How to find new Panasonic AG-1980?

[latreche34]

I agree, The difference from the RF into the vhs-decode and the RF going through the VCR's own processing boards is going to be very minimal, Using a newer generation VCR that has in chip processing like the late 90's high end JVC S-VHS machines would probably tip the balance towards the VCR processing being better than the VHS-decode, The AG-1980 is an old beast, still uses some analog processing. No need to an external TBC if looking only for visual differences. And yes, so far, most comparisons done used cheap easycaps, We challenged them to a fair comparison using the same tape passed around and none of them accepted the challenge.

[lordsmurf]

Quote:

Originally Posted by latreche34

I agree, The difference from the RF into the vhs-decode and the RF going through the VCR's own processing boards is going to be very minimal,

I would more agree that "RF in software (vhs-decode) is unable to surpass RF into dedicated VCR chips". Sometimes it's the same, sometimes it's very much not the same.

Quote:

Originally Posted by latreche34

Using a newer generation VCR that has in chip processing like the late 90's high end JVC S-VHS machines would probably tip the balance towards the VCR processing being better than the VHS-decode,

Do you have any documentation on this? I need to go into R&D mode again, orchestrate some actions.

[latreche34]

There is a JVC technical guide floating somewhere, I think it was uploaded by OLN here and on VH.

[Slushly]

I'm going to use the same 1980p for both vhs-decode and traditional capture. The one I just bought is in very good shape, original, just needs a recap (so far lol). Yea, if you follow the discord, it's overwhelming how many tweaks people are posting on there and the learning curve is wild. I don't understand why people are going out of their way to use that crappy ali-express card because if they're already this deep into it, they might as well just save the money and buy the duplicator. It has better quality. Plus, people seem to forget that vcr heads vary in quality and signal strength, so I don't know why they're buying these combo thrift store decks.

Also, whats the deal with the vhs-decode subreddit? It's like the biggest echo chamber known to man. I get it. You're skeptical. But for people to call you a scammer and just be all around rude? I don't understand that.

When/if I finally get all the tools I need, I can give you information, results, and comparisons if you're interested.

[latreche34]

Some engineers spent their entire career developing the VCR technology from the early top loading ancient decks all the way to SoC (system on the chip), JVC in particular went great lengths on making sure they stay on top of everyone else, Their B.E.S.T (a.k.a video calibration) makes sure the signal is at its peak even with worn heads or worn tapes by adjusting heads' coils current, TBC and chroma noise reduction ensures stable and clean images, None of this can be achieved by a cheap VCR, If the RF signal is screwed by a cheap VCR, there is little you can do to iron it out. All what the vhs-decode system is doing in my opinion is just changing the way the scan lines are digitized and saved, that's all.

[lordsmurf]

This is one of my more textbook-y type posts...

My biggest problem with vhs-decode is all the nonsense about software TBC.

I was researching TBCs before most of them were even born. Some of those guys use technobabble tactics (ie, repeating jargon to "sound smart") to hide their lack of knowledge. Harry, the now-22-year-old, that only just started learning video in the past 4 years (ie, a bored teenager in the pandemic), is very guilty of it. He fools some newbies, but that's about it. As a Pickles comic read years ago, "if you can't explain it in plain English, you don't understand it yourself".

TBCs are essentially A<>D not too different from the RF/FM/decode/whatever methods. But the key ingredient is how that's done, and how the signal is processed. It's not a simple matter of a generic FPGA and off-the-shelf scripting. To even suggest that shows total lack of knowledge, akin to wanting to attach a lawnmower engine to a bicycle to make a motorcycle. But actual TBCs, actual motorcycles, require lots of R&D. There is a secret sauce to making a safe and effective product.

The basic concept of a semiconductor (semis, chips) is that it is designed with a base instruction set (essentially hardcoded/hardwired), and some user customizable area within those instructions (essentially sub-instructions). Well, video is complex, lots of data, lots of instructions.

