Panasonic AG-1980P, Kona LHi, best practices?

[latreche34]

Quote:

Originally Posted by mrmuy97

Exactly. It was yet another warning about the type of people found online who have seen/used/heard of "professional" video gear and think that an SDI or HDMI converter must be the best quality tool simply because the device outputs via a "modern" interface in comparison to analog. The insane amount of YouTube videos and forum posts to mislead others with incorrect information and plainly wrong "facts" is painful. But aramkolt seems to have a good grasp on reality.

You're confusing SDI with HDMI, SDI as I said is not a new port like HDMI, Also SDI is mainly a stream/capture port that can be used as a display port, HDMI is a display port that was made to be captured by 3rd party gadget manufacturers under no specific standard.

SDI was the port that defined the analog to digital conversion under the rec.601 standard back in the mid 80's that all modern consumer capture cards work according to, except the chinese fake devices and HDMI devices.

[lordsmurf]

Quote:

Originally Posted by latreche34

You're confusing SDI with HDMI,

I don't think he is.

Quote:

SDI as I said is not a new port

I would suggest this: it's new to others, both home/hobby users, as well as newly minted professional users. In any industry, older methods can get displaced for years or ever decades. And then, oddly, the older method makes a comeback. Sometimes for valid reasons, sometimes "just because".

As the costs of traditional known-quality s-video based workflow gear has increased, the cheaper elements of society have sought out alternatives. Most of them are harebrained stupidity (example = cheap Chinese HDMI converters made for DVD players and video game consoles), but some are sound under the right conditions, such as SDI I/O broadcast/studio gear. Keyword, key phrase = "the right conditions",

I highly doubt anybody here knows what NAB is/was (the Comicon of broadcasters, sort of like CES or defunct E3), or has attended it, but I remember when Blackmagic came to the broadcast scene 15+ years ago at NAB. It was interesting, and I was highly looking forward to adopting some of their higher end gear for both my work and hobby. I was so disappointed and disallusioned a few years later. BM gear, overall, truly sucks, for what we do (converting consumer analog SD to digital lossless). Nice on paper, PITA in practice.

Now, 10+ years later, non-pro users are "discovering" these SDI (and HDMI, and BNC) items, often to their own detriment. Same for the "cheap" (relatively speaking) eBay-found used boxes from Snell & Willcox, Aja, or others. Trying to force analog consumer formats (home footage) into pro devices rarely works well.

Now you know that, I know that .... but others don't know that.

They see a few "new" (not really that new, just updated old gear) items on Amazon and B&H, and think everything about it is "new" or (even dumber) "the future". So the cockamamie idea of "SDI = new" hails from there.

But as mrmuy97 pointed out, it's the device that is new, and seemingly never in a good way. It can use HDMI, SDI, or something that fell off a UFO, and it doesn't matter. The I/O wiring doesn't fix the crap device.

People sort of back into SDI this way, then "discover" all the dreadful pizza box rack items from the 1980s, which is precisely where some threads get derailed or sidetracked. That's the real rabbit hole of video, but not the fun Bugs Bunny kind.

[latreche34]

People who bash SDI and compare it to HDMI just don't understand how it works, let's agree to disagree and move on.

[lordsmurf]

Quote:

Originally Posted by latreche34

People who bash SDI and compare it to HDMI just don't understand how it works, let's agree to disagree and move on.

... then move along we shall.

[aramkolt]

The benefit of 10 bit for VHS is something I have yet to do any testing on. However, the more bits you have makes "banding" issues less likely. I believe the technical term is "quantization artifact" where a variety of analog sample points will end up "rounding" to the same digital output which creates bands of the same color when it should have actually been more of a gradient. This is mainly evident in areas that are mostly the same color such as sky.

This does a better job of describing it but should also apply to capture bit depth:
https://www.avsforum.com/threads/determining-display-panel-bit-depth.2424330/page-5

You should be able to tell if you feed the right gradient test pattern image into your capture process to see if it causes banding in areas with gradual color change.

