Do I really need vhs-decode/Domesday, compile software?
 

I finally managed to grab a test RF file from a Matsui VN9730SIL, (which appears to have the exact same circuit board layout as the "Bush VCR925NSIL (Orion made)" images shown on the github tap list) using DomesdayDuplicator 2022.zip because all the later windows builds are encrypted (and I will deliberately use that word, despite the fact it's a technical exaggeration) with QT6.

I'm currently running an attempt to decode the raw 16-bit signed capture (having to adjust the command to "vhs" insteal of "--vhs" because nobody can be bothered to update the documentation.

It's an experimental SECAM capture on a PAL player so I've not got great hopes, although it does appear to detect a fair number of frames, maybe I'll let you know how that went...

To put things bluntly, there is a reason people disregard this otherwise promising technology - nobody is maintaining it to a sensible standard and it often ironically requires the latest operating systems (despite a manual build attempt of part of the software demanding python 3.8.10 - maybe I should compile that bit on my XP machine?!)

I'm hoping someone can come up with a proper build for the gui that doesn't demand version QT6 that's locked to windows 10 and dx12 .dlls because I know full well there is nothing in the gui that requires it and windows 10 has plenty of backward compatibility for native W7 software.

I'd be surprised if this thing couldn't crunch numbers on windows 95 to be honest, as I'm not expecting live video decodes, here!

Apologies for the rant but I've been at this for VERY MANY hours now and I'm tired, to be honest. :D

That's exactly the problem.

Most of the self-named "leaders" ("documentarian", devs, etc) are just young GenZers/20-somethings, who have no real-world experience at video. They refuse to learn proper methods, shun it, and make fun of people that have been doing this for decades. It's extreme immaturity and snark, and their communities are toxic cesspools that make Twitter look inviting and friendly.

I get comments on this frequently. For example, just last week:
And I have so many more.

Those kind of comments are common in private. "In private". That's because the immature leader TheRealHarrypm (Harry Munday, 24, of Oxford UK) will belittle the person, then ban them from their little communities. They remove any criticism, silence any voices with any degree of criticism. Harry seems to be unemployed, so he spends his day trolling online.

They've designated me enemy #1 because I won't be silenced. I have decades of experience in video, both in the hobby community (going back to the 90s) and the pro/studio community (going back to 00s). I was literally doing this stuff, to a serious level, before they were even born.

But let me re-visit this phase:
Domesday86 is old tech, created for digitizing Laserdiscs, based on Domesday originally developed by the BBC. Professionals created it, and it works stunningly.

Domesday86 is not vhs-decode. That's a hobby attempt to get Domesday86 to also work with VHS. And it sorta-kinda does, and sorta-kinda doesn't. It has many issues, both in quality of output, and quality of process (the gear, software, etc). Halo (ringing) is the worst part of vhs-decode. Attempts to compensate merely result in a blurred image that looks like it came off a bad composite VCR output.

vhs-decode has been out for almost 10 years now, nearly a full decade, and serious quality issues simply have not been addressed. So I would argue that the window for "promising" ended at least 5 years ago. It's no longer promising. At best, they're trying to "circle the wagons", and salvage what they can, and ignoring or poo-poo'ing the many drawbacks.

vhs-decode output quality is very dependent on the deck head quality, and the transport quality. vhs-decode modulates the data, but the VCR head does the pickup. Cheap consumer VCRs, their gear of choice, is notorious for having inferior heads and sloppy loose transports. And in 10-30 years of aging, typical lack of consumer maintenance, and it's not pretty.

For vhs-decode to ever succeed, as I have said for the past ~10 years since it was birthed as a project, dedicated hardware will be required. Both in terms of head reading (not using VCRs) and with a "black box" to process the data much faster than a desktop/laptop computer can. But no, that goes against their #1 "primary objective" (Robocop reference) that all of their process "must remain as cheap as possible".

So that's why you get
- thrift stores VCRs
- Aliexpress parts
- along with poorly-documented homebrew duct-tape-and-chicken-wire software

All for a low price that results in excessive amounts of time. However, for many of us, time is worth more than saving $100 or even $1000. These young 20-something do not understand this. They've maybe not experienced loss or bad timing, showing them how futile it is to spend gobs of time to save relatively little money (in the grand scheme of life). It's even more egregious when those same people have $5k+ "gamer" computers, $1k+ cell/mobile phones, etc. In that light, it not even about absolute dollar value, but willingness (or stubbornness) to spend what it needed for quality work. They have a "negative economics" problem.

Your "VERY MANY" hours simply will not end anytime soon, if ever.

That's why this project remains virtually still-born, near-vaporware noise, ignored by 99.99%+ of people, which includes most professionals and archivists (including ones working for large organizations like LoC). They want nothing to do with this mess. Nor do I, nor do many.

If you continue down this path, good luck.

And my condolences to those many sacrificed hours, frustration, and lost sanity. :screwy:

No need to apologize, many people feel with same, especially when they find out they've been lied to about how good vhs-decode supposedly is.

