Audio Delay from Frame Synchronizer TBCs?
 

My capture card is a Kona LHi using breakout cables; this integrates audio and analog video together but does not take in to account any audio synch issues created by the frame buffer in a Frame Synchronizer.

I read though the forum to see if there is a device that will delay the audio to match the frame buffer in the Frame Synchronizer? I hear this sync amount is small maybe 1/60th of a second so it sounds like for most people it is too much trouble to synch the audio? Devices like a Leitch DPS-475 or a For.A FA-9100 have integrated Audio though Phoenix Connectors and are supposed to correct for these audio delays along with SDI output. If the delay is irrelevant matter why do these devices correct for loss of audio synch?

Devices like the TBC-1000 and For.A FA-125 ouput in S-video, so if this is connected to a capture card in S-video doesn’t that mean the image is going from Analog to Digital and then back to Analog and then Digital? Does this “conversion” cause any undesirable effects on the output video?

I am giving this a closer look because it seems like the Frame Synch might be adding a reddish to the skin tones, I want to try another Frame Synch but I am still troubleshooting that phenomena. Which is the straightest path to digital with the least number of devices and failures? Any other suggestions? Is there anything that will put a delay on the audio to match the Frame Synch?

For single sources, it's just anal retentiveness.
For multiple sources (ie, genlock required, etc), it's needed.

Yes, but not in the way you're thinking. You're looking at it from a capture card stance. It's more along the lines of SDI fully uncompressed 4:4:4 conversion.

Rarely.

What model?

It is a Leitch DPS 475, low hours, purchased from original owner back in 2011. I just sent it to the fellow that used to make the repairs for the Leitch Corporation he said the unit was in great shape and he finely tuned it and said he found no issues. I just got it back, I need to test and see if the reddish skin tint reappears.

I only notice the reddish skin tint when I input through component or S-Video and ouput with Component video, consistent on the AG-1980, HR-S9600U and a Sony SVP-5600, and disappears without the frame synch. The effect is slight and in some cases even desirable. When I output SDI I only noticed a slightly darker image, but no reddish skin tone shift. I am also troubleshooting my cables to see if there could be an issue there. Originally I was using the breakout box but it positively malfunctioned so I want to simplify the chain by making it as simple as possible.

For academic and educational purposes is there a way to make up for the audio synch lost to the 1/60 second buffer in a Frame Synchronizer?

In hardware, maybe some audio mixers will allow it, or have delay as a byproduct of the mixing.
In software, just offset it in the editor by 1/60th of a second (~16ms).

^ See that. 16ms. It usually takes at least 100ms to see any change in lip sync. I'm good, and I have trouble detecting offset below 50ms. And even then, only some kinds of content allows detection below 100ms.

That's probably why. Don't do that.

You should normally try to output the same as the input
- composite > composite
- s-video > s-video
- component > component

You rarely have good reason to swap midstream. Only in special circumstances is it suggested.

Is it generally preferable if you input S-Video or Component to output SDI if available? A Kona LHI will capture from S-Video, Composite, Component or SDI & even HDMI.

Just capture SDI, to avoid the extra conversion back and forth.

I don't agree with this. Composite, for example, must be separated into at least Y/C to process it. Outputting composite then involves recombining, and then the capture device must re-separate. Each step adds an opportunity for degradation.

If you suspect the TBC is messing up chroma, the best way to test is to throw up some color bars and scope the output if you have access to that equipment. You can also do it post capture with AviSynth or VirtualDub plug-ins.

Hmm... :hmm:

On paper, I agree with you. In fact, in practice, I'd probably agree as well.

But I've seen some video that hates composite>s-video switching ... just not in a few years now, and not something I can re-test now. So I think I'd restate it as "never crossover" to be safer. When you see an issue, don't crossover. Stick to the path from source to output. I've seen it matter.

Reading though the forum about how a line TBC clean up the lines in the picture and Frame Synchronizers clean up the signal suggests that a Frame Sychronizer is only needed in specific situations? If this is true, how do you determine when a Frame Synchronizer is necessary?

It seems that every thing that is added to the chain creates a new potential for issues, so in a sense, the shorter the cable, the less devices, everything minimal to the capture card seems to be best unless the image is significantly improved? This philosophy suggests that you only need decent cables and a line (built in) tbc or ES15 TBC and a capture card. So the Frame Synchronizer or Processing amplifier are likely just for videos with an obvious problem?

Frame sync is useful when you encounter any type of signal that the capture card is unable to decode on its own, which would lead to frame drops and possible A/V desync. You're not likely to encounter this when capturing retail tapes like the Titanic example, setting aside the whole Macrovision thing.

This philosophy of cutting out the middleman doesn't really apply when you're capturing the SDI output. The DPS is effectively acting as your "capture card", with the digitized signal just being shuttled through the Kona to your hard drive. (This is assuming the Kona presents the data as-is. I have seen an SDI card add further processing to the video, with no neutral setting for its internal digital proc amp.)

That said, different capture cards can produce better or worse image quality, and the same goes for DPS vs Kona digitization. The only way to really know is to compare.

No.

Analog video signals are chaotic. Capture cards (and DVD recorders, etc) do not understand chaos. They want to digitize a clean-timed signal. This is why all DVD recorders have some sort of frame sync on input (though not necessarily frame sync TBC). All capture cards need some sort of signal purification as well.

You always need one. It's not optional.

It can. But you need to solve it, not quit.

Also not true. Certain pieces of hardware need to exist in a workflow. Without them, a shorter workflow will look worse.

Not everything is about image. It's also about signal integrity. Sometimes audio hardware is also desired.

No.

The ES10/15 recorders also have unwanted side effects, so these should only be used sparingly on an as-needed basis. The ES series were terrible DVD recorders. But these were great restoration tools intended to trade errors, when used as passthrough (a rare ability that I discovered in 2005). Tearing is removed (yay!), but it is replaced by overaggressive NR and posterization. It's a good trade. But it's not something you'd want all the time on all video, if tearing is not present on the source.

... which is basically ALL capture cards.

Not true. Retail tapes vary wildly, and are just as chaotic as a homemade tape. It's all just VHS. In fact, some retail signals can be worse than homemade, depending on the condition of the recording equipment and tape grades.

Yep. And I'm not sure that SDI (or the DPS) is needed. In fact, I;d say SDI is undesirable.

The Leitch DPS is a Frame Synch Earlier you said it was needed then later you say it is not? Are you suggesting a different Frame Synchronizer? As I understand all Frame Synchronizers convert to digital, so why go back to analog then back to digital? I can try some captures with the Kona LHI without the DPS.

AJA tech support said the Kona card does well with shaky signals and that he said you needed a TBC but that a Frame Synchronizer is not required.

For SDI = no.
As TBC = yes.

He's saying nonsense. That's not uncommon from a tech support desk.

Using SDI offers the benefit of avoiding an extra D-A-D step. Capturing the analog output instead offers no possible advantage.

But you have to trust the SDI capturing is of good quality.

And that assumes that SDI is needed or desired. It's not easy to work with, and is overkill for VHS source.

Most SDI workflows used in professional settings are for BetacamSP, U-matic, etc -- not VHS.

In fact, when SDI is used for VHS, it is usually the lower-quality conversions you see. I often see a VHS tape used in a documentary, and wonder how they manage to screw it up so badly. SDI is one strong possibility of where it started to go wrong.

SDI puts you on an NLE path, and that's not what you want for VHS ingest. VHS needs to be imported to an NLE after separate capturing and cleaning.

VHS is special because, compared to other formats, it was really crappy. :depressed: