Go Back    Forum > Digital Video > Video Project Help > Restore, Filter, Improve Quality

Reply
 
LinkBack Thread Tools
  #1  
03-21-2011, 06:22 AM
lordsmurf's Avatar
lordsmurf lordsmurf is offline
Site Staff | Video
 
Join Date: Dec 2002
Posts: 13,508
Thanked 2,449 Times in 2,081 Posts
Quote:
Hi LS,

I'm keeping my eye on a Snell & Wilcox TBS24 for VHS to Digital transfers with my non-TBC VCR's. This is probably higher-end than I really need... but as the auction gets closer to the end, the price is still great, and it seems S&W has a reputation for being top-of-the-line.
Here it is: http://cgi.ebay.com/ws/eBayISAPI.dll...m=290541393564

From messaging back and forth with the seller, he says I'll probably need an external sync source because the TBS24 doesn't have a "local" or "internal" sync setting and the studio where he worked always had this unit hooked up to house sync.

Would you agree with that statement? If so, would this be a good option for sync:
Leitch SPG-130N: http://cgi.ebay.com/ws/eBayISAPI.dll...m=230586490241

I realize you may not have worked with either of these before, but even if that's the case you'd probably have a better guess than me.

Thanks in advance!
Obviously I'm replying to this after the auction ended, but the information I have to share may still be useful for you.

To swipe a quote from a Creative Cow user post:

Quote:
a tbc in it's most native form cannot syncronise a signal it can only stabelize it. your snell can frame syncronize a signal that's why it was sold seperate. it was mostly used to syncronize remotefeeds to studio ref.
I've tried to write a succinct definition of a TBC vs frame sync for a couple of years now, and I think that's the best one I've heard of yet. It would also explain why certain older rack mounted TBCs behave as they do.

This Snell & Willcox, for example, appears to be a TBC + proc amp. You're correct, I don't have specifics on this particular model, nor first-hand experience with its use. But I do have some general observations about rackmounted TBCs.

These were built for non-VHS analog workflows that pre-date digital conversion work. As such, most of them make for poor additions to your VCR + capture card. At worst, this may turn out to be a great proc amp, and it has that added noise reduction.

It may be because of the simple reason given in the above quote: a base TBC can correct a signal, but it cannot "frame sync" (not the same as genlock).

Modern TBCs, designed during and for the digital conversion era, were made by companies like DataVideo and Cypress. These are essentially a version of TBC + frame synchronizer, to provide a steady and stable stream of video data to the capture card or DVD recorder. Without a stable/continuous signal, the capture session will barf itself and halt. The one issue with these is they appear to do nothing for horizontal jitter, unlike line TBCs in higher-end VCRs.

I don't think the Leitch SPG-130N is a frame sync, but rather a genlock. That's of no use to you.

The reason these rack mounted TBCs tend to have "good prices" is because nobody wants them anymore -- they're next to useless for consumers and hobbyists, as well as broadcasters (and even colleges) that moved up to all-digital workflows. The few holdouts for such gear are transfer and post houses, when dealing for broadcast sources (Betacam, S-VHS, etc). But even then, work has dwindled, and they have all the gear they could use -- including backups and extra parts.

If anybody has contrary information, please post it. Opinions are also welcome, be it in agreement or in dissent. All discussion is welcome.

Thanks.


You may also find this article/post useful: What is a TBC? Time Base Correction for Videotapes
I wrote that a couple of years ago.


This question was asked in a private message. Rather than hide our tech advice in private conversations, Site Staff will often answer PMs (from any site) here in the digitalFAQ.com forum, so that others may read and benefit from our expertise. Please continue the conversation here. Either login or join as a Free Member, and we can continue troubleshooting your video, photo or web related issue. Thanks for understanding our tech Q&A policies.



- Did my advice help you? Then become a Premium Member and support this site.
- For sale in the marketplace: TBCs, workflows, capture cards, VCRs
Reply With Quote
Someday, 12:01 PM
admin's Avatar
Ads / Sponsors
 
Join Date: ∞
Posts: 42
Thanks: ∞
Thanked 42 Times in 42 Posts
  #2  
03-21-2011, 08:45 PM
jbd5010 jbd5010 is offline
Free Member
 
Join Date: Feb 2011
Location: Philadelphia, PA Area
Posts: 181
Thanked 15 Times in 11 Posts
Interesting. I thought standalone TBC's would also take care of horizontal jitter (in fact, I thought that was the main purpose of creating a new sync signal!)

I was wondering why the NovaTrol NovaMate2 TBC I had was doing worse than my SR-V10U's TBC on one of my tapes with really bad horizontal distortions. So far the AG-1980 wins in this category among my limited amount of equipment.

