Capture Cards with reasonably good Y/C separation (comb filter)

[pitsie]

I recommend to use S-video connection from your vcr.

For S-VHS tapes or also for plain VHS tapes?

I think I will put sharpness in the middle... AI does not seem to do much

I am having sync problems with audio on long captures... any ideas? Should I split into smaller chunks?

Thanks again!

[dannotto]

For S-VHS tapes or also for plain VHS tapes?
For both.

By default VirtualDub try to resync when audio is captured with the help of sound card. Are unnecessary for capture cards with integrated audio ADC.
With these settings and TBC On (jitter free) I don`t have such issue, even with long 4 hours tape or 12 hours long recording over RF input.
At Capture tick only > Timing > Do not resync between audio and video stream
Automatically disable resync when integrated audio / video capture is detected

[pitsie]

I think I have exactly that. I unchecked everything else but those options.

I don't think I have as much a drift but a constant issue (does not seem to increase with time)

My audio is a whisper ahead of my video from the start. How can I slightly increase the delay?

At "Audio latency determination" I have "Automatic" and at "Number of audio blocks to use at start" by default is 30, what if I increase that number?

:thanks:

[dannotto]

I don`t know. Try and see.
With these types of card sync problems are gone. During capturing can be some sync issue but not in captured file. This can be related with directshow filters. Also if you play lossless captured file, sync issue may apear because this format is not intented as playback because is huge and hard to render at real time. Maybe these can help: At Video > Crossbar
http://www.mediafire.com/file/bmtwnzdyfww/crossbar.png

[Ghitulescu]

On playback it act as Auto Sharpness. On recording mode act as tape calibration.
Made some test if you don`t believe me and also check the manual.
From manual When Ai in ON the picture sharpness cannot be ajusted
I know this, but there are also other things AI does, also on playing. Auto-sharpness is one of them, auto-noise is another one. The manual is for normal users, and do not forget that no company in the world would disclose you all the "secrets". You need probably to read some internal documents (classified) or patents. In short, AI is not AutoSharpness, but AutoSharpness is one of the effects AI has.

1.For S-VHS tapes or also for plain VHS tapes?

2. I think I will put sharpness in the middle... AI does not seem to do much

3. I am having sync problems with audio on long captures... any ideas? Should I split into smaller chunks?

1 - S-VHS is not very different from VHS, just a little bit more of B/W bandwidth and some other small technicalities (OT). S-Video is the recommended way to go.

2 - try first using the EDIT setting (both in Menu and on the front panel).
I'm not familiar with this model, as it's just a mixture of profi/consumer (was directed to small enterprises to be used by non-specialists; since the enterprises must pay for each tuner unit, as opposed to private persons, in most countries I know, they'd simply taken out the tuner and painted it in dirty white, otherwise it's just the regular NV-HS1000).

That you don't see a big difference is not an indication that there's no difference, you can judge this only after having it's output encoded (MPEG-encoders are very sensitive to noise and fake sharpness). For good tapes there's always difficult to find a difference anyway ;) ...

3 - no; you shouldn't have synch problems, but if you do, then you have problems with your capturing hardware. Fix this first.
The mechanics of Ag4700 still remains that of a consumer gear, and if the cassette is old or presents various mechanical defects that prevent it from running smooth, you may have synch problems within the deck.
I ordinarily do a "dust removal" run, which also acts as a FWD/REW step, then capture in one go.

[dannotto]

I know this, but there are also other things AI does, also on playing. Auto-sharpness is one of them, auto-noise is another one. The manual is for normal users, and do not forget that no company in the world would disclose you all the "secrets". You need probably to read some internal documents (classified) or patents. In short, AI is not AutoSharpness, but AutoSharpness is one of the effects AI has.


Are so secret that not even their effect (auto sharpness) can not be revealed.