- In the early/mid 90s, the reason for TBC weakness/inaccuracy was lack of the user customizable area, paired with non-robust/slow I/O.
- In the late 90s to early 00s, we got massive strides in RAM, some boosts to processing. TBCs really hit their stride, with great promise to come in just one more hardware generation -- TBCs that acted as capture devices with many options! (Yes, there are SDI boxes, but those are primitive tech compared to what should have existed.)

Sadly, this promising TBC generation never happened. The 08-09 recession (aka, the GFC, the Great Financial Crisis) harmed (or bankrupted) a lot of companies. Anything that was legacy (like analog related gear) at those companies was cut to reduce expenses, for the future survival of the company. That's why a lot of devices ceased to exist within 1 year of 2009 (2008-2010). In the now-2020s there is no ROI for the R&D.

So post-GFC, now we had a new problem. The chips needed were no longer being fabbed. Analog was old anyway, right? So no more hardcoded instructions needed. Some TBC companies tried to adjust to similar chips, but all failed (for example, the Cypress "black" generation of TBCs -- as I think Cypress tried to use old instructions with new/different chips).

The datasheets/specs for many of these chips is online, freely given out -- though not all, and some require expensive paid relationships (especially anything Intel/Alterra/etc). So there's a lot of amateur/Dunning-Kruger when it comes to old semiconductors. The vhs-decode community reeks of it, like rotting fish.

In the 80s-00s (and there was no 10s-20s), this was all treated as "trade secrets". Not just the on-chip code, but sometimes even the chip used (ie, sanded chips). Almost everything from that era is bitlocked, sometimes with self-destruct code if intrusion is detected.

The FM/RF/decode/whatever projects all attempt to recreate hardware, but they're only able to do it from a very rudimentary level, based on the bare hardcoding of the chips. They have no access to, often no concept of, the additional user-added instructions. So no secret sauce, maybe some ketchup at most.

To effectively craft TBC code, you need a code wizard that can extrapolate from a test library of thousands of tapes, combined with scoping. Harry/others have this naive idea that "open source will solve it", but there's zero control over the quality of the coders, or the quality of the testing library. It's all random. Thus why we see random results from the project, and why everything must be tweaked and re-tweaked ad nauseum. Tweaking in itself is not progress, sometimes it's just spinning your wheels, ie "development hell".

I've had a lot of people contact me over the years, wanting to remake TBCs/VCRs/whatever. But all it took for me was a single conversation with them (often over the phone) to realize they had no clue, they didn't know what they didn't know. I always kindly declined. (Well, there was one person that did impress me, "hmm, a smart labradoodle!", but that's a story for another day.)

vhs-decode is essentially fully reinventing the TBC wheel, and they're literally decades behind the curve. We now have unlimited RAM, but the ARM/x86 processing power is still vastly inferior/slower to FPGA/SoC/GPU/whatever from even a decade ago.

My initial thought process for vhs-decode -- long before Harry was involved, then mostly just hodgey/oln's own project -- was to attempt the next-gen "capturing TBC" that never existed. But they wanted to screw around with thrift store junk, act like asses on Reddit, and tinker for the sake of tinkering. Fine, whatever, I have videos to convert, no time for BS.

Somebody will eventually fork this commercially, more attuned to my idea, and I've already heard whispers in backchannels.

Unlike what some of these Reddit fanboys think, I'm not married to TBCs, or VCRs, or capture cards. I just use the tools available for the quality needed, as well as refurb gear (and yes, sell it, capitalism, the horror!) for quality-seeking members/readers of this site. If the new tool was better, I'd adopt it. But it's not. vhs-decode is a new inferior tool. I have no doubts that tests will prove this fact, so I'm all for it. At best, it's just different, as latreche34 writes. At worst, it's a primitive attempt to recreate 20-30 years old hardware tech.

I actually want to get with you Slushly, so I can confirm your findings.

(At very most, I do see some niche situations where VCR modulation causes bleeding not correctable in Avisynth. That bleeding/noise reduces sharpness in a roundabout way. In those atypical scenarios, I have seem vhs-decode samples that better allow the NR, and retain the sharpness better. So it has some use, and could be just another tool in the toolkit. This is something I've stated for years now. I have test tapes waiting for this.)

[latreche34]

When I look at the story of SingMai/Daniel, it reminds me of the VHS-decode project, Daniel was an ambitious guy, Besides other video projects related to his business, there was this VCR project that I was interested in, He actually approached it better than the vhs-decode team, He made everything hardware, Took the guts of a VCR, put it in an enclosure, attached two RF boards (I guess the second one was reserved for HiFi in the future), He built the decoding and digitization stages on a physical circuit board and made the output a legal rec.601 SD 4:2:2 interlaced video with digital audio over SDI, Not sure exactly what happens and the projects was abandoned.

He then focused on an old project of digitization of Y-C, composite and analog audio in 2 or 3 devices, namely the SM03, He almost perfected it, then the pandemic hit. Fast forward few years later, he gave it another shot but he had trouble finding the original chips he was using due to the chip shortage back then, The new ones did not perform as expected and were like 3 times more expensive, With little bit of patience, modification and tweaking he was able to overcome most of the problems he had with these new chips and then.... it was total silence, completely disappeared, no answer to e-mails, the website is gone.

The moral story is, if a company with engineers on the payroll could not pull it off, no one can. Don't get me wrong, I know there is Apple, Youtube, Facebook, Tesla ...etc started by dreamers working in their garages, But this is an obsolete task, aimed to a very specific group of people, Not the masses.

I hope they succeed, I really do, but not something I'm crossing my fingers for. Looking forward to comparison tests from the OP.

[lordsmurf]

Quote:

Originally Posted by latreche34

he was able to overcome most of the problems he had with these new chips and then.... it was total silence, completely disappeared, no answer to e-mails, the website is gone.

This is the common vaporware scenario that I've seen countless times over the years, even from large companies.

Quote:

The moral story is, if a company with engineers on the payroll could not pull it off, no one can.

I'm not sure about "no one can", but you definitely cannot as a broke kid with an attitude. This stuff takes time and funding -- and realize that time is a valuable asset, even more than money, something that vhs-decode fans seem to not understand.

Quote:

I hope they succeed, I really do, but not something I'm crossing my fingers for. Looking forward to comparison tests from the OP.

I've said this for years now.

So it's odd that Harry (and his flock) direct all this hate my way (and lies, and BS). You'd think I'd have told them I hope it fails (or worse). But no, I did not. They're just unhinged, seemingly pissed off that I don't worship at their alter, and will not join their cult that sees only perfection and no flaws.

Quote:

Don't get me wrong, I know there is Apple, Youtube, Facebook, Tesla ...etc started by dreamers working in their garages,

Ah, but you are overlooking something! The pivot. (Well, actually two things. "Garage" is a common myth. It's almost always bedrooms, rarely garages..)

Companies start with an idea, but it almost never happens that way.

- Youtube was started by lonely/single/horny Paypal employees wanting to make a dating site.
- Facebook was started by lonely/single/horny Harvard student to rate how hot a campus chick was.
- Tesla was more about the "E" than the "V", and years before Musk showed up.
- Apple did start as a niche computer hardware company, then floundered for about 25 years, almost collapsing twice. But then they pivoted and started to see extreme success in the past 2 decades. Those niche computers (Mac) sales are now only 7.5% of revenue, while iPhones and apps are about 75% revenue.

Successful companies, or even careers, almost never end up where initially conceived. My own video career was an accident -- the result of a hobby (video), a job (in another field) that ended before my first day (due to 9/11), and wanting to eat.

It's a lack of a pivot -- lack of leadership, lack of vision -- that ends in failure. Burying your head in the sand is far easier to do that to foresee what comes next. Many of the vhs-decode guy think they're the "way of the future", but analog videotapes are the past. They're fighting a war that ended almost 20 years ago. It's their hobby, their pet project, and they're not too different from (U.S.) Civil War reenactors.

[alitek12]

I've purchased multiple new AG-1980's before. I've found them by luck and did pay the premium. Just know though, a new 1980 is still a 25-year-old AG-1980 and those require the same recap work that a used one requires. I've confirmed this myself.