So what I'm wondering, then, is if it's possible to use a VCR without a TBC and still use some external equipment to cure horizontal issues? For example, I think I'd like the results I get with my older 9400U if I could correct horizontal jitter.

I have TBC-1000 on the way, but from the sound of it that might not do the trick...
Reply With Quote
  #3  
03-21-2011, 09:05 PM
NJRoadfan NJRoadfan is offline
Premium Member
 
Join Date: Sep 2010
Posts: 1,155
Thanked 357 Times in 293 Posts
Quote:
Originally Posted by jbd5010 View Post
So what I'm wondering, then, is if it's possible to use a VCR without a TBC and still use some external equipment to cure horizontal issues? For example, I think I'd like the results I get with my older 9400U if I could correct horizontal jitter.

I have TBC-1000 on the way, but from the sound of it that might not do the trick...
The AVT-8710 did a wonderful job eliminating jitter from my Betamax home movies, the output was rock solid and jitter free for the most part. Some of the worst parts of the tape still had some curling, but its better then nothing. Its a good thing too, since no Betamax machines with built in line TBCs were ever made.
Reply With Quote
  #4  
03-21-2011, 09:17 PM
jbd5010 jbd5010 is offline
Free Member
 
Join Date: Feb 2011
Location: Philadelphia, PA Area
Posts: 181
Thanked 15 Times in 11 Posts
Ah, Betamax. None of those floating around my house anywhere... I almost wish there were because I've never seen/used any Beta tapes or VCR's.
Reply With Quote
  #5  
03-22-2011, 01:01 AM
lordsmurf's Avatar
lordsmurf lordsmurf is offline
Site Staff | Video
 
Join Date: Dec 2002
Posts: 13,508
Thanked 2,449 Times in 2,081 Posts
Quote:
So what I'm wondering, then, is if it's possible to use a VCR without a TBC and still use some external equipment to cure horizontal issues? For example, I think I'd like the results I get with my older 9400U if I could correct horizontal jitter.
Why not use the Panasonic ES10 method?
It's also discussed here: What is a TBC? Time Base Correction for Videotapes

I use that method, and it's quite good.
One workflow is currently: Panasonic AG1980 > Panasonic ES10 > JVC DR-M10

VCR for tape transport, ES10 for jitter, DR-M10 for chroma removal (XP mode capture).
After the hardware has done its work, video is further processed in Avisynth + VirtualDub, then finally Premiere CS4.
(Audio has hardware/software work too, but concentrating on video for conversation.)

And it does quite well.

Note that our workflows are temporary, based on the project. Gear is swapped around, re-wired, etc, as needed. Nothing stays connected for very long. Most items don't even stay in the same spot on the custom racks more than a few months max.

I'd be curious to see what the S&W NR controls can do, if anything of significance.

- Did my advice help you? Then become a Premium Member and support this site.
- For sale in the marketplace: TBCs, workflows, capture cards, VCRs
Reply With Quote
  #6  
03-22-2011, 07:49 AM
jbd5010 jbd5010 is offline
Free Member
 
Join Date: Feb 2011
Location: Philadelphia, PA Area
Posts: 181
Thanked 15 Times in 11 Posts
Quote:
Originally Posted by lordsmurf View Post
One workflow is currently: Panasonic AG1980 > Panasonic ES10 > JVC DR-M10
Well I've got the AG1980 and the JVC DR-M10... although I'm still waiting for the remote I had to order separately for the DR-M10. If there's a way to change the channel (i.e. get it off of default channel 2 and over to an A/V input) without the remote, I haven't figured it out. I was hoping I could do AV/C control via firewire... no dice.

I've been eyeing up some ES10's, but some (I think including Lordsmurf) say that it over-processes video and even introduces artifacts when used as a pass-through? Or is that only if you actually use the ES10 as the authoring burner...
Reply With Quote
  #7  
03-27-2011, 12:12 PM
robjv1 robjv1 is offline
Free Member
 
Join Date: Jun 2009
Posts: 187
Thanked 37 Times in 33 Posts
Yeah, the ES10 is not totally transparent on passthrough. I don't notice any artifacts in light scenes, but in very dark scenes it tends to make the light parts of the picture noticeably posterized. That's my experience given my ES-10 interacting with my other equipment, YMMV.

If there was a device that can do the miracle work the ES-10 can do, but more transparently, I'm all ears
Reply With Quote
  #8  
03-27-2011, 11:02 PM
NJRoadfan NJRoadfan is offline
Premium Member
 
Join Date: Sep 2010
Posts: 1,155
Thanked 357 Times in 293 Posts
Quote:
Originally Posted by jbd5010 View Post
Ah, Betamax. None of those floating around my house anywhere... I almost wish there were because I've never seen/used any Beta tapes or VCR's.
The curse of being an early adopter. We all had Beta machines in the family before VHS. Our first camcorder was the Sony BetaMovie BMC-110, purchased as soon as it came out. My grandfather was a big gadget person and had to have the first all in one consumer camcorder.

