Acquired good equiptment. can't figure out an archiving workflow.

[PVideo]

I've been reading forums here and DF for years and finally took the leap and purchased a sought after S-VHS player and an inexpensive alt via a Vhs/DVD Recorder. With the combination of usb dongles, usb capture devices, a folder full of software, and scores of youtube videos; I'm having trouble etching out a workflow. I just want to make sure that I'm taking the right steps with the right stuff on day 1.

First of all, I should start with my ultimate goal. My intent is to archive old VHS videos, rare commercials, and other findings primarily to youtube. Is there something else i should get, like an affordable elgato or avermedia capture hub? Do i need both the VCR and VCR/DVD Recorder?

VHS Equiptment:
Panasonic AG-1980 Desktop Editor VHS Recorder
Panasonic DMR-EZ48V VHS/DVD Recorder

Capture Devices:
Mini- AV2HDMI RCA Upscaler 1080P
Avermedia ET113 Dongle Component (YPbPr) to HDMI
Guermok HD AudioVideo Capture HDMI
Redyo S-video/RCA to USB Capture

Software:
OBS
VirtualDub2
StaxRip
MPC-BE
Brovicon

I've noticed that the VCR/DVD Recorder via HDMI output is a better quality than the SVHS Player, and i understand that i should want to do post-processing on the pc instead of the Recorder doing it. I'm including pics of my VHS player's video heads, i worried they might need to be replaced.

Any help and/or guidance would be appreciated! Hearing so many different takes on Youtube and the forums has me a bit overwhelmed.

[aramkolt]

I think you're probably getting better results from the 48V via HDMI is because the digitizing cards you are using are generally considered poor, and the other thing is IF your AG1980 hasn't been refurbished yet (heavy recapping and mechanism rebuild), then I wouldn't expect it to perform well. I do full refurbs for $750 and it usually takes around 8 hours solid at a minimum even having done 50+ of them before. PM me if interested.

You do also have a lot of dirt buildup right above the heads which is easily visible on your photos. That can be cleaned with the alcohol and paper method with 90%+ isopropyl alcohol. You can't really tell head wear from pictures though. I will say it's relatively uncommon for AG1980 heads to wear out as I think they used harder material than the JVCs did, even though most AG1980's have probably seen significantly more hours of use as they were used in more professional environments.

My general experience is that the 48V produces a significantly softer picture than the AG1980, but if softer/more smoothed out image is your preferred "look" for the final capture, you may be ok with that.

If you're curious what the difference would actually be for a different capture card and one of the tapes you've captured with your machines vs one of my refurbished AG1980's and the capture setup I'd normally use, you can send me a tape or two that you've already captured along with a flash drive containing your capture and I can capture a portion and send it back with my capture to see if you think it's worth changing/upgrading your setup or not. I'd do it for just the cost of return shipping.

[llansing]

You should had look further into vhs-decode if you're willing to spend $1000+ on equipment. It's less than half the price that can do one to one copy from the RF capture.

[The 14th Doctor]

Though keep in mind Decode isn't for everybody, and there are major drawbacks. Such as painful long decoding times (took me an hour and 45 minutes just to decode 5 minutes and 40 seconds worth of video) and not at all easy to simply jump into, requiring one to modify an existing VCR, adding in external components and whatnot.

Captures with that method tend to also introduce ringing/haloing artifacts that are not present in a traditional capture workflow, unless the tape already has such artifacts baked into it. If so then the decode method might make those artifacts even more pronounced and noticeable.

Using a suggested workflow can surpass Decode as it stands, way less of a headache if all the components are in refurbished shape, and without the extra problems decode brings to the finalized capture.

Sure with decode you can preserve the VBI/Line 21 data, but the reality is that sort of information isn't that all important in grand scheme of things.

[aramkolt]

I don't personally have a problem with VHS_Decode, it just isn't quite as user friendly (both hardware and software) as traditional captures. The original pitch where it was a slightly modified CX card which could be had for around $40 and 20 minutes of work apparently is no longer the recommended method as they now recommend a clockgen mod (requiring a custom PCB) which is apparently needed to keep the audio in sync. Generally it's also recommended to use an RF amplifier which essentially buffers the original RF and gets it to a more uniform level that the duplicator will do better with and not affect the original video output of the player for live previewing purposes. Both of these can be sourced online from HarryPM, but prebuilt ones are often out of stock and ship from overseas if you're in the USA. Not insurmountable barriers, but additional barriers nonetheless. There are newer devices in the pipeline that I believe have one or both of those built in, but prebuilt ones are not easily available/in stock to my knowledge.

