Quantcast Myth or fact: timebase corrector (frame synchronizer) should always be used? - digitalFAQ Forum
Go Back    Forum > Digital Life > Tech Myths, Misinformation

Reply
 
LinkBack Thread Tools
  #1  
08-11-2014, 02:48 PM
premiumcapture premiumcapture is offline
Free Member
 
Join Date: Dec 2013
Location: Boston, MA
Posts: 581
Thanked 68 Times in 62 Posts
I have recently started to question the idea of always using a TBC. In addition to my personal experiences, I cite the following two articles:

http://www.studio1productions.com/Articles/TBC.htm
http://www.htforum.nl/yabbse/index.p...=27890.0;imode

Perhaps I am misunderstanding terminology here, but it seems that a time base corrector corrects distorted and wavy video, including flagging. This would mean that what is commonly referred to as a 'line' time base corrector is in fact just a regular time base corrector.

A frame synchronizer, on the other are what the TBC 1000 and AVT 8710 perform - they regulate the framerate for better capture.

Here's the question here - if a TBC as included in a VCR and a frame synchronizer perform different tasks, and VirtualDub resyncs video and audio on the fly, then technically speaking, shouldn't a TBC in a VCR be necessary, while the external units should not be? Isn't VirtualDub performing the same job that these units are?

Perhaps I am confusing terms here, but it doesn't make sense to me that an external synchronizer should be necessary if the software is synchronizing on its own, and perhaps even doing a better job.
Reply With Quote
Someday, 12:01 PM
admin's Avatar
Ads / Sponsors
 
Join Date: ∞
Posts: 42
Thanks: ∞
Thanked 42 Times in 42 Posts
  #2  
08-11-2014, 03:20 PM
lordsmurf's Avatar
lordsmurf lordsmurf is offline
Site Staff | Video
 
Join Date: Dec 2002
Posts: 8,274
Thanked 1,370 Times in 1,205 Posts
99%+ of the time, it's fact -- TBC is needed for analog video.

Those articles you've linked to have some errors and missing information. For example, that second link is confusing a frame sync with a gen lock -- both of which can be TBC functionality. But they're not the same.

VirtualDub is just software. It cannot do anything.

I'll explain more later. Busy right now.

- Did my advice help you? Then become a Premium Member and support this site.
- Find television shows, cartoons, DVDs and Blu-ray releases at the TVPast forums.
Reply With Quote
  #3  
08-12-2014, 04:23 PM
msgohan msgohan is offline
Free Member
 
Join Date: Feb 2011
Location: Vancouver, Canada
Posts: 888
Thanked 197 Times in 171 Posts
Quote:
Originally Posted by premiumcapture View Post
Perhaps I am confusing terms here, but it doesn't make sense to me that an external synchronizer should be necessary if the software is synchronizing on its own, and perhaps even doing a better job.
VirtualDub is forced to deal with the capture hardware you're using, which can generally only adapt to a signal that's mildly outside of the accepted range. The external units are designed to have a wider tolerance for signals that deviate from the expected standards.

Confusion of "Time Base Corrector" vs. "Frame Synchronizer" isn't something that originated with hobbyists; it's all over the pro world. Lordsmurf has always preferred to use terminology based on the apparent effect of a given unit rather than its advertised or technical name; if you go back through the VideoHelp archives there are some informative debates between him and Davideck.

Dana Lee (Ryerson University) - Time Base Correctors and Frame Synchronizers

Reply With Quote
  #4  
08-27-2014, 05:12 AM
lordsmurf's Avatar
lordsmurf lordsmurf is offline
Site Staff | Video
 
Join Date: Dec 2002
Posts: 8,274
Thanked 1,370 Times in 1,205 Posts
I want to partially revisit this right now. But later on, I want to expand the TBC sticky even further to better clarify things.

RE: What is a TBC? Time Base Correction for Videotapes

The video above is excellent for explaining some quick tidbits about the technical side of TBC. Everything mentioned there is accurate. The only problem with this video is that implementation in devices is not always so precise. Hence all the confusion and disagreements about what a TBC is.

As msgohan stated, this happens professionally, nut just among the hobby crowd.