Back to TBCs. One reason why pro external units seem to be unloved is by the mid to late 90s almost all pro VTRs had one built in. Built in TBCs have the advantage of being able to directly servo lock the VTRs transport to an external genlock source (helps avoid frame dropping by TBC). The best TBCs can do the following:

1. Correct severe horizontal jitter (including flagging/curling). Rumor has it, the JVC SR-VD400US (pro version of the HM-DH40k DVHS deck) can fix curling, the Digipure circuit on this model differs from past units having 6MB frame buffer.

2. Provide dynamic noise reduction, particularly chroma noise reduction (DigiPure)

3. Provide clean continuous sync output. External genlock is nice, but not needed for DVD transfers.

4. Provide dropout compensation. This one is TOUGH to find. JVC's Digipure and the popular AVT-8710/Datavideo TBC-1000 DON'T do this.

5. Basic proc-amp controls.
Reply With Quote
  #9  
03-28-2011, 09:53 PM
jbd5010 jbd5010 is offline
Free Member
 
Join Date: Feb 2011
Location: Philadelphia, PA Area
Posts: 181
Thanked 15 Times in 11 Posts
Quote:
Originally Posted by NJRoadfan View Post
Rumor has it, the JVC SR-VD400US (pro version of the HM-DH40k DVHS deck) can fix curling, the Digipure circuit on this model differs from past units having 6MB frame buffer.
You know, NJRoadfan, I was ALREADY kicking myself that I missed the end of an SR-VD400US auction that ended at $80 shipped. People don't recognize the SR-VDA300U and SR-VD400US as the HM-DH3k/4k's that they really are.

Now I'm really wishing I hadn't missed it!

http://cgi.ebay.com/JVC-SR-VD400U-D-...item4157c13f03

In any case, I just picked up a Snell & Wilcox NRS30 noise reducer. It has a built-in TBC, but I'm excited to see how it stacks up against NeatVideo. In Hardware vs Software processing, hardware almost always wins. I'd love to have something like the VDub Neatvideo plugin that runs as hardware in realtime.

More info on NRS30: http://home.tiscali.nl/rajduim/Video...Processing.pdf
(See page 7 of PDF)

EDIT: Also, I have an HM-DH40000U sitting in my living room that I was getting ready to sell. Any idea if this has that 6MB buffer circuit also, or is it just the pro version? I didn't seem to get any results from it that made it worth keeping in my obnoxiously large stack of equipment.
Reply With Quote
  #10  
03-28-2011, 10:44 PM
NJRoadfan NJRoadfan is offline
Premium Member
 
Join Date: Sep 2010
Posts: 1,155
Thanked 357 Times in 293 Posts
Quote:
Originally Posted by jbd5010 View Post
You know, NJRoadfan, I was ALREADY kicking myself that I missed the end of an SR-VD400US auction that ended at $80 shipped. People don't recognize the SR-VDA300U and SR-VD400US as the HM-DH3k/4k's that they really are.

Now I'm really wishing I hadn't missed it!

http://cgi.ebay.com/JVC-SR-VD400U-D-...item4157c13f03

EDIT: Also, I have an HM-DH40000U sitting in my living room that I was getting ready to sell. Any idea if this has that 6MB buffer circuit also, or is it just the pro version? I didn't seem to get any results from it that made it worth keeping in my obnoxiously large stack of equipment.
That SR-VD400US is on its way to my house, would have been here already if the seller wasn't a slow poke shipping it. Kind of a fluke purchase, I put a not very serious bid in and won it anyway. Then again the lack of remote scared some people away, I just happen to have a JVC remote here. The HM-DH40000U is identical to the SR-VD400US aside from the "pro" branding, same 6MB frame memory.

The only reason I picked this up was the price and to go through these 80+ DVHS tapes I got... one of them has to have a decent movie on it... I hope. I'm still on the lookout for a nice AG-1980 and/or Mitsu HS-HD2000U for transfers though. No rush since the big 20 tape job is finished... now to deal with that guy's stack of MiniDV tapes...
Reply With Quote
  #11  
03-28-2011, 10:53 PM
jbd5010 jbd5010 is offline
Free Member
 
Join Date: Feb 2011
Location: Philadelphia, PA Area
Posts: 181
Thanked 15 Times in 11 Posts
Well then in that case, nice snag! I probably would've bid you up to $75 had I not been out somewhere with friends
Reply With Quote
  #12  
03-30-2011, 03:25 AM
robjv1 robjv1 is offline
Free Member
 
Join Date: Jun 2009
Posts: 187
Thanked 37 Times in 33 Posts
Quote:
Originally Posted by jbd5010 View Post
In any case, I just picked up a Snell & Wilcox NRS30 noise reducer. It has a built-in TBC, but I'm excited to see how it stacks up against NeatVideo. In Hardware vs Software processing, hardware almost always wins. I'd love to have something like the VDub Neatvideo plugin that runs as hardware in realtime.
Very interested in your observations on this unit, please post about it here if you get an opportunity!
Reply With Quote
  #13  
03-30-2011, 04:59 PM
jbd5010 jbd5010 is offline
Free Member
 
Join Date: Feb 2011
Location: Philadelphia, PA Area
Posts: 181
Thanked 15 Times in 11 Posts
Okay, the unit arrived today. I came up with a good test to do right from the start. I have an EP tape that is a second-generation transfer from VHS-C to full size VHS. It has a lot of chroma noise at the beginning of the recording, I think the full-size tape was also recycled (I was 9 years old so it's hard to remember!)

I tried the Chroma Noise Reduction feature, and it had pretty good results! Picture 1 attached is without noise reduction, and picture two is with Chroma NR set to "High" (Available settings: Off, Low, Med, High)

Notice how the purple vertical bar across the brown paper shopping bag and drawer on the right, as well as the horizontal one across the middle-left of the picture are significantly reduced.

During the black portion of the video a few seconds before this, noise was also significantly reduced (although this would be "easier" for the filter since it is filtering obvious noise using a series of all-black frames).

I will post more noteworthy results as I get into using this unit more.

I found, though, that just as expected, the TBC on this unit does absolutely nothing for horizontal jitter, which is very severe on this tape.

(Even though the AG-1980's TBC was off, though, that VCR still seems to reduce jitter immensely when compared to non-TBC JVC units like my HR-S9400U).

snell_off_chromaNR.jpg

You must be logged in to view this content; either login or register for the forum. The attached screen shots, before/after images, photos and graphics are created/posted for the benefit of site members. And you are invited to join our digital media community.


Reply With Quote
  #14  
03-30-2011, 05:13 PM
robjv1 robjv1 is offline
Free Member
 
Join Date: Jun 2009
Posts: 187
Thanked 37 Times in 33 Posts
Interesting! Not quite as dramatic as I expected, but it seems to do a good job, at least on certain aspects of the image. Looking at the date-stamp reveals a big improvement in the waviness of the image as well.

One thing I noted though -- look at the purple hue on the white cushion(?) under the cabinets on the left side of the screen, it appears to show a bit more in the after image than in the before.

The reduction of edge enhancement is more prominent for certain aspects of the image than others between the shots too. Looking at the lines in the wood paneling underneath the cabinet show a big difference between shots, with the appearance of the grooves nearly vanishing as you go down the wall. However that effect does not seem nearly as pronounced when looking at the floor tiles as they move from away from the camera. Perhaps the dark color of the paneling makes the effect more pronounced. I'd be curious to see the same shot on the "low" and "medium" settings.
Reply With Quote
  #15  
03-30-2011, 05:17 PM
jbd5010 jbd5010 is offline
Free Member
 
Join Date: Feb 2011
Location: Philadelphia, PA Area
Posts: 181
Thanked 15 Times in 11 Posts
Now a test of the Median Noise Filter.

Same tape as before, but in a darker scene with some blank wall in the background and also some objects with detail to get a good overall feel for the noise reduction. Same settings as the Chroma NR test... the picture setting on the AG-1980 is "DETAIL."

Decent noise reduction, but it can't keep up with NeatVideo. Not bad for a realtime solution, though. The NeatVideo example was profiled using the auto-profile feature and a blank wall in a different, brighter video frame.

Picture 1: Median filter off.
snell_off_medianFilter.jpg

Picture 2: Median Filter on.
You must be logged in to view this content; either login or register for the forum. The attached screen shots, before/after images, photos and graphics are created/posted for the benefit of site members. And you are invited to join our digital media community.


Picture 3: VirtualDub NeatVideo plugin, for reference.
snell_on_medianFilter.jpg


Reply With Quote
  #16  
03-30-2011, 05:20 PM
jbd5010 jbd5010 is offline
Free Member
 
Join Date: Feb 2011
Location: Philadelphia, PA Area
Posts: 181
Thanked 15 Times in 11 Posts
Yeah, I think what might be a good idea is to start a new thread to talk about this unit. I don't have major plans during the day this Saturday... I can perform some more thorough tests and share the results.

This unit also has horizonal and vertical enhancement filters that I'd like to test out, along with a few other things.
Reply With Quote
  #17  
04-01-2011, 10:05 PM
lordsmurf's Avatar
lordsmurf lordsmurf is offline
Site Staff | Video
 
Join Date: Dec 2002
Posts: 13,508
Thanked 2,449 Times in 2,081 Posts
Going back a few posts...
Quote:
say that it over-