This is the most recent head to head test I found from one of the admins:
https://www.youtube.com/watch?v=B2ldecEC_jA

I will agree that the left decode process looks sharper, but the right was also captured via composite (not S-Video) and also passes through a DVD player which will essentially stabilize the video, but also cause a generation loss/noise reduction to some degree (and noise reduction = detail loss). It is unknown to me if the DVD player processes internally at 8 bit or 10 bit and if the staxrip is working at 8 or 10 bit. It is clear that the decode pathway used in the demo is 10 bit from start to finish.

I'd also say that decode is about trading cost for time and complexity while achieving a likely moderate bump in quality over a good capture setup. A 1 hour capture with a traditional setup to the time you can watch it can take 1-2 hours at most whereas the same capture may take 10+ hours in processing time per hour depending on your computer.

Here's a roundabout review about the actual process for capturing a 1.5 hour movie with decode from about a year ago:
https://www.youtube.com/watch?v=pEzmbw_Y-Tw

Per this video, for this 1.5 hour movie, it took about 15 hours to process to get to a viewable video file and required a total of 700GB for the final video plus intermediate files, though at any given time it sounds like you'd need 500GB at a minimum. Don't get me wrong, for a rare film/media preservation, I'd make that investment also.

If decode could be done in twice that of realtime, at twice the hard drive space, and with twice the complexity both hardware and software wise, I'd consider it as a standard process, but as it stands, it's closer to 10x for each of these in my opinion and based on the Ricky1 video above. Perhaps there have been improvements in the past year that improve upon those, but I think it has a ways to go to being used mainstream as a standard process.

Just my opinion of course, but would love for you to try it and report back with your experience.

I think you'll find plenty of that people that did get the setup working and are quite pleased with the results, but for each success story, my guess is there are several more that gave up in frustration.

I actually have the DdD hardware, but I've had issues with the Mac compatibility and my newer computers happen to be Macs, so I haven't been able to do much in the way of direct comparisons myself as of yet.

[The 14th Doctor]

I don't personally have a problem with VHS_Decode, it just isn't quite as user friendly (both hardware and software) as traditional captures. The original pitch where it was a slightly modified CX card which could be had for around $40 and 20 minutes of work apparently is no longer the recommended method as they now recommend a clockgen mod (requiring a custom PCB) which is apparently needed to keep the audio in sync. These can be sourced online from HarryPM, but prebuilt ones are often out of stock and ship from overseas if you're in the USA.

This is the most recent head to head test I found from one of the admins:
https://www.youtube.com/watch?v=B2ldecEC_jA

I will agree that the left decode process looks sharper, but the right was also captured via composite (not S-Video) and also passes through a DVD player which will essentially stabilize the video, but also cause a generation loss/noise reduction to some degree (and noise reduction = detail loss). It is unknown to me if the DVD player processes internally at 8 bit or 10 bit and if the staxrip is working at 8 or 10 bit. It is clear that the decode pathway used in the demo is 10 bit from start to finish.

I'd also say that decode is about trading cost for time and complexity while achieving a likely moderate bump in quality over a good capture setup. A 1 hour capture with a traditional setup to the time you can watch it can take 1-2 hours at most whereas the same capture may take 10+ hours in processing time per hour depending on your computer.

Here's a roundabout review about the actual process for capturing a 1.5 hour movie with decode from about a year ago:
https://www.youtube.com/watch?v=pEzmbw_Y-Tw

Per this video, for this 1.5 hour movie, it took about 15 hours to process to get to a viewable video file and required a total of 700GB for the final video plus intermediate files, though at any given time it sounds like you'd need 500GB at a minimum. Don't get me wrong, for a rare film/media preservation. I'd make that investment also.

If decode could be done in twice that of realtime, at twice the hard drive space, and with twice the complexity both hardware and software wise, I'd consider it as a standard process, but as it stands, it's closer to 10x for each of these in my opinion and based on the Ricky1 video above. Perhaps there have been improvements in the past year that improve upon those, but I think it has a ways to go to being used mainstream as a standard process.

I don't have a problem with it in theory, it's just in practice is has less than desirable results. There are some posts of users voicing their issues and overall opinion on the method

Well that and all the drama filled in the community, I'd rather not use a method that is being promoted by somebody who tried to dox and leak the personal information of somebody else, with them ending up getting the personal info for somebody completely unrelated to who they were trying to go after. Kind of ruins their trust and credibility to be honest.

