failing bearings). On the playback of that part, the image started developing tracking lines, then horizontal jitter, and eventually a horizontal tear. Finally, in the last couple minutes, it went to a full-on vertical roll, before recording completely quit. I got to thinking how that tape would surely test the robustness of a TBC. Just how much of that error could it correct?

[dellsam34]

The point of this story, lies in the results of that tape's playback (on a good Hi8 unit). During the last few minutes of the junker's recording, the head drum had begun to slow (due to drag from the failing bearings). On the playback of that part, the image started developing tracking lines, then horizontal jitter, and eventually a horizontal tear. Finally, in the last couple minutes, it went to a full-on vertical roll, before recording completely quit. I got to thinking how that tape would surely test the robustness of a TBC. Just how much of that error could it correct?

None, You can't make chicken salad out of chicken shit, If the camcorder is broken nothing can magically fix the signal. The TBC has to receive a recognizable signal to work properly, Damage to the tape and/or to the camcorder cannot be recovered. I have a tape with frame roll and no TBC on earth could play it back properly.

[anachronon]

Oh, I wasn't about seriously wanting to fix that video. I just figured that it might make an interesting test. Due to the dragging head drum, the field stripes on the tape are seriously off-angle. I doubt that even the most profession equipment could compensate for that (though the FBI or NSA might have such secret equipment in their basement).

[anachronon]

P.S. Y'all take this way too seriously. Remember, most folks visiting this site, are doing this for fun! Are we having fun yet?

[dellsam34]

No, I'm not taking the bait.

[dellsam34]

One thing I haven't experimented with is a D8/DV camcorder in pass-through. Unlike Hi8/S-VHS camcorders and VCR's which their internal TBC cannot be used in pass-through mode, D8/DV/minDV decks and camcorders don't have a pure analog signal path, They all take the analog input (if such feature exist), digitize it and one output will be encoded to DV and passed to firewire, the other output is converted back to analog and passed to A/V output. As you can see the analog signal has been processed by the ADC chip at which point is being digitized, time base corrected and converted back to analog after leaving the DAC chip, I'm hunting down for schematics and diagrams to better understand this.

If this is proven to be effective this will be the alternative to the ES10/15 DVD recorders and a list of camcorders capable of this feature will be listed.

Keep in mind this is going to be very limited list if not at all because most consumer camcorders don't offer separate inputs and outputs analog jacks.

[anachronon]

Does that apply to S-video as well?

[lordsmurf]

line TBC = clean the image
frame TBC = clean the signal

You need both.

In some uncommon situations, with pristine tapes (signal-wise, not something you can see), an ES10/15 strong+crippled line TBC with a mere non-TBC frame sync can suffice. But you also get all the downsides of the ES10/15 (posterization, agreesive NR even when "off", luma changes, just an overall degraded+digital look, though not DV'ish or MPEG'ish or outright crappy hardware in use). More ideal is to chase the ES10/15 with the DVK, as it's a true weak frame sync TBC in a non-TBC device (be careful with in/out path, some bypass TBC).

I'm not a purist, I simply want quality results. If you tapes are good, and you can take the quality hit (without it cascading), then go for it. But most tapes won't check that box.

The original intent of ES10/15 as passthrough (as discovered by me, may years ago, 2005) with as anti-tearing, chased by actual frame TBC, an apparent unintended hapy bonus/accident use. Panasonic designed it as pre-processor in the recorder to attempt good DVD transfers, but that failed because Panasonic MPEG encoders and encode settings suck.

It's like this:
You want to convert videos with quality buy a car.
So you get the S-VHS VCR with line TBC (or Hi8 camera with line TBC), and frame TBC go to a reliable dealer.
Or you can get the ES10/15 find some dude on Craigslist selling a beater that gets you from A to B.
Not quite that bad, but you get the idea.

[dellsam34]

Does that apply to S-video as well?

Yes, For miniaturization purposes camcorder manufacturers opted for one S-Video, one composite and one RCA audio port that are used as both inputs and outputs selectable by the mode the camcorder is in, VCR mode will turn ports into inputs, Playback and recording mode turn the ports into outputs. Then later on replaced by mini jacks for more miniaturization, Though high end camcorders kept one S-Video port and replaced composite and audio stereo with a 3.5mm jack.

