digitalFAQ.com Forum

digitalFAQ.com Forum (https://www.digitalfaq.com/forum/)
-   Capture, Record, Transfer (https://www.digitalfaq.com/forum/video-capture/)
-   -   BetaMax workflow with TBC? (https://www.digitalfaq.com/forum/video-capture/12642-betamax-workflow-tbc.html)

latreche34 03-24-2022 12:59 AM

Quote:

Originally Posted by lingyi (Post 83675)
FYI, in my Betaphile years, I'd wait until the top models dropped below $1000 before I got them. This was the main reason I never got a SL-HF1000 or EDV-9500. They never dropped below $1000. <GRIN>

I never had a Beta machine in their hayday but I've owned a Sony EDV-7500 in recent years and never really got any use of it so I put it on ebay last year and was sold for little over $500. Shortly after, someone asked me if I can transfer some ED tapes and realized I made a big mistake, Went back looking for an ED Beta machine and got an EDV-7300 from Canada, first owner, very clean with 13 ED Beta home recorded tapes, cost me $400 shipped to the US, Now I'm back in business.

Closecall 03-24-2022 04:35 PM

With all this said, knowing that there are S-video machines out there, through some more digging and research it does appear that base Betamax tapes would not benefit from S-video. How they were made/originally recorded would not benefit from the split of S-video vs composite.

lingyi 03-24-2022 05:24 PM

It's your call, but to my eyes, different machines and S-Video out definitely made a difference in the visual quality between the 20+ Betamax(s) I had and viewed on my Sony PVM2530 back in the day.

lingyi 03-24-2022 05:32 PM

It's your call, but to my eyes, different machines and S-Video out definitely made a difference in the visual quality between the 20+ Betamax(s) I had and viewed on my Sony PVM2530 back in the day.

Edit: Oddly, while someone (sorry can't remember who), said that his SL-HF860 had the best picture quality, the one I owned had a clearly soft picture. I chalked it up to the digital processing and effects.

Edit 2: These threads talk about Beta machines.

http://www.digitalfaq.com/forum/vide...cr-buying.html
http://www.digitalfaq.com/forum/vide...-guide-16.html

As the others have said, no matter how much you read and compare, the only real way to judge what is best is your eyes. Just a with VHS, no one machine will definitely be able handle every tape the best.

The most sage advice I ever received was "You only think he knows a lot, because of what you don't know."

latreche34 03-24-2022 10:48 PM

Quote:

Originally Posted by Closecall (Post 83698)
With all this said, knowing that there are S-video machines out there, through some more digging and research it does appear that base Betamax tapes would not benefit from S-video. How they were made/originally recorded would not benefit from the split of S-video vs composite.

A lot of people think that S-Video is an extra step splitting Y and C signals after being composite right after being played back from tape, It is the other way around, Y-C is the original signal on tape after the chroma under is recovered from tape and up converted back to its frequency with a slight loss, Then comes composite which is an extra step combining Y and C in one wire for convenience, Thanks to S-VHS who standardized the Y-C output into a socket called S-Video.

RobustReviews 03-25-2022 06:24 AM

Quote:

Originally Posted by latreche34 (Post 83706)
A lot of people think that S-Video is an extra step splitting Y and C signals after being composite right after being played back from tape, It is the other way around, Y-C is the original signal on tape after the chroma under is recovered from tape and up converted back to its frequency with a slight loss, Then comes composite which is an extra step combining Y and C in one wire for convenience, Thanks to S-VHS who standardized the Y-C output into a socket called S-Video.

Yes, this is something that's easily overlooked.

U-Matic used a similar method but without a (re)hetrodyne stage during dubbing, but that wasn't intended for the domestic market. U-Matic can output FM tape-carrier modulated 'C' signals directly into another machine rather than at baseband so there was little C-loss on direct UMatic transfers if they were made correctly.

They're at baseband on S-Video but it's still a substantial improvement over composite and better suited for domestic use. There is still a demodulation stage on S-Video for the C-Subcarrier so it's still far from a perfect solution but it's a substantial improvement on anything else.