I also wouldn't personal invest in it because I'm in the same camp as Ethan Nunn, I am rather satisfied with the results I am already getting. That and I don't want to mess with soldering, I don't want to bother doing electronic work. Heck I'm too worried to even clean my own decks and would rather have a local pro who has been in the scene for 40+ years (and also thankfully dismisses Decode, so I'm not alone) to do that task.

That and even if I wanted faster decode times, I'd have to invest in a 1 to 2k gaming desktop setup since I primarily use laptops for my daily and work life. With that cash I'd rather buy a refurbished VCR and suggested capture card.

And yeah, remember hearing that Latreche34 called that video comparison unfair, and I agree.

[llansing]

The OP is looking for archiving, not a quick job. And nothing is beating a direct copy of the tape in the RF level. The quality of the decoder and the processing time will improve over time.

[The 14th Doctor]

I beg to differ

Besides isn't that a problem, that it isn't perfect yet, and we are just waiting for it "to get better" while these tapes sit here in storage, slowly rotting if not kept in good condition.

Yes using less than ideal equipment such as HDMI upscalers and OBS is a quick job, but not a suggested workflow that is comprised of components that have been the standard for around 20 years.

[lollo]

And nothing is beating a direct copy of the tape in the RF level. The quality of the decoder and the processing time will improve over time.

False.

We all had great expectations on vhs-decode, but so far there is not a single evidence that is superior to the standard approach, i.e. capturing with one of the recommended cards a 4:2:2 Y/C time base corrected signal into a lossless codec from a high end S-VHS VCR.

If you have such evidences, just post it.

[llansing]

I beg to differ

Besides isn't that a problem, that it isn't perfect yet, and we are just waiting for it "to get better" while these tapes sit here in storage, slowly rotting if not kept in good condition.

Yes using less than ideal equipment such as HDMI upscalers and OBS is a quick job, but not a suggested workflow that is comprised of components that have been the standard for around 20 years.

I think you confused yourself with the decoding part of the process to the capturing part. Capturing the RF means that you're archiving a one to one copy of the current state of the tape, you don't need to recapture again. Decoding is the process of turning that RF signal into viewable format, which will get better over time, just as to any other developing software.

[lordsmurf]

You should had look further into vhs-decode if you're willing to spend $1000+ on equipment.

Why is it that vhs-decode fans always jump immediately to costs, when it wasn't mentioned, and clearly wasn't a problem so far?

You get what you pay for.

I think you confused yourself with the decoding part of the process to the capturing part. Capturing the RF means that you're archiving a one to one copy of the current state of the tape, you don't need to recapture again.

Well, somebody is sure confused.

There is nothing special about vhs-decode. It simply attempts to recreate, in software, what a with-TBC VCR did 30 years ago (ie, the "decode"). But worse, as it adds artifacts that are not from the VCR.

And then it entirely relies on the quality of the VCR used (ie, for the "RF capture"), including head condition, transport stability, etc. But when you use thrift store VCRs (as 90%+ of the vhs-decode fans do), there's no real quality or stability to be had.

Many people have had to re-capture (and re-capture, and re-capture) footage, often with no useful results for their excess invested time. At least with a standard workflow, you'll immediately know if a problem exists, and can you fix/compensate. With vhs-decode, you won't see the problems until much farther along in the process, hours later.

Not to mention the constant changes to the RF hardware needed, to fix/patch method failings.

Decoding is the process of turning that RF signal into viewable format,

Yes, but
- normal VCR = instant
- vhs-decode = hours later

which will get better over time, just as to any other developing software.

This has been stated for at least 6 years now, and it's still not proven to be true. It's a "kick the can" attitude to the failings of the method, and has been pure propaganda thus far.

The entire process is an ordeal, a huge time vampire.

[llansing]

You should had look further into vhs-decode if you're willing to spend $1000+ on equipment.

Why is it that vhs-decode fans always jump immediately to costs, when it wasn't mentioned, and clearly wasn't a problem so far?

You get what you pay for.

Looks to me the OP is finding out now with that $1000+ VCR not performing as expected. And there's another person offering a $750 invoice to have a look, Jesus...And surely it's the vhs-decode fans that always talk about money.

And then it entirely relies on the quality of the VCR used (ie, for the "RF capture"), including head condition, transport stability, etc. But when you use thrift store VCRs (as 90%+ of the vhs-decode fans do), there's no real quality or stability to be had.

Lies. As the Harry guy already proved it here with both the high end S-VHS VCR outputting virtually the same image quality as an lower end VCR. It's all the same at the RF level.

Many people have had to re-capture (and re-capture, and re-capture) footage, often with no useful results for their excess invested time.

Lies.

With vhs-decode, you won't see the problems until much farther along in the process, hours later.

There is option that you can decode just a short clip for snippet.

Decoding is the process of turning that RF signal into viewable format,

Yes, but
- normal VCR = instant
- vhs-decode = hours later

From my experience with LD-decode, decoding 100 minutes of lds on my Ryzen 7950x took about 3 hours, not good but not too bad.

which will get better over time, just as to any other developing software.

This has been stated for at least 6 years now, and it's still not proven to be true. It's a "kick the can" attitude to the failings of the method, and has been pure propaganda thus far.

I don't know what world you live in because I didn't really see the many "problems" with the decoder that you are complaining. For example, look at the captures of this guy, they all look pretty good to me.

[The 14th Doctor]

[Attachment 88491 - Click to enlarge]
[Attachment 88492 - Click to enlarge]

Even with Twitter image compression you can see the ringing/ghosting around objects, it is subtle yes, but it is there. I can recover details in the shadows too just by adjusting the proc amp of my capture card so the blacks and whites are being captured within legal values, both images have the contrast boosted to 200% just to demonstrate.

[Attachment 88493 - Click to enlarge]
[Attachment 88494 - Click to enlarge]

https://x.com/AndySaysHiBTW/status/1954081674842542155

This Andy guy knows the right way around here, good for him!

From my experience with LD-decode, decoding 100 minutes of lds on my Ryzen 7950x took about 3 hours, not good but not too bad.

Well from my experience it took me an hour and 45 minutes to decode just 5 minutes and 40 seconds of a tape that somebody else archived because I couldn't be bother to keep it decoding the rest of the tape, and this was on a Lenovo Thinkpad P15v Gen 3 (AMD) Mobile Workstation Laptop running a AMD Ryzen™ 7 PRO 6850H.

Also LaserDisc decode is wildly different from VHS decode, that has way more of a success rate and is the proper way to archive LaserDiscs, there is no argument on that front at the very least from what I've seen.

Looks to me the OP is finding out now with that $1000+ VCR not performing as expected. And there's another person offering a $750 invoice to have a look, Jesus...And surely it's the vhs-decode fans that always talk about money.

Shame on you for dissing Aramkolt like that, he is trying his hardest to keep these Panasonic 1980s alive and well. They are great decks for archiving tapes, especially EP mode, mono audio and VHS-C tapes from what I've been told. These decks were probably around 1 to 2 thousand dollars new, these were high end professional machines meant for editing.

But they are a bitch and half to get properly working. As Lordsmurf has said in the past they are money pits. You want to know why Aramkolt charges almost $800 for a full refurbishment? Well it's a full refurbishment, he can comment in greater detail if he wants to.

He goes through that deck top to bottom and does around 8 hours of work to future proof these aging decks as much as he can. Unlike TGrantPhoto who only do a partial refurbishment from my understanding (Aram has needed to re-refurbish decks), and he charges twice as much money for a 1980 as Aramkolt does ($2400 vs $1200)

[lordsmurf]

Looks to me the OP is finding out now with that $1000+ VCR

$1000 is your random number, not the OP. We have no idea what he paid for this, it wasn't stated anywhere. Odds are, it was a $200 fixer-upper, or a lying SOB $500+ eBay seller. In which case +$750 repair is in-line to a $1000+ total deck cost (for a pro deck that had $2000+ MSRP).

Lies. As the Harry guy already proved it here with both the high end S-VHS VCR outputting virtually the same image quality as an lower end VCR.

Great. Now try again with an S-VHS VCR that is actually suggested.

There is option that you can decode just a short clip for snippet.

That's never how video errors work.

From my experience with LD-decode

Off-topic.

because I didn't really see the many "problems"

Most cults have willful blindness to obvious issues.

[llansing]

Even with Twitter image compression you can see the ringing/ghosting around objects, it is subtle yes, but it is there. I can recover details in the shadows too just by adjusting the proc amp of my capture card so the blacks and whites are being captured within legal values, both images have the contrast boosted to 200% just to demonstrate.

If there were ghosting in the original source, then it's not the job of the decoder to remove it. You may suggested that the ghosting was introduced by the decoder, but looking at his other capture, there were anime that have no ghosting like this one.

From my experience with LD-decode, decoding 100 minutes of lds on my Ryzen 7950x took about 3 hours, not good but not too bad.