How much you'd notice a difference on most tapes is another question. There should be some scenes where a true 10 bit conversion would result in a visibly better result. This is limited to the lowest bit conversion in your chain though, and I don't think it's really documented at what bit rate that internal line TBCs actually convert at.

[nightowl3090]

I'll toss my hat in the ring here to keep on topic. I've been using the Kona LHi for 2 years now and have been very satisfied with the results with Hi8 capture. Even more satisfying is that you can pick them up on eBay for less than $80 and the software: AJA Control Room is free. I'm now moving on to VHS capture and still evolving my workflow. Feel free to break down my setup and offer suggestions.
Video
JVC-MV45 (TBC: ON NR: NORM) ->S-Video-> DMR-ES10 (Black Level Input: Brighter, Black Level Output: Darker NR: OFF) ->Component-> Kona LHi
Audio
JVC-MV45 ->RCA-> DMR-ES10 ->RCA/XLR unbalanced cable-> Kona LHi

After reading this thread I just ordered the Rolls MB15b Promatch, mostly because I do find the low input gain an annoyance, not because I believe it will add any additional quality.

My main complaints with this setup have been discussed here: https://www.digitalfaq.com/forum/new...-dmr-es10.html

Having to weigh the benefits of the frame sync with the luminance crunching is a necessary evil. The Kona LHi isn't particularly happy when capturing S-Video directly from the JVC-MV45, preview lag, frame doubling. The Kona LHi seems far more stable and happy when it receives video from the ES10 in either S-Video or Component.

I'd like to get your take on the Control Panel settings. What color space and Analog Black levels are you all using?

controlroomsettigns.jpg

[nightowl3090]

Returning the Promatch after this scathing review: https://www.audiosciencereview.com/f...nverter.19865/

[latreche34]

Be careful when applying 7.5 IRE, For VCRs usually it is not needed, The VCR applies it to the output. That setting is put there for pro camcorders, as there are analog camcorders that apply 7.5IRE on their own, some don't and rely on the recording software. There is no need to capture in 10bit unless you're anal about restoration, I personally capture at 8bit, also because script software work on 8bit only.

This card is not built in any signal timing, that's why it relies heavily on your DVD recorder. For audio, It's a wise thing you used a balanced to unbalanced converter, it helps boosting the gain for low gain tapes. Personally I've managed to control the gain from the MediaExpress app with BE75 and shorting the left and right's ground to cold.

[mrmuy97]

Quote:

Originally Posted by mrmuy97

I consulted a friend who is a high-level audio engineer and electrical engineer and general A/V nut. He recommended:

#1 the ART CleanBox Pro [https://artproaudio.com/product/clea...el-converter/]
He highly suggested this and that's what I got and the thing is awesome. Build quality, sound quality, all the little details, just excellent. I'm a stickler for QA and this is a purchase I've been extremely happy with.

#2 If I couldn't get that, he said to grab a Rolls unit instead and that would be fine. Can't remember model info for that but if you end up needing it I can ask him again.

Quote:

Originally Posted by nightowl3090

Returning the [Rolls MB15b] Promatch after this scathing review: https://www.audiosciencereview.com/f...nverter.19865/

Hard for me to imagine a better unit than the CleanBox Pro for the price. Still recommending it 100%.

Quote:

Originally Posted by nightowl3090

After reading this thread I just ordered the Rolls MB15b Promatch, mostly because I do find the low input gain an annoyance, not because I believe it will add any additional quality.

A good idea for a comparison between audio capture with and without a level converter, to find out for sure.

Quote:

Originally Posted by aramkolt

The benefit of 10 bit for VHS

There's a consensus (no benefit) but I'm always interested to see an argument with evidence to the contrary. Latreche34 also asked if you can process 10-bit AVI files with QTGMC and got no answer: https://www.digitalfaq.com/forum/vid...t-version.html And he more recently stated it's not currently possible.

Quote:

Originally Posted by nightowl3090

I'd like to get your take on the Control Panel settings.