All VCR brands have a fancy name for a minor tehnology - auto sharpness. At Panasonic first was AI then renamed to CVC (Cristal Viev Control). At the last generation of Panasonic vcr this feature cannot be turnet off.
With JVC vcr the name ist BEST, for Sony is Trilogic.
The sharpness affect the noise on image but is not true noise reduction. It is a very simple way to cut down the high frequency, the place where is a part of noise but also the detail. DNR is true noise reduction where the noise removed and in the same time the detail is preserved as much is possible. The noise can have have all type of frequency (low, midle and high) and sharpness control can only remove the high frequency and in the same time all the detail.

[Ghitulescu]

There are several other noise types on the band than simply that one that goes with detail/sharpness. AI builds a sort of hiytogram (you know it probably from photo) and optimizes it further. Its most evident effect remain the sharpness (which is immediate to most people), then comes the background noise which is not really noticed, finally comes other small improvements on colours, even less noticed.

So saying AI is auto-sharpness is like saying "an auto is Ferarri", because that would exclude the Trabis.

[dannotto]

There are several other noise types on the band than simply that one that goes with detail/sharpness. AI builds a sort of hiytogram (you know it probably from photo) and optimizes it further. Its most evident effect remain the sharpness (which is immediate to most people), then comes the background noise which is not really noticed, finally comes other small improvements on colours, even less noticed.

So saying AI is auto-sharpness is like saying "an auto is Ferarri", because that would exclude the Trabis.

A analogue VCR made in 1993/1994 is so smart to perform a histogram and analyse the background?! LOL
It`s hard to accept the truth. My dialogue about a dumb feature stops here.

[Undead Sega]

Hi guys! I'm so sorry for bringing this old thread up, but this caught my eye:

@dannotto

Can I download that test pattern from somewhere? I'd be interested to compare Pixelmagic PDI Deluxe(SAA7118) and Datavideo TBC-1000(composite->s-video) with your results.

Is this even possible? Because I am so sure I did this myself and it doesnt work! I was really looking forward to what the results would've been from my HS-1000 via. composite and outputting S-video from my TBC-1000 :(

[juhok]

Works for me. :) I don't have the pictures/urls at hand but composite -> s-video passthru with TBC-100(0) sucks.

[Undead Sega]

Really?

And really??? Can you probably elaborate abit on why you think it's that bad? Surely the separation should visually improve the image quality somewhat.

[juhok]

I think it's bad because I've experimented with it and seen it so. Not many devices work like they could ideally do and TBC-100(0) most definitely doesn't. HS-1000 outputs good s-video, I haven't seen a device that would output better PQ from it's composite output compared to s-video. I can make some samples later if I remember. Would you rather see HS-1000 or DVD-player as a source if I compare TBC-100(0) separation quality?

[Undead Sega]

Ahhh i see, you own a HS-1000 yourself?

So are we trying to say that the TBC-1000 Y/C Separation is not good at all?

And for now I would say DVD player as a source purely because the material can be of higher quality and thus would be easier to judge :)

[juhok]

Yup, I would say it's not good at all. With DVD-source it's easier to see the problems. But HS-1000 might prove more useful for your scenario. I try to do both when I have time.. really overloaded atm.

[Undead Sega]

Ohhh okay then, well that would really be appreciated! I look forward to your samples when you get the time :D

So, what would you say is a good Y/C separation device which is definitely not the TBC-1000 as you're saying?

[lordsmurf]

Yeah, don't use the TBC-1000 for y/c separation.
I would use into only for s-video.
That's true of pretty much any TBC, to be honest.

Good gear, BTW. :cool:

And good to see you juhok. It's been a while. :thanks:

[juhok]

Hey LS. Indeed it's been a while since I roamed the videoforums. I'm glad you're doing better. :-)

Undead Saga; I haven't personally seen a really good device and I've tried a lot of different ones, but I've never really chased that workflow much. Because I rarely need it, I've typically used the one that my capture device provides. The one in BrightEye 75 AD converter is the best I have, ie. "better than nothing" (nothing = like Blackmagic Design cards etc).