You have several technical items:
- a line TBC
- a multi-line TBC, aka full-field TBC, aka infinite window TBC
- a frame sync
- a frame sync TBC

From there, things quickly get murky.

The sort of buffer attached to the TBC can help define what it does as well.

It's location in the workflow path can help define what it does -- and that includes inside components like cameras, DVD recorders, and VCRs

Many more can be crippled to allow error to pass (anti-copy like Macrovision). Or rather, not "crippled" but not built 100% properly to begin with. Thus, not really a TBC or even a non-TBC frame sync. It's a quasi/pseudo TBC-like device, which has signal leaks. They fail about 50% of the time. These are also cheaper to make, since most have poor buffers, and missing chips/components. Some DVD recorders, most DV boxes, and other items fit here. A true TBC is at least $100 or more to build, so it should be obvious that the prices of those cheap devices do not have one. (Note: The ADVC-300 is expensive because of the poor NR and proc amp, and because you're buying the Canopus name. So no TBC there either.)

Most DVD recorders just have basic frame syncs on input. Few have the more unusual insides.

Various errors interact differently with different types of TBCs or non-TBCs.

That's why explaining things is best done with examples by device type.

For example,

S-VHS VCRs have line TBCs or multi-line TBCs, with decent (but still somewhat small!) 2mb-4mb buffers, and can clean up about 90% of the errors. These have no frame sync, either TBC or non-TBC type, and thus do nothing for this. Capture cards grab video in frames, so when the signal is not frame synced, errors can occur. This fix is to add the external frame sync TBC after the VCR.

So: FACT.

- Did my advice help you? Then become a Premium Member and support this site.
- Find television shows, cartoons, DVDs and Blu-ray releases at the TVPast forums.
Reply With Quote
The following users thank lordsmurf for this useful post: Winsordawson (08-28-2014)
  #5  
08-28-2014, 09:30 PM
premiumcapture premiumcapture is offline
Free Member
 
Join Date: Dec 2013
Location: Boston, MA
Posts: 581
Thanked 68 Times in 62 Posts
So it sounds like a TBC is necessary due to the nature of VHS, with such hardware only being included in VCRs, camcorders, and DVD recorders to a degree. These devices affect how the original picture quality is presented and captured.

I still question the necessity of the frame sync, however. If a tape is sending garbage in, I don't see why software couldn't perform the same task as the TBC-100/0 and AVT perform. These devices don't perform the line corrections necessary for stable playback, and capture rate can be configured through software. If there is no usable picture sent by the VCR, I don't quite understand what the devices above do to correct those errors.

I realize that broadcast houses use these devices on a regular basis, and switching between one source and another can cause timing errors that need to be fixed by frame synchronizers to guarantee continuous, synced broadcast, but when capturing one source, I don't quite see the need. What am I missing?
Reply With Quote
  #6  
08-29-2014, 08:42 AM
sanlyn sanlyn is offline
Premium Member
 
Join Date: Aug 2009
Location: N. Carolina and NY, USA
Posts: 3,396
Thanked 1,066 Times in 889 Posts
I can't offer a scientific, by-the-numbers explanation for the following, but it is one example from my own experience.

A nephew gives me a VHS tape of a 3-hour football broadcast recorded off cable on a decent Panasonic VCR. Had a nice sharp image even at 6-hour speed, but the usual EP noise problems, some chroma shift, a little color bleed, and spots all over the place. The problem with the tape was that the nephew and enthusiastic roommates kept using the fast-foward/stop and replay features, physically bruising the tape at many points. I didn't have my AG-1980 or other tbc unit around, so I played this tape on a Panasonic PV-S4670 with s-video out. What I got at many points was a variety of effects including upper border tearing, ripples, dropped frames, frame hopping, dropouts, blue screens, angular stretching, etc., etc.

Hooked this up to a Toshiba RD-XS34 for its line tbc pass-thru. It solved a lot of problems. The tearing was gone and most of the ripples except some of the worst that were due to tape damage, and image distortion and line wiggles were cleared. I've used that pass-thru on copy protected tapes, and it always defeated Macrovision. But this time I still got some dropped frames, a few blue screens, and some frame hops and stutter. Ultimately I got smooth playback after adding the AVT-8710 to the circuit.