Well from my experience it took me an hour and 45 minutes to decode just 5 minutes and 40 seconds of a tape that somebody else archived because I couldn't be bother to keep it decoding the rest of the tape, and this was on a Lenovo Thinkpad P15v Gen 3 (AMD) Mobile Workstation Laptop running a AMD Ryzen™ 7 PRO 6850H.

Also LaserDisc decode is wildly different from VHS decode, that has way more of a success rate and is the proper way to archive LaserDiscs, there is no argument on that front at the very least from what I've seen.

I'm just giving my side of the decoding speed. Faster PC, faster processing time.

Looks to me the OP is finding out now with that $1000+ VCR not performing as expected. And there's another person offering a $750 invoice to have a look, Jesus...And surely it's the vhs-decode fans that always talk about money.

Shame on you for dissing Aramkolt like that, he is trying his hardest to keep these Panasonic 1980s alive and well. They are great decks for archiving tapes, especially EP mode, mono audio and VHS-C tapes from what I've been told. These decks were probably around 1 to 2 thousand dollars new, these were high end professional machines meant for editing.

But they are a bitch and half to get properly working. As Lordsmurf has said in the past they are money pits. You want to know why Aramkolt charges almost $800 for a full refurbishment? Well it's a full refurbishment, he can comment in greater detail if he wants to.

He goes through that deck top to bottom and does around 8 hours of work to future proof these aging decks as much as he can. Unlike TGrantPhoto who only do a partial refurbishment from my understanding (Aram has needed to re-refurbish decks), and he charges twice as much money for a 1980 as Aramkolt does ($2400 vs $1200)

I'm just being real here, prop him up all you want, but $750 is a lot of money.

[llansing]

Looks to me the OP is finding out now with that $1000+ VCR

$1000 is your random number, not the OP. We have no idea what he paid for this, it wasn't stated anywhere. Odds are, it was a $200 fixer-upper, or a lying SOB $50+ eBay seller. In which case +$750 repair is in-line to a $1000+ total deck cost (for a pro deck that had $2000+ MSRP).

It's wouldn't be "a leap of faith" like the OP said if it's just $200.

Lies. As the Harry guy already proved it here with both the high end S-VHS VCR outputting virtually the same image quality as an lower end VCR.

Great. Now try again with an S-VHS VCR that is actually suggested.

Point has already been proven, they produce the same image quality in the RF level.

From my experience with LD-decode

Off-topic.

No it's logically relevant. If the decoding speed on my 100 minutes LD source is 3 hours, then logically the decoding speed of a 90 minutes VHS source is not going to take 15 hours.

[The 14th Doctor]

Well those screenshots still show ringing/ghosting, it is more noticeable when it is up against a solid white object, like the eye or Yu-Gi-Oh card. That and these captures have a lot of ugly rainbowing. Same deal as when Slushy compared a Decode and traditional capture using his own Panasonic 1980.

And yeah you are right, but what about captures that might have ringing baked into it (IE a copy)? Well I bet decode would amplify the already baked in ringing and/or halos, therefore making it even worse. Then you get results like this where it's double ringing.

This is peak example of the phrase "Now that I see it, I can't unsee it"

No it's logically relevant. If the decoding speed on my 100 minutes LD source is 3 hours, then logically the decoding speed of a 90 minutes VHS source is not going to take 15 hours.

But you just said it depends on PC specs, and more importantly, single CPU core performance. Still, LaserDisc Decode and VHS Decode are different because the formats are vastly different. One is the 100% best way the archive LaserDiscs, the other is debatable and is being outright dismissed by some users and even professionals who have been in this field for over 20 years. (Source, I know one)

I'm done, I don't want this to turn into a big fat argument that is ripe to be taken out of context to push their agenda. Sorry if I upset anyone.

[dellsam34]

Capturing the RF means that you're archiving a one to one copy of the current state of the tape, you don't need to recapture again.

Very doubtful that capturing RF is a one to one copy of the current state of the tape, All we know from the current refined decoding process that it appears it is not the case, due to the artifacts introduced in the video that are not present in the direct display from the VCR which clearly is not a result of the decoding process but rather the RF capture process, The best method so far, is a S-VHS VCR built in LTBC connected via S-Video (YC) to a capture device with built in TBC that can output digital directly via SDI, lossless firewire (requires Apple OS) or HDMI (as in some legacy HDMI VCRs).

On the Audio side, I have yet to see a direct comparison between HiFi RF and a conventional analog stereo capture with downloadable audio files since youtube crams everything down to their lossy audio codec, But I doubt the difference in quality will be distinguishable. If it is, it will be useful for music videos.