[Undead Sega]

Yeah, don't use the TBC-1000 for y/c separation.
I would use into only for s-video.
That's true of pretty much any TBC, to be honest.

Well I still find it kinda strange that my TBC-1000 doesnt do Composite -> S-Video the last time I tried it out. But since you're saying this yourself, what would you recommend?

Undead Saga; I haven't personally seen a really good device and I've tried a lot of different ones, but I've never really chased that workflow much. Because I rarely need it, I've typically used the one that my capture device provides. The one in BrightEye 75 AD converter is the best I have, ie. "better than nothing" (nothing = like Blackmagic Design cards etc).

Woah...are you trying to say the Black Magic cards are somewhat inferior in performance???

By the way, just to inform you guys here, I also now have my HVR-4000 TV card installed, I dont know if this make a difference to any of you lot.

[juhok]

Oops, I don't mean Blackmagic cards are inferior to TBC-1000 in this regard, they are *better*. But they have basic dumb composite input. BM doesn't introduce extra softness etc like TBC-1000 but it's still not a very good. Not something you seem to be looking for here.

[Undead Sega]

I see, but with the TBC-1000 what causes this softness in the picture? I didnt think it was that evident when using S-Video.

Another piece of equipment that I probably want to hook up as well is my newly received LD-V8000 and was thinking of passing that through my TBC-1000 too.

[Undead Sega]

Hello juhok, how are you? any luck on those tests by any chance?

Also, I gathered the time to hook my LD-V8000 and (for some people strangely) TBC-1000 passing through via. composite input and S-video output, and it does work! In addition, I noticed a very minor improvement, this being where bry subtle rainbowing effects that appeared or were shimmering on the subtitles were somewhat eliminated.

Now this isnt a sciencetific test as it was very hard to tell if there was an overall (minor) improvement in the video which a doubt but seeing that result does make me wonder, however it did very much compare to having the player connected via. composite through the TBC-1000 input and output and having the 3D Comb Filter turned on on my HDTV. I suppose the best way is to capture them using my HVR-4000 TV/Capture card.

Anyways, I thought I would share that result, iw ill trying it again on a better laserdisc (Star Wars Trilogy Special Edition) that is coming very soon :)

[juhok]

Hmm, we never said comp->s-video doesn't work, just that the quality sucks. :)

Anyway, got queue of work but I managed to squeeze this together.

First I had to disassemble HS1000 upper drum and clean it and the rest of the tape path because it seems once again customer had sent me crap which muddied the deck. I hope there's not too many dropouts in the picture now.

Hardware used:
- Pana HS1000
- Modified TBC1000 (stock connectors replaced by professional BNC, internal distribution amplifier bypassed, replaced internal crappy wire with RG174A coax)
- Ensemble Designs Bright Eye 75 AD
- HQ cables

HS1000 -> BE75 (s-video):
http://www.siluriformes.net/doom9/BE75_HS1000_svideo.png

HS1000 -> TBC1000 -> BE75 (s-video all the way)
http://www.siluriformes.net/doom9/BE75_TBC1000_HS1000_svideo.png

HS1000 -> BE75 (composite, BE75 comb filter on)
http://www.siluriformes.net/doom9/BE75_HS1000_composite.png

HS1000 -> TBC1000 -> BE75 (composite to TBC1000, then s-video to BE75)
http://www.siluriformes.net/doom9/BE75_TBC1000_HS1000_composite-svideo.png

Ignore the chroma saturation difference between direct VCR->BE75 and VCR->TBC1000->BE75, that's another story.

I've ranked pictures from best to worst (IMHO). I would rank direct composite with BE75 comb filter to 2nd but it has some weird artifacts at times which I don't like. S-video from HS1000 rocks. It's even more clear with playback. I can upload samples if you want, about 1GB total.