Composite with adequate filtering and demodulation can look impressive (our For-A units are very good at it when they're set up properly) but as I've said many times, it's akin to 'unmixing paint' and it's a complex process to get right in hardware. You can get a fair way there with a handful of components but the devil is in the detail here, and fragile C signals need very precise separation if you don't want a lacklustre transfer full of dull details and dot-crawl everywhere.

I'm certain that you 'could' get the soldering iron out and attack a Betamax machine to extract base-band Y&C but that's beyond the scope of this topic. I need to have a look at this one day.

Then there's the SCART connector and it's myriad configuations, but again, this not for discussion here.

lordsmurf 03-25-2022 06:34 AM

Quote:

Originally Posted by RobustReviews (Post 83707)
Composite with adequate filtering and demodulation can look impressive
... but as I've said many times, it's akin to 'unmixing paint'

This needs repeating.

Too many folks hear/read "composite", and immediately dismiss the item as crap.

But I've said it many times, that composite itself isn't inherently bad. The main problem is that devices almost all do it badly. A few items do it good -- mostly some TBCs (not the usual suspects), and the rare VCRs (also not the usual suspects). Not DataVideo, Cypress, JVC, Panasonic. Though the crossover comp>svid of the green AVT-8710 isn't bad.

Unmixing paint. I like it.
Almost like unringing a bell.

RobustReviews 03-25-2022 06:52 AM

Quote:

Originally Posted by lordsmurf (Post 83710)
This needs repeating.

Too many folks hear/read "composite", and immediately dismiss the item as crap.

But I've said it many times, that composite itself isn't inherently bad. The main problem is that devices almost all do it badly. A few items do it good -- mostly some TBCs (not the usual suspects), and the rare VCRs (also not the usual suspects). Not DataVideo, Cypress, JVC, Panasonic. Though the crossover comp>svid of the green AVT-8710 isn't bad.

Unmixing paint. I like it.
Almost like unringing a bell.

Computationally doing it is almost certainly the answer as parallel processing using FFT is almost the preserve of modern computing hardware, but it's costly to do in bespoke prehistoric specific hardware.

I disagree with Panasonic, I think the Y/C separation is more than good enough for most purposes but we'll have to just disagree there.

hodgey 03-25-2022 09:43 AM

Well it depends on what you mean by Panasonic and JVC here - A analog Y/C filter in a early 90s vcr, a old video mixer and say a 2000s digital 3D Y/C in a TV or dvd-recorder are going to be vastly different, and may or may not be designed by themselves or others.

For somewhat stable PAL sources - the BBC transform decoder like used in ld-decode beats the pants off everything else though of course that's not something found in any commercially available equipment.

For more jittery stuff like VHS it gets rather complicated as fancier 3D YC filtering (and also transform for PAL) is somewhat reliant on the color signal lining up between lines so they will often fall back to something more primitive methods. Also due to the low bandwidth of VHS there isn't going to be all that much overlap between Y and C so very dumb/simple filtering (that would kill detail on a higher bandwith source) may sometimes work better given that won't be much content in the chroma frequency range that is luma.

As for S-Video modding VCRs - haven't looked into it for betamax, but have looked a bit at VHS diagrams and in the case of PAL decks it seems it would often be doable as most even in most newer video ICs where Y and C is in the same IC, the chroma signal out and then back into the IC right mixing to allow for external chroma processing in the case of french SECAM (which is normally a separate IC only present in secam decks). So, I would think one would just need to add some buffer/amplifying stage and not send the signal back in to extract the chroma separately, though it might be more complex in practice. The main video IC was often shared between vhs and SVHS variants in later decks as well (especially in JVCs) so one can see how the SVHS and VHS variants differ The further back you go the more split up processing is into multiple ICs.

Closecall 03-25-2022 11:48 AM

Quote:

Originally Posted by latreche34 (Post 83706)
A lot of people think that S-Video is an extra step splitting Y and C signals after being composite right after being played back from tape, It is the other way around, Y-C is the original signal on tape after the chroma under is recovered from tape and up converted back to its frequency with a slight loss, Then comes composite which is an extra step combining Y and C in one wire for convenience, Thanks to S-VHS who standardized the Y-C output into a socket called S-Video.

