ADVC-300 Questions and Concerns

[SixFiftyThree]

Hi all,
I know this product has been discussed to death over the years here, but bear with me. To clarify, this isn't a thread about capping to DV or not. I've made an educated choice according to my workflow needs and budget, which offers me high bitrate, manageable filesize and editability, in 4:2:0 colorspace (PAL). With that settled, I'd like some help deciding on the capture device.

My first choice was the ADVC-110, but I believe the ADVC-300 could be better, depending on the feedback I get here. Despite the negativity towards its NR, which I won't use, and the confusion over its TBC, which I don't need, I'm drawn to it for its other features. Namely video and audio levels adjustments, as I understand it's best to correct levels before A/D. I should mention that for the sake of a short capture chain, I'd prefer to use the 300 over a seperate proc amp, and insert only my Datavideo full frame when needed.

But I do have my concerns. First is the 300's always-active line TBC. I've seen video examples of it and I'm convinced it works - but my decks do it better. What I'm wondering is whether the 300 will play nice with a signal that's already been LTBCed by a VTR. Is feeding one LTBC into another a bad idea? If anyone's been in this situation, please share your findings.

Next is the auto gain. As I'll be using a vectorscope to ensure levels are within spec, this is more of a curiousity, but could be handy for some quick and dirty captures. I've already read tests on the audio AGC, but I'd like to know how the video AGC performs, particularly if it's prone to creating out of range values. Finally, white step, white peak and black expansion look good in writing, but how do they hold up in practice? How effective are these functions, along with the rest of the proc amp controls?

Hope you guys can help me out, thanks for your time.

[manono]

First is the 300's always-active line TBC. I've seen video examples of it and I'm convinced it works - but my decks do it better.

Either it doesn't work or is too weak to be of any use at all. I'd like to see some before and after video samples from someone that says it works. I use the TBC in my PAL Panasonic NV-HS860 VCR and the ADVC-300 doesn't disrupt it or nullify it at all.

Next is the auto gain. As I'll be using a vectorscope to ensure levels are within spec, this is more of a curiousity, but could be handy for some quick and dirty captures. I've already read tests on the audio AGC, but I'd like to know how the video AGC performs, particularly if it's prone to creating out of range values. Finally, white step, white peak and black expansion look good in writing, but how do they hold up in practice? How effective are these functions, along with the rest of the proc amp controls?

I don't trust or use most of that stuff. I'll do some test caps to adjust the brightness and contrast, and that's all I use from its proc amp. The rest I do in AviSynth later on. I'm not much help with that part of your questions, I'm afraid (and maybe not the first part either).

[SixFiftyThree]

Either it doesn't work or is too weak to be of any use at all. I'd like to see some before and after video samples from someone that says it works.

Here's someone's clip of an ADVC-300 LTBC test: http://www.youtube.com/watch?v=ZPE8cOvgZFE
My JVC and Panasonic decks would give better results, but it's evidently correcting the h-sync.

I use the TBC in my PAL Panasonic NV-HS860 VCR and the ADVC-300 doesn't disrupt it or nullify it at all.

That's good news, but I wonder if this is due to some Panasonics having a different kind of LTBC, specifically full field/multi-line. I believe this only exists in the AG-1980 for NTSC models, but that model comes in different flavours in the PAL region; closest is the NV-FS200 but there's others that work similarly. Perhaps someone with a technical understanding of line TBCs could pitch in.

I don't trust or use most of that stuff. I'll do some test caps to adjust the brightness and contrast, and that's all I use from its proc amp. The rest I do in AviSynth later on. I'm not much help with that part of your questions, I'm afraid (and maybe not the first part either).

Not sure what you mean by not trusting those functions, unless you mean the AGC, which is understandable. I realize the importance of capping with correct levels, and I'm particularly interested in controls like white step to recover more details, which would otherwise be impossible in post. If you or anyone else has experimented with those settings, I'd appreciate it you could comment on their effectiveness.