Overkill/inefficient ... Unless:
- You've seen a discernible difference comparing the same video capture in 8-bit versus 10-bit (many people would be eager to see such evidence)
- You've heard a discernible difference comparing the same audio capture in 16-bit versus 24-bit (many people would be eager to see such evidence)
I'm just being blunt about it, not attacking you. Capturing at higher quality than the original format could store is just wasting lots of space -- again, unless there's evidence to the contrary, which would be very interesting to see.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = =

Quote:

Originally Posted by latreche34

SDI is not modern

Quote:

Originally Posted by latreche34

You're confusing SDI with HDMI

Quote:

Originally Posted by latreche34

People who bash SDI and compare it to HDMI just don't understand how it works, let's agree to disagree and move on.

Unfortunately, we can't agree to disagree, because there's no disagreement in the first place.
You invented 2 imaginary arguments for yourself.
I never said "SDI is modern."
I never made any comments about SDI versus HDMI.

[latreche34]

Fair enough, Though that last post was a general statement, I was not addressing anyone in particular.

[aramkolt]

Quote:

Originally Posted by mrmuy97

There's a consensus (no benefit) but I'm always interested to see an argument with evidence to the contrary. Latreche34 also asked if you can process 10-bit AVI files with QTGMC and got no answer: https://www.digitalfaq.com/forum/vid...t-version.html And he more recently stated it's not currently possible.


Overkill/inefficient ... Unless:
- You've seen a discernible difference comparing the same video capture in 8-bit versus 10-bit (many people would be eager to see such evidence)
- You've heard a discernible difference comparing the same audio capture in 16-bit versus 24-bit (many people would be eager to see such evidence)
I'm just being blunt about it, not attacking you. Capturing at higher quality than the original format could store is just wasting lots of space -- again, unless there's evidence to the contrary, which would be very interesting to see.

= = = = = = = = = = = = = = = = = = = = = = = = = = = = =

Agree could be unnecessary and inefficient indeed. I'm still gathering all the stuff up do to the test, but I will post results once I do. I suspect that most internal line TBCs or line-like devices such as the ES10 may only convert at 8 or less bit, in which case that would be the bottleneck regardless of whatever other capture hardware you'd use later in the chain.

Should be pretty easy to actually test though, just needs something like a grey sweep pattern recorded onto a VHS tape vs sent directly by the pattern generator. If banding is seen with line TBC on, but not when it is off or when sent by the pattern generator, that would the internal line TBC having less bits of color depth than your capture card.

I think 10 bit captures would bee more commonly part of ProRes capture chain, so I'll try doing a capture with that, 8 bit AVI, and DV (which is also 8 bit, though is also 4:1:1) so if anything I'd expect the most banding to occur with DV which may be why it gets a bad rap for being blocky under certain conditions.

I haven't tried to see if QTGMC will deinterlace 10 bit ProRes, but I'll do that as well as part of the test. Agree doesn't make sense if your editing process requires a 8 bit for things like Avisynth. I've seen mixed postings about ways to get 10 bit to work and apparently it is possible with a lot of things. I think SD interlaced NTSC ProRes 422 with 10 bit depth takes up something like 15GB an hour from what I've read.

By the nature of the test, should be pretty obvious if the ProRes 10 bit of a grey sweep patter is having bit depth reduced during the QTGMC deinterlace as banding would show up after the deinterlace.

[nightowl3090]

I'll pick up the CleanBox Pro and report back.

Great input all around. And great catch on the 7.5 IRE. Those are the settings I was hoping people would look over and steer me in the right direction.
I went the overkill route on the Hi8 tapes because they were quite literally disintegrating after 1 play. So I did full bore on bit rate. I could probably roll both video and audio back now that I'm working with more reliable source material.

As for QTGMC goes, I've been using it regularly via Hybrid with these ProRes 422 HQ 10 Bit captures and results are excellent for typical broadcast media. I haven't done a gray sweep pattern or any sort of scientific assessment. But it passes the No crash, No glitch and looks good criteria. I am a bit concerned now that further filtering in Vapoursynth is being hindered by the 10 bit source? Again, I'm happy with the results I'm getting so I won't be losing sleep over it.