Candidate camcorders will be some prosumer DV/DVCAM and some consumer HDV camcorders and maybe some early D8 camcorders, Will dig it further.

[anachronon]

So, if I am reading this properly, "If it ain't broke, don't fix it!" Can I gather from what has been said, that using the Panasonic ES-10 or ES-15 (VHS to DVD recorders) on pass-through is not a "lossless" process? Or, is it that certain settings need to be used? I intend to store all of my master copies in a raw, lossless format. That gives me the option to later make copies, and process and compress each new copy for a particular use---burn to DVD, convert to 60fps for Youtube or media player, etc...

There is really only one quality issue that I have noticed with my capture tests. This is the occasional/intermittent swapping of fields (top on bottom, bottom on top). I suspect that this is actually a problem with a crappy (slow) video capture card (I have a hopefully better one on order). Sure, I could fix this in VirtualDub. But, the fewer "fixes" that I have to do, the better.

[dellsam34]

For the best quality use as little pieces in the chain as possible, Every piece you enter in the system incur a loss, This is what made me look for a digital TBC. They are more efficient and they don't convert back to analog, When the signal is digitized for TBC it stays that way and sent out in its lossless state via an SDI port. But very few were designed for consumer formats and they are rarer than the unicorn unless you want to rent it or pay the online broadcast stores premium price of $1000-$1500 for one.

One thing is more important in the entire workflow than any other piece and it is using a S-Video capable machine that is equipped with line TBC, Capturing is different than playback so line TBC is a must for a proper archiving (Notice I didn't use the word digitizing), Back in the day TV's had well designed line delays that worked well with consumer VCR's and camcorders, Not the capture cards and certainly not the Chinese ones.

Hopefully the VHS-decode team will come up with a solution that allows people to use any regular VCR and the board will take care of video processing and timing, Until then don't throw your tapes away.

[orsetto]

Many of us have tried to thread the needle between lowly VHS/DVD combo recorder and full-bore LordSmurf/DigitalFAQ workflow, sometimes for budgetary reasons and sometimes when unusual tapes perversely resist "proper" capture methods. While it is possible to spend less and obtain a median quality between the two extremes of workflow, too many variables present themselves to ever definitively say "use this one specific relatively affordable midrange setup".

Camcorder TBC passthru runs the gamut from amazingly good to ghastly, depending on each camcorder sample, their condition, and the tapes being passed thru. Panasonic ES10/ES15 passthru solves many problems but not all, and can add artifacts of their own as a tradeoff. A "forgiving" capture dongle will be more amenable to not using a true frame TBC, but the majority of capture cards/dongles available today are viciously unforgiving. The non-reference setup that occasionally works for me and some odd tapes won't necessarily suit you, what works for you may not suit dellsam34, what works for him may not suit LordSmurf, and even LordSmurf has made changes to his reference systems over the years as hardware evolved (or regressed), or specific tapes demand.

While expensive and a huge PITA to second-hand source these days, the LordSmurf/DigitalFAQ/VH recommended "Dual TBC" system (line TBC equipped VCR + external frame TBC) ensures the best most consistent capture results from the widest range of tapes. Once you start deviating from that system for either budgetary or oddball tape reasons, all bets are off and things get very unpredictable very fast. Some ordinary consumer VCRs will track certain problem tapes better than premium JVC/Mitsu/Panasonic VCRs with built-in line TBC, but when fed thru a capture dongle the result is unusable garbage.

Adding a DataVideo TBC alone may or may not fix this, often one needs to also loop thru a Panasonic ES10/ES15 to simulate some of the corrections the premium VCRs handle. Or you may get along well for some time with "just" an affordable plain-jane VCR and generic EZcap, until the day you hit a run of tapes that choke your capture device unless you finally add the DataVideo TBC (oops, there goes your budget). Or maybe you make a significant change to your display or playback system, and suddenly flaws that were completely concealed become horribly visible (again, oops).