[manono]

That was an interesting video. Thanks for pointing it out. So I guess the ADVC-300's TBC does do something. The kinds of things I see without the Panasonic's one turned on look different from that, though. Those are really big wobbles it's correcting, while mine are on a smaller scale and the ADVC-300 does nothing to them. I don't know enough about it (nothing, really) to say that guy's source being a Hi8 camera is the reason, or what.

As for testing the proc amp settings, no, I haven't. I'd much rather do it manually in AviSynth.

[lordsmurf]

The ADVC is so weak to essentially be useless.
DVD recorders and S-VHS VCRs do a much better job.

This video just further proves that.

[SixFiftyThree]

The ADVC is so weak to essentially be useless.
DVD recorders and S-VHS VCRs do a much better job.

This video just further proves that.

With all due respect, I think the video proves it can stabilize the picture to some degree. That's beside the point though. My question was how two line TBCs interact with each other, regardless of strength.

Once again, I'd appreciate it if someone could post some feedback on the ADVC's proc amp and other functions as mentioned in my original post.

[lordsmurf]

My question was how two line TBCs interact with each other, regardless of strength.

Whatever the first one does is permanent. The second one won't have any effect, aside from potential signal sync (non-visual improvements).

That's why S-VHS + external TBC are always suggested. The VCR's TBC handles the jitter, the second TBC handles signal sync.

In some cases, you disable the VCR, and passthrough to another device (DVD recorder, etc), and then pass it to the external TBC.

You can't stack TBCs for visual improvements. Doesn't work that way.

The ADVC proc amp is one of those pathetic little color adjusters, similar to the AVT-8710/CTB-100 proc amp.

I've been doing this for more than a decade now, just FYI.

[SixFiftyThree]

And I've been following your advice closely for over a year now, just FYI
I know you're not fond of the ADVC units or DV format, but it suits my needs.

As for the ADVC's line TBC, it's not my intention to stack LTBCs to improve quality, but rather avoid quality loss from unnecessary processing. As long as it doesn't disrupt the already corrected signal from my high-end decks, then I couldn't care less what it does. Given that both line and full frame will be applied before input to the ADVC, there would already be sufficient correction. Looks like there's no problem here.

Also, the AVT-8710 has far less controls compared to the ADVC-300, which has both audio and video levels adjustments, including specific controls for black and white values. It'd be nice to know in detail how these work and how well they work. But for basic levels setup with a vectorscope, it seems like it'd do nicely. If you have more info to share then I'm keen to hear it.

[lordsmurf]

Do you have the ADVC-300 manual?
That might help.

[SixFiftyThree]

Do you have the ADVC-300 manual?
That might help.

I do have the manual, so I know the technical description of all the functions, but would like some detailed feedback from those who personally use the device and those particular features. As I said they look good in writing, but how they hold up in practice will decide whether I choose it over the ADVC-110.

[davideck]

In some cases, you disable the VCR, and passthrough to another device (DVD recorder, etc), and then pass it to the external TBC.

A DVD Recorder will provide the same frame synchronization as an external TBC. Unless copy protection is an issue, there is no need for an external TBC after the DVD Recorder.

[SixFiftyThree]

In some cases, you disable the VCR, and passthrough to another device (DVD recorder, etc), and then pass it to the external TBC.

A DVD Recorder will provide the same frame synchronization as an external TBC. Unless copy protection is an issue, there is no need for an external TBC after the DVD Recorder.

It'd still be a safer bet using a TBC for non-copyright material though, in case a card falsely detects MV due to tape errors. Some of those ATI models are quite touchy I hear. But I can see the merit in what you're saying, it'd be the ideal method to avoid more signal degradation in this scenario. In any event, I don't believe a DVD recorder's frame sync would be as reliable as a dedicated TBC. Although the AVT-8710 is a piece of crap...

[JVRaines]

Although the AVT-8710 is a piece of crap...

I have to disagree. As a simple frame sync it does a great job for the price. I routinely run U-matic through it for horizontal and vertical stability. The proc amp isn't much, true, and that's why I have a real amp from SignVideo. This combo with ADVC110 has got me through thousands of hours of NTSC material.