Realize that vhs-decode quality is NOT better than known-good workflows. :warning:

Namely:
- JVC/Panasonic S-VHS with line TBC (certain models)
- DataVideo/Cypress type frame TBCs (certain models)
- quality capture card -- where ATI/Pinnacle is known good (certain models), GV-USB2/Hauppauge is arguable, and Easycap/Elgato/ClearClick etc is known bad.

And all the gear is in proper refurbished condition, not just random "used" (abused) from eBay. Which is where most of the bad "comparisons" come from. Again, too many young users in that project that have never seen or used this gear, and have no idea what it can and should do. For example, that pathetic wiki has an image of a Panasonic AG-1980P with very obvious caps issues, to show a fake "comparison". It's been point out now, many times, by many people, and Harry refuses to remove it, or replace with an image from a recapped AG-1980P.

You've been misled, lied to, and fed nothing but FUD, from the vhs-decode community. Ranting is warranted. :congrats:

VHS-decode is like a communist government running a chain of grocery stores, Nothing works when offered for free, Profit ambitions can make things work but not to the extent of greedy capitalism, Greed had killed nice products as well in the past. They should have set up a crowd funding campaign, pay some smart people to get things done and have a nice product that works.

I forgot about RTFM (Read The Fucking Manual) that really does apply to Harry specifically. Any and all questions are responded with "just read the wiki idiot, the info is right here I ain't gonna dumb it down any further" Not to bash other members there, especially the project's proper author and creator Oln.

It sucks because what from I've heard from everybody else in these circles Oln comes across as a down to earth professional who understands that the project is just a hobbyists thing, he understands it's faults. We aren't shitting on him but how his project turned out to be after letting an arsehole run around.

The promoter's antics makes him look like an unprofessional dickhead and ruins whatever good will the project had. If they seriously want it to be the new standard, it should be more easier to setup and not require the average guy to become a electronics engineer.

But alas it feels like they are done, I think at this point their focus is just capture hardware and software updates. Yeah that's cool and all but that ain't gonna fix the biggest problem which is the wiki being a mess for a newbie and the complicated setup phase which requires soldering.

We need more people rightfully calling this crap out, it's not better for christ sake, they are all stuck in their own little alternate reality. It's different yes, but not as any different as using DV25 is.

It's like George Lucas making the Star Wars prequels and being surrounded by yes men, and look what that gave us..

I'd use it sure, but I'd use it rarely and for a small tape collection, why? Because I know it brings me no meaningful jump in quality, it doesn't give me a reason to redo my whole tape collection. Hell I'm only using it because I'd rather pay somebody to mod one of my S-VHS VCRs for me since all I can figure out is the software itself.

It's nice to see such quick (and sensible) replies.

Believe it or not my cmd window has literally just finished processing the 1hr25 SECAM recording as I was typing this (so now you have an idea how long that takes on Windows 7 with an Intel Core i7-6700K @4.00GHz). :eek:

I'm about to go out so I can't comment on quality (although it has skipped a bunch of frames).
Next time I'll just do a five minute test capture and use the smaller 10-bit filetype, maybe) :D

At least I have a couple of laserdisc players and some anime discs that could be spruced up so it isn't hasn't all been wasted experiments, even if I do have to go back to the s-video rips for the tapes. :)

That's where Domesday86 should really show a composite-vs-FM quality difference.

Laserdisc was a composite format, non-digital/non-file FM data saved to the optical platter. Laserdisc players, even "the best players", were composite-only or primitive-comb s-video out.

VHS didn't have Laserdisc problem, by way of using S-VHS players with line TBCs (mostly JVC, Panasonic). vhs-decode is entirely redundant, and actually looks worse than the best S-VHS decks, paired with quality capture cards. All that vhs-decode really does is make a low-end junk VCR look less bad. It doesn't exceed the best gear.

Laserdisc never had better gear, and is why nobody argues that Domesday86 is superior.

- For LD, it's best, it's just fact.
- For VHS, it's just FUD/BS from a bunch of cheapskates, mostly 20s kids --- but yet who still have high-end "gamer" computers, with fancy Nvidia video cards, gobs of RAM and SSD/HDD storage, to process the data. They honestly make no sense. To them, this is just a fun tinker project, proof of concept, their end goal isn't really about archiving video in the best quality.

With the exception of the actual dev, Oyvind/oln/hodgey who is a great guy, and doesn't get involved in all the BS drama.

Harry is the one who has turned vhs-decode into a cult, the "MAGA of VHS". The world really likes its cults lately. Too many people need/yearn to belong to tribes -- even if those tribes shun facts, science, truth, even common sense, as is the case here. Don't believe your eyes and ears, only the Dear Leader of the cult! So @Bypassall, you've now become a blasphemer for seeing the truth! It's so insane.

Brad/msgohan is probably the one person who disappoints me. He's become "one of us" (one of them), including making defaming comments towards others. He wasn't like that in the 2010s, when he posted here and on VH. It really is a cult. So sad what it's done to these people.