OTOH, some get lucky using just a generic Amazon capture dongle with their grandmother's old VCR (or so they often claim here in threads). The results are typically mediocre or worse, but its a certainty most of the consumers who purchased DVD/VHS combo decks or use Granny's VCR with an EZcap would never dream of dropping the $2K or more that a good used JVC DigiPure VCR + DataVideo TBC costs today, nor would they entertain the thought of capturing with VirtualDub and post processing. Those that care enough to need or appreciate the best result find the money and learn the additional effort, those with less demanding standards throw together a cheap random system, use a combo unit, or ship their tapes off to a mass market service like Legacy Box (which is no better than using a combo deck, really, but does take the job off your hands).

Short version: spend the cash for a VCR with line TBC plus an external DataVideo frame TBC for the best chance at good quality with fewer nasty surprises. Omit either TBC, and the variables multiply with the quantity of tapes you need to archive. Nothing is perfect or foolproof: TBCs and VCRs can get defective, and neither TBC is totally transparent (the vcr line TBC/DNR cleans up noise and flagging at the expense of some detail/motion smearing, and as dellsam34 noted external frame TBCs add a redundant A-D-A step). But for a large project of widely varying tapes, its better to have all possible tools on hand to juggle until everything clicks.

[lingyi]

So, if I am reading this properly, "If it ain't broke, don't fix it!"

More like you if you don't know it's broke (i.e. how much potentially better your captures could be), it's fine. But be aware when someone shows you how it's "broke", you may regret your earlier choice.

In the early 2000's, many of the Hong Kong movies I wanted were available on VCD or extremely hard to find and expensive Laserdiscs. I knew the VCDs were "broke", but was satisfied with the quality. When they started to become available on DVD and now sometimes Blu-Ray, I realized how 'broke" (i.e. extremely poor quality) the VCDs were. Note that many LDs were also broke, due to poor press quality, poor mastering and laser rot.

The most sage advice I got from one of my bosses was: "You only think he knows a lot because of what you don't know!"

[anachronon]

That gets into the question of exactly what is "broken". There are the obvious things, like horizontal tear, tracking lines, etc. But, what are some of the more subtle cues?

I have read that "aliasing" (jagged edges on diagonal lines) is one cue. However, every case of aliasing that I have encountered on my captures, has been due to random field swapping. Is this a problem that a frame TBC would fix? Or, is this a capture device and/or software issue? I suspect the latter. At least, this is an easy fix, when using a lossless codec. Still, I would prefer not to have to "fix".

For reference (I know that I have mentioned this before) I am using the S-video output on my Hi8 machine---even for standard Video-8 tapes.

Which, gets into the issue of "sharpness". From what I recall, analog video could only resolve about 400 lines in the horizontal direction (out of a possible 720) even with S-video. S-video improved sharpness, by separating the chroma signal from the 'Y', and eliminating the need for a chroma subcarrier. This allowed the chroma to have a resolution close to that of the 'Y'. Still, old analog simply didn't have the bandwidth. Broadcast analog had only half of the "400". It should be noted, that analog video never had discrete pixels. The number "720" simply refers to the number of horizontal phosphors on a standard color picture tube (I believe in both PAL and NTSC). Later VGA computer monitors had better bandwidth and resolution. But, analog TV's, VCR's, and camcorders still had that same 400-line limit, right up to the end.

That also gets into why old camcorder images are often not that sharp. It did not make sense to build high-end optics into a device that had such limited resolution. Or, so the thinking went. In reality, the difference between videos shot on a cheap camcorder, and ones shot with a high-end or professional model, can be quite striking.

[dellsam34]

Consumer video tape formats are around 200 luma levels per scan line, 400 figure is for Hi8/S-VHS/Super Beta, ED Beta resolved about 500 levels per scan line, Chroma in the other hand is about 30-40 levels per scan line for all.

Signs of needing a frame TBC are audio drift (if not related to computer issues), Frame jump, flashes or loosing signal, frame roll. Though a frame TBC cannot fix those problems 100%, If your frame is stable, your audio is in sync you don't need a frame TBC.

[lollo]

If your frame is stable, your audio is in sync you don't need a frame TBC.

So, assuming that a "frame not stable" is recognized, this perfectly summarize the topic. Thanks!

[orsetto]

That gets into the question of exactly what is "broken". There are the obvious things, like horizontal tear, tracking lines, etc. But, what are some of the more subtle cues?

In most forum sites and threads, the topic of "subtle cues" in great vs good or poor captures is divorced from the topic of basic capturing itself. You're more likely to glean info on subtleties from more specific threads where the poster presents a sample clip and requests suggestions for improvement. Sometimes the suggestions will involve a complete re-capture from scratch, but most often you'll find script recommendations or other post-capture advice. Your field swapping alias issue would be considered intolerable and steps would be suggested to eliminate it during or after capture. Innumerable threads and posts bemoan the "halo-ing" caused by many VCR models (or aggressive sharpening in post): this halo issue could definitely be categorized as one of the "subtle cues" noticed by dedicated video hobbyists but largely ignored by Average Joe/Jane. There are others of course, more visible/annoying in specific types of video: animation and concert bootlegs can present challenges not seen in wedding videos or tv show recordings.

To the absolute horror of many perfectionist regular forum contributors, Average Joe and Jane really do not care one bit about the "subtle cues" that differentiate a really good VHS capture from an average to barely-adequate capture. All they care about is that the capture plays without massively blatant defects like excessive tearing, jumping, jitter, blackouts, or loss of lip sync in the audio. In recent years especially, 80% or more of capture-related threads started here and elsewhere come from ordinary consumers who simply want to know why their $39 Amazon dongle won't produce stable (or even recognizable) video. These people have been conned by the USB dongle industry into expecting any random dongle connected to any random VCR thru any random capture app will "just work" out of the box. So they assume they're simply misreading instructions or need an updated driver or some other trivial corrective will solve their problem.

When informed that USB dongles are insanely sensitive to the grossly crude VHS signal format, and that they likely need additional very pricey second-hand hardware to actually get reliable performance out of their impulse-buy USB dongle, they naturally balk (very often never bothering to reply or return to their thread). We're so far into the era of smartphones, streaming and apps now that the many vagaries of analog are long forgotten and baffling to the mass market: if they can't get a capture dongle to work at the push of an app button, they're "over it" and don't want to discuss it any further. All they wanted was an inexpensive warts-and-all digital copy of their VHS tapes that doesn't black out, jump or lose audio sync. Few give a rats ass if the cap is grainy or smeary or discolored or aliasing as long as it plays uninterrupted and can be sent to youTube or their phones (a month later the capture is utterly forgotten and never viewed again anyway). Artifacts or instabilities that bother many of us here on VH don't faze the typical USB dongle customer: most of these knockoffs aren't mfd with us in mind.

[dellsam34]

Keep in mind that capture cards from back in the day are totally different from the USB dongles being sold today, They were designed with better processing chips, These modern knock offs USB dongles are more suitable for retro game captures and possibly a clean Hi8 signal from a S-Video socket, If you have composite forget about it.

[anachronon]

Actually, I have been rather pleased with the results from my first Hi8 capture. Needless to say, I had researched all of the necessary capture settings, beforehand. The couple problems that I encountered were quite fixable in editing (I'm using a lossless codec). Still, I suspect that the problems that I had, all stemmed from a crappy, slow capture device, and the fact that I was saving to my main hard drive, and not directly to my external 6TB drive (which will be my main video storage).

I mostly asked about "subtle cues", because someone implied that those minor issues could suddenly pop-up as major problems, down the road. So far, my captured clips have looked way better than they ever did on an old CRT television. I was really expecting disaster with a 30-year-old tape. Luckily, it had been stored well. Still, it's hard to achieve perfection with a recording that wasn't that great in the first place. Of course, if anyone has suggestions on capture settings that I may have overlooked...

It seems ironic that the best capture devices were made at a time when home computers were barely able to handle video capture/processing. Now that computers can process video without breaking a sweat, and storage devices have grown large enough to hold uncompressed video files, the selection of capture devices has shrunk to almost nothing. The fact that I could get such good results from a 1990 recording, proves that the market is still out there. But, the tech industry always seems in such a rush, to leave everything "old" behind.

[lordsmurf]

It seems ironic that the best capture devices were made at a time when home computers were barely able to handle video capture/processing.

You're confusing encoding/editing with capture (ingest).

The speeds and codecs of capture have barely changed in 20 years.

- graphics cards don't matter
- RAM doesn't matter -- 1gb was good minimum (actual minimum was less), 2gb was plenty
- CPU barely matters (for SD), and CPU is mostly about MPEG capture without frame drops at the higher broadcast bitrates,

Video processing has come a long way. But capture is evergreen. That's why "tell me newest best method to capture video" from newbies is often met by eye rolls. Some technology doesn't move fast whatsoever, if ever. The younger users tend to be the ones that think everything always changes, but it's not accurate to reality.

[lordsmurf]

That gets into the question of exactly what is "broken". There are the obvious things, like horizontal tear, tracking lines, etc. But, what are some of the more subtle cues?

In most forum sites and threads, the topic of "subtle cues" in great vs good or poor captures is divorced from the topic of basic capturing itself.

Part of the issue with this conversation is that it approaches the topic with misinformed assumptions. Namely that video capture has good/better/best type issues.

This is how newbies think:
-best
-excellent
-great
-good
-good enough
-not perfect
-kinda bad
-bad

But this is reality:
-best
|
- some issues, but overall good (using better gear like non-TBC S-VHS, ES10/15+DVK, good card or recorder, etc)
|
|
|
- not quite unwatchable aka "good enough" (tape half-ass cooperates in old VCR, with crap capture cards)
- bad (tape will not cooperate, still crap gear)

There's a gulf. Rarely are issues and island. Dropped frames, tracking, timing wiggles, audio sync -- it's all related. It's not either/or, but all or none.

Innumerable threads and posts bemoan the "halo-ing" caused by many VCR models
but largely ignored by Average Joe/Jane.

I disagree. Large HDTVs make it obvious now. They may live with it, but it's not unseen or ignored. At most, brushed off with the ol' "good enough" excuse. But it also depends on how bad it gets. Some is tolerable, other halo-ing is so ghastly that it really does ruin the image with ghosting and weird contrast errors.

To the absolute horror of many perfectionist regular forum contributors,

I really disagree with this.

For whatever reason, I'm sometimes painted as a "perfectionist", but that's silly. For example, I find some of the Avisynth threads ridiculous, trying to literally remove pixel-sized noise in a still image, or trying to correct a single bad frame. I don't lose the forest for the trees. I always have a macro view of projects. I try to take a 95/5 type approach, where 95%+ of problems are resolved, and that last 5% resolution will depend on time/work/effort curve and the ROI. And while some may think "95 is high!" it's really not. The hardware resolves much of this, and then basic software techniques can knock out the rest. It's the "last mile" concept. Sometimes folks insist on that last 5% no matter the cost.

Average Joe and Jane really do not care one bit about the "subtle cues" that differentiate a really good VHS capture from an average to barely-adequate capture.

Perhaps. But it depends on how subtle.

A common gripe is "black clipping" or "crushing blacks" (and blamed on the capture card), which is an absurd way to describe what's actually happening. The video had illegal levels, and were discarded. The tape/video is at a fault, not the capture card. Most people will not care about this, and indeed often I don't either. Illegal values are really only useful for restoration, when the footage is underexposed. As-is capturing doesn't need that 0-15 YUV. When you monkey around with black/IRE/brightness values, you reduce contrast and saturation. Most folks find that more offensive. That's a photography/videography concept, and "losing" levels into extreme shadow can be style. I'd much rather have an ATI 600 USB card (captures legal values only), and not a random POS card that capture 0-15 (and screws up other values, has other issues).

Mild dot crawl is another. (There's also confusion about "all composite is bad" when the reality is that the device (with composite) is bad. Yes, composite inherently has downsides, but too many crap devices amplify it. That's the device, not composite.)

Everybody disagrees (even whines) about sharpness, pro and and laymen alike. Not just fake sharpening (both for and against), but resolution (including both fake resolution, and true optical resolve). But the reality is that this is overblown. Throwing a VCR into edit mode, to get a tiny % extra sharpness, only to increase noise (and thus NR work post-capture, if possible) is somewhat silly. I'm long been firmly in the AUTO/NORM camp, not the EDIT/DETAIL camp, with some exceptions when possible, when needed.

All they care about is that the capture plays without massively blatant defects like excessive tearing, jumping, jitter, blackouts, or loss of lip sync in the audio.

Image wiggling, chroma noise, and color loss (ie DV capture) is getting a more vocal opposition as well. Again, large displays, easier to notice these problems there days.

In recent years especially, 80% or more of capture-related threads started here and elsewhere come from ordinary consumers who simply want to know why their $39 Amazon dongle won't produce stable (or even recognizable) video. These people have been conned by the USB dongle industry into expecting any random dongle connected to any random VCR thru any random capture app will "just work" out of the box. So they assume they're simply misreading instructions or need an updated driver or some other trivial corrective will solve their problem.

Yep.
That USB thing is crap. You need the standard recipe: VCR/camera > TBC > capture card/recorder (not just any random devices, but specific ones)

When informed that USB dongles are insanely sensitive to the grossly crude VHS signal format, and that they likely need additional very pricey second-hand hardware to actually get reliable performance out of their impulse-buy USB dongle, they naturally balk
if they can't get a capture dongle to work at the push of an app button, they're "over it" and don't want to discuss it any further.

Not really. Not now. Most people pursuing this in the 2010s/2020s are trying to correct bad work, or do it the right way. In the 2000s, sure, lots of cheapskates and stubborn folks.

[anachronon]

What is overlooked in this discussion (here in 2021) is that PCI capture cards for old SD analog video are no longer manufactured. Even on eBay, one would be hard pressed to find a PCI card for analog video. Modern PCI cards are all HDMI input, and geared toward gaming. No one is selling the older cards hailed here. Even if one did become available, would it work with a modern motherboard and OS? It would seem that the USB devices (that are being so maligned) are currently the only option.

I do have an old Pentium 4, XP machine, that would likely run one of those old PCI cards. It even has a second, internal, 1TB HDD, used purely for storage. I still use it for photo editing. Even so, it is near the end of it's useful life. That PC dates back to the days of the Radeon AIW.

[lordsmurf]

What is overlooked in this discussion (here in 2021) is that PCI capture cards for old SD analog video are no longer manufactured. Even on eBay, one would be hard pressed to find a PCI card for analog video. Modern PCI cards are all HDMI input, and geared toward gaming. No one is selling the older cards hailed here. Even if one did become available, would it work with a modern motherboard and OS? It would seem that the USB devices (that are being so maligned) are currently the only option.

Nonsense.

"No longer manufacturer" = who gives a rat's ass. There is extant supply.
In fact, right now, I actually have some of the rare good (it had low production) EOL ATI AIW PCI cards available (not the garbage you find on eBay/etc). I came across warehouse NOS, many now gone.

PCIe (PCI-E, PCI express) cards are extremely hit-or-miss, mostly miss. But then again, all comms are the same: PCI, PCIe, USB, etc. The comms doesn't determine cards quality, the card does. So it's a silly argument. Newer isn't better, older isn't better -- better is better.

I do have an old Pentium 4, XP machine, that would likely run one of those old PCI cards. It even has a second, internal, 1TB HDD, used purely for storage. I still use it for photo editing. Even so, it is near the end of it's useful life. That PC dates back to the days of the Radeon AIW.

It can work. Depends on specs. Not just "P4", but exact CPU, RAM, motherboard, chipsets (VIA is awful, nothing but problems).

I use AGP cards in my Asrock board that has SATA2, dual-core E7xxx CPU, and is not junk. Until 2015, I used it for restoration. It's not an "old" system. In the tech world, "old" means worthless. "Legacy" is older but useful.