I tried the ADVC300 for its processing abilities but wasn't impressed and returned it.

[davideck]

It'd still be a safer bet using a TBC for non-copyright material though, in case a card falsely detects MV due to tape errors...In any event, I don't believe a DVD recorder's frame sync would be as reliable as a dedicated TBC. Although the AVT-8710 is a piece of crap...

...and the DataVideo softens the image. As a preservationist, I would much rather wait for a particular tape to fail before subjecting it to another analog-digital-analog conversion. I've been lucky, I guess. I have not had MV issues, except for false detections on a JVC DVD Recorder.

[SixFiftyThree]

Although the AVT-8710 is a piece of crap...

I have to disagree. As a simple frame sync it does a great job for the price.

Then you have an older unit that works as intended - the newer batches are blown. Nothing but a frustrating experience, personally.

I tried the ADVC300 for its processing abilities but wasn't impressed and returned it.

Some elaboration on this would be appreciated, unless you're talking about the NR.

...and the DataVideo softens the image. As a preservationist, I would much rather wait for a particular tape to fail before subjecting it to another analog-digital-analog conversion. I've been lucky, I guess. I have not had MV issues, except for false detections on a JVC DVD Recorder.

I agree to some extent, and the DV format has its advantages here. With foolproof sync, and providing its a stable tape without MV, the TBC could be skipped in favor of just one A/D conversion. But in favor of efficiency, it's better to keep the TBC in the chain to fall back on.

The best results would be a TBC with SDI out, but that's easily over 1k by itself, and then there's still the SDI card to buy.

[davideck]

Have you considered capturing to 1394 with a mini DV Camcorder? My JVC unit provides exceptional TBC performance, and with a single analog to digital conversion, it doesn't get more transparent than that.

[JVRaines]

I tried the ADVC300 for its processing abilities but wasn't impressed and returned it.

Some elaboration on this would be appreciated, unless you're talking about the NR.

Yeah, the NR wasn't worth the price and I didn't like having to keep the TBC on all the time. Plus the TBC looked a touch softer than my 8710, so with the same encoder chip, I decided to stick with the ADVC110.

[lordsmurf]

...and the DataVideo softens the image.

No, the TBC doesn't do it. The VP-299 distro amp in some TBC-1000 units is the actual culprit. And even then, it's not a guaranteed error, but a random one. You can bypass it and just use the actual TBC-100 card inside the TBC-1000. Some guides and example photos are in the digitalFAQ.com forums.

[juhok]

I've run tests with ADVC-300 controls but I don't have the notes or samples anymore. As far as I can remember, the adjustments were done after AD and had no benefit compared to just doing it in editing software. ADVC-300 is so much overprized compared to what it delivers that IMHO you'd be better off getting real capture card and capture to DV using cedocida codec if DV is what you want. I've used the cheaper ADVCs too and AFAIK there wasn't really any difference in the results, sans digital proc amp (if you have Frame Sync before the ADVC like you said you do). And yes, TBC-1000 does nothing to soften the image when working properly.

[davideck]

...and the DataVideo softens the image.

No, the TBC doesn't do it. The VP-299 distro amp in some TBC-1000 units is the actual culprit. And even then, it's not a guaranteed error, but a random one. You can bypass it and just use the actual TBC-100 card inside the TBC-1000.

The TBC-3000 softens the image as well.

[juhok]

Maybe TBC-3000 shares some of the design flaws of the VP-299 In any case, TBC-100 and TBC-3000 are different PCBs and not comparable.

[davideck]

The TBC-3000 is basically two TBC-100s that share a common reference output timing. It uses many of the same ICs as the TBC-100.

Some people notice the softening, others do not. The source material and playback VCR have a lot to do with it. My tapes were mostly all higher detail S-VHS.

[juhok]

TBC-3000 also has more stuff added to the signal path.

But in any case you're right. There's some softening, atleast with my current setup. I don't remember it being like this last time I tried, but that's no excuse. I see the effect. This is an very old specimen, second hand from eBay, and modded to bypass the dist amp.

Test was made with TVblink test DVD, basic sony DVD player, modded TBC-1000 and Blackmagic Design Studio capture card. Captured to lossless file and upsampled to png.

Comparing Blackmagic quality against digitally decoded file is not meaningful because DVD player DA plays it's role. The file is included for curiositys sake.

TBC-1000 shows some ghosting that wasn't there in my last test. Might bad soldering, bad cable, bad something else. Have to examine it further.

edit: From top to bottom; mpeg2, BM direct, TBC-1000

[davideck]

Thanks for posting the images!

TBC-3000 also has more stuff added to the signal path.

Are you sure? The proc amp controls on the TBC-3000 simply allow the user to modify the setup registers in the Analog I/O devices that are otherwise always defaulted to unity by the TBC-100 firmware. No extra components in the signal path are required.

There's some softening, atleast with my current setup. I don't remember it being like this last time I tried, but that's no excuse. I see the effect. This is an very old specimen, second hand from eBay, and modded to bypass the dist amp...TBC-1000 shows some ghosting that wasn't there in my last test. Might bad soldering, bad cable, bad something else. Have to examine it further.

The ghosting may be a result of improper 75 ohm termination or cable impedance. It would be very interesting to restore the dist amp and compare that to your other captures.

[juhok]

Are you sure? The proc amp controls on the TBC-3000 simply allow the user to modify the setup registers in the Analog I/O devices that are otherwise always defaulted to unity by the TBC-100 firmware. No extra components in the signal path are required.

I'm not sure. PCB looks more populated with more ICs and components than 2x TBC-100 would warrant. Without service manual I cannot say more.

The ghosting may be a result of improper 75 ohm termination or cable impedance. It would be very interesting to restore the dist amp and compare that to your other captures.

Sadly my mod is not very modular and dist-amp is recycled long time ago. I tried different cables, same result. Problem might be in the connectors which seem flimsy from day one. Maybe it's time to replace that annoying S-video connector with 2 BNCs + custom cable. Luckily this phenoma is not easily observed outside this kind of test chart (done a lot of VHS with this specimen).

[lordsmurf]

I have a standalone VP-299 and TBC-100, and can try testing at some point in time.
But it won't be soon, due to scheduling.

[SixFiftyThree]

I've run tests with ADVC-300 controls but I don't have the notes or samples anymore. As far as I can remember, the adjustments were done after AD and had no benefit compared to just doing it in editing software.

That's great info, thanks. I was hoping to recover some detail with the levels adjustments, but if the proc amp sits after A/D then it'd be no better than s/w as you said. In fact, Enosoft has a s/w proc amp that works in realtime for DV acquisition, with vectorscope and waveform included. Are there any good proc amps available new these days? I'm sick of buying used gear.

ADVC-300 is so much overprized compared to what it delivers that IMHO you'd be better off getting real capture card and capture to DV using cedocida codec if DV is what you want.

If I'm capping to DV then I prefer dedicated DV hardware such as the ADVC which has the added benefit of being bi-directional. I also have an ATI 600 to use if lossless is necessary. Not that I consider either of those as real capture cards. Not compared to something like an AJA Kona. But that's more than I'm willing to invest.

Anyway, I've not read anything to convince me that the 300 is a good investment, so I'll be passing on it for the 110 instead.

[lordsmurf]

Are there any good proc amps available new these days?

No.
SignVideo recently stopped manufacturing the PA-100.

[SixFiftyThree]

Are there any good proc amps available new these days?

No.
SignVideo recently stopped manufacturing the PA-100.

That's unfortunate, I suppose I'll just have to make do with software. Thanks for the quick reply.

[juhok]

Replacing S-Video/DIN connectors with good quality BNC connectors and using BNC->S-Video cable removed the ghosting seen in post #23.

edit: Optimally the other end would be BNC too. I'll propably mod some of my VCRs this way in the future.