So my conclusion from this experience is that the pass-thru solved some in-frame problems, but there were frame-by-frame discrepancies that the unit's frame sync circuit didn't address. The pass-thru got a few frame level problems correct, but not all of them. So I think lordsmurf's description of some of the shortfalls of pass-thru units applies here.

Couple of years later I played this tape on an AG-1980 without the pass-thru unit. I had similar frame-level results and had to add the AVT to get better frame sync. The AVT also fixed the problem of audio momentarily and slightly losing exact sync or giving some occasional wow or sputter in some badly damaged segments. To test this, I used an ES15 for pass-thru instead of the AVT, to compare frame sync problems. It was no contest: the AVT did better work than the pas-thru.

I haven't tried this with decent tapes that haven't been so badly abused. But I did have one tape with Macrovision that kept fooling the pass-thru device so that I had a white 'flash" every now and then, telling me that Macrovision was still doing its dirty work. Had to add the AVT for that tape.

My AVT is the original I purchased in 2004. Still works. Knock wood.
Reply With Quote
  #7  
08-30-2014, 04:07 PM
lordsmurf's Avatar
lordsmurf lordsmurf is offline
Site Staff | Video
 
Join Date: Dec 2002
Posts: 8,274
Thanked 1,370 Times in 1,205 Posts
Quote:
Originally Posted by premiumcapture View Post
I still question the necessity of the frame sync, however.
What am I missing?
Well, stop doubting.

What sanlyn has posted is a pretty good example of what happens with a frame sync TBC vs. just using other devices (including non-TBC frame sync). Analog signals are just dirty, and this is what is used to fix them.

- Did my advice help you? Then become a Premium Member and support this site.
- Find television shows, cartoons, DVDs and Blu-ray releases at the TVPast forums.
Reply With Quote
  #8  
08-31-2014, 05:35 PM
msgohan msgohan is offline
Free Member
 
Join Date: Feb 2011
Location: Vancouver, Canada
Posts: 888
Thanked 197 Times in 171 Posts
Quote:
Originally Posted by premiumcapture View Post
If there is no usable picture sent by the VCR, I don't quite understand what the devices above do to correct those errors.
Rewording what I said before: what one device considers an unusable signal can be recovered using hardware designed to handle such garbage, up to the second device's own threshold.

This short piece has some diagrams showing synchronization signals that I think most devices would consider to be noise. If a capture device can't recognize a frame exists because it can't decode the timing, it will be dropped no matter your software settings about frame rate.


Reply With Quote
  #9  
10-07-2014, 10:25 PM
jazz57 jazz57 is offline
Free Member
 
Join Date: Jan 2012
Posts: 7
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by msgohan View Post
Rewording what I said before: what one device considers an unusable signal can be recovered using hardware designed to handle such garbage, up to the second device's own threshold.

This short piece has some diagrams showing synchronization signals that I think most devices would consider to be noise. If a capture device can't recognize a frame exists because it can't decode the timing, it will be dropped no matter your software settings about frame rate.
What I find interesting about this is Analog Devices sells chips containing their new ADLLT circuitry described in that paper. The chips, called SDTV decoders, cost only a few dollars. Pairing one of these with a matching encoder should make a TBC to beat all those usually discussed here. I wonder if anyone has tried it.
Reply With Quote
  #10  
10-08-2014, 01:06 AM
msgohan msgohan is offline
Free Member
 
Join Date: Feb 2011
Location: Vancouver, Canada
Posts: 888
Thanked 197 Times in 171 Posts
Analog Devices' offerings are no replacement for the TBCs suggested here, since they are made to respond to Macrovision.

For the chips that include ADLLT (like ADV7280), they describe the feature as "mini TBC functionality." In theory a person could simply buy one of their Evaluation Boards with analog input & output (Product Category: Video Decoders) but the price is more than a few dollars ($152-$1100+).

For the ADV7840 present in my AVR-890, they refer to "Advanced time-base correction (TBC) with frame synchronization" and make no mention of ADLLT or mini-TBC.
Reply With Quote
  #11  
10-08-2014, 02:45 AM
jazz57 jazz57 is offline
Free Member
 
Join Date: Jan 2012
Posts: 7
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by msgohan View Post
Analog Devices' offerings are no replacement for the TBCs suggested here, since they are made to respond to Macrovision.