Thank you for this perfect explanation that helps to answer my question. Is it true then when ANY format of "tape", i e. VHS, S-VHS, Beta, Hi8, Digital-8, etc., is originally recorded that the Y/C signals are indeed separate? And is it for this reason that if a VCR (type) player has S-video output, the Y/C signals from the tape are not disturbed making this a preferred method of playback/transfer?

latreche34 03-25-2022 12:05 PM

The Y signal is easy to pull, The C one is always the problem, not only the up conversion part but the timing as well, Most add on Y-C kits and mods for uMatic pull the luma (Y) signal from the DUB connector (used to dub tapes using the chroma under signal to avoid generational loss of up and down converting chroma) and the chroma (C) from the composite out since it is already timed to the signal and its resolution is very low to begin with, Basically this solution focuses on providing a clean Y signal with no chroma interference. There are mods that pull the chroma from inside the main video board but it is model dependent.

As to modifying consumer gear is worth it or not, Most low end consumer VCR's don't have line TBC so it is not worth modifying them, The ones that have line TBC are most likely S-VHS machines, therefore do have a proper S-Video socket.

Beta in the other hand is Sony and it's very proprietary, modifying it to enable a socketed Y-C pins is not going to be easy without studying how the processor work.

latreche34 03-25-2022 12:08 PM

Quote:

Originally Posted by Closecall (Post 83724)
Thank you for this perfect explanation that helps to answer my question. Is it true then when ANY format of "tape", i e. VHS, S-VHS, Beta, Hi8, Digital-8, etc., is originally recorded that the Y/C signals are indeed separate? And is it for this reason that if a VCR (type) player has S-video output, the Y/C signals from the tape are not disturbed making this a preferred method of playback/transfer?

Digital8 is not an analog format, it works differently using DV codec. Yes the Y and C are separated in the frequency domain on analog tapes.

RobustReviews 03-25-2022 12:13 PM

Quote:

Originally Posted by Closecall (Post 83724)
Thank you for this perfect explanation that helps to answer my question. Is it true then when ANY format of "tape", i e. VHS, S-VHS, Beta, Hi8, Digital-8, etc., is originally recorded that the Y/C signals are indeed separate? And is it for this reason that if a VCR (type) player has S-video output, the Y/C signals from the tape are not disturbed making this a preferred method of playback/transfer?

Don't worry about the digital stuff for a moment, that's quite different and has a myriad of different obfuscations.

Very, very simply (we can build from here) - think of it like an audio cassette.

1) To record a signal on to magnetic tape, an electromagnet and a tape and moved relative to one another.
2) Slow speeds relative to one another result in lower quality recordings, this is because of loss of high frequencies. In audio recording 15ips (inches per second) was used in studios, it was unfeasibly fast for home compact cassette players (1 7/8ips) but it lead to the best quality reproduction. Long story short, the faster you move a tape relative to the head, the higher the frequencies you can record.

Now imagine recording a television program from your roof aerial, that came in as an NTSC modulated on to whatever channel you were recording. We'll take out the broadcast frequencies and we're left with a straight NTSC signal, this is what we want to record onto the tape.

NTSC contains two parts, a luminance component (the black and white bit) and the chroma (colour). The chroma part is considerably higher frequency to the luma, whilst the tape can record all of the luma signals, it's not moving fast enough to be able to absorb the colour information..... Bummer.

What is employed in the videotape is a technique called 'colour-under' whereby the colour signal is converted to a lower frequency which fits nicely in the tape's recording zone. This means some colour information has to be discarded, but effectively our eyes and brains handle most of that, but it does give that classic 'videotape' look. It can be placed 'under' the luminance signal and stored on the tape.

This really is about as simple as it can be made, trust me, it'll make your head spin* when you dig into how it works in cold science and mathematics, but if you just want a simple explanation that's a basic guide to the 'why' if not the how.

The alternatives would be to move the tape supremely fast or use more video heads, neither of which are effective strategies, especially in low-price domestic machines. A few early experimentations were made recording colour without the technique but speeds and tape consumption were unbelievable.



Digital is a different kettle of fish, colour is often discarded but that's going to confuse the topic here.

*ha, it's not lost on me.


All times are GMT -5. The time now is 06:57 AM

Site design, images and content © 2002-2024 The Digital FAQ, www.digitalFAQ.com
Forum Software by vBulletin · Copyright © 2024 Jelsoft Enterprises Ltd.