For the chips that include ADLLT (like ADV7280), they describe the feature as "mini TBC functionality." In theory a person could simply buy one of their Evaluation Boards with analog input & output (Product Category: Video Decoders) but the price is more than a few dollars ($152-$1100+).

For the ADV7840 present in my AVR-890, they refer to "Advanced time-base correction (TBC) with frame synchronization" and make no mention of ADLLT or mini-TBC.
From what I was able to to glean from the datasheets "responds to Macrovision" means they detect it but that doesn't necessarily mean you can't use the chip to eliminate it. Just ignore the detection flag in your design. It can also detect a bunch of other stuff like closed-captioning.

By the way I found a unit selling on eBay called the DP-X7001 which does just what we're talking about---a decoder chip followed by an encoder. This one, though, is based on a Phillips SAA7114H from the year 2000. Datasheets says it can handle "weak and distorted" signals. They even give a description on their eBay page.

Too bad those evaluation boards aren't less expensive. I'd hate to try one only to discover poor performance.
Reply With Quote
  #12  
10-08-2014, 10:12 AM
msgohan msgohan is offline
Free Member
 
Join Date: Feb 2011
Location: Vancouver, Canada
Posts: 888
Thanked 197 Times in 171 Posts
The SAA7114H itself is nothing extraordinary. It's used in the AVT-8710 and others. There are capture cards that use it, though its predecessor is the one that was in tons of cards for years. I believe configuration matters a lot, though.
Reply With Quote
  #13  
10-08-2014, 01:53 PM
jazz57 jazz57 is offline
Free Member
 
Join Date: Jan 2012
Posts: 7
Thanked 0 Times in 0 Posts
Quote:
Originally Posted by msgohan View Post
The SAA7114H itself is nothing extraordinary. It's used in the AVT-8710 and others. There are capture cards that use it, though its predecessor is the one that was in tons of cards for years. I believe configuration matters a lot, though.
That's too bad. I wish someone would put a modern chip in one of these devices. From the paper you cited it looks like the newer AD devices do an amazing job cleaning up the sync signals--some of those waveforms looked unrecognizable yet the Analog Devices chip was able to generate a usable signal. I'm particularly impressed by how it solved the "top curl" problem, usually referred to as "flagging" here on the forum.
Reply With Quote
  #14  
03-20-2017, 03:37 PM
msgohan msgohan is offline
Free Member
 
Join Date: Feb 2011
Location: Vancouver, Canada
Posts: 888
Thanked 197 Times in 171 Posts
Quote:
Originally Posted by msgohan View Post
Confusion of "Time Base Corrector" vs. "Frame Synchronizer" isn't something that originated with hobbyists; it's all over the pro world. Lordsmurf has always preferred to use terminology based on the apparent effect of a given unit rather than its advertised or technical name; if you go back through the VideoHelp archives there are some informative debates between him and Davideck.

Dana Lee (Ryerson University) - Time Base Correctors and Frame Synchronizers

Another YouTube video: Time Base Corrector vs. Frame Synchronizer

Reply With Quote
Reply




Similar Threads
Thread Thread Starter Forum Replies Last Post
Audio Delay from Frame Synchronizer TBCs? Cingular Capture, Record, Transfer 14 06-02-2016 02:03 PM
Dropped frames - What difference does frame synchronizer/TBC make? premiumcapture Capture, Record, Transfer 6 08-12-2014 10:37 PM
Manual for the AVT-8710 TBC/Frame Synchronizer [DOWNLOAD] DeeSeven Video Hardware Repair 0 10-29-2013 02:57 PM
AVT-8710 Multisystem TBC / Frame Synchronizer with Color Corrector, for sale [SOLD] ieee488 Marketplace 10 10-12-2010 12:27 PM
Timebase corrector (TBC) FAQ admin Restore, Filter, Improve Quality 0 02-11-2005 05:29 AM

Thread Tools



 
All times are GMT -5. The time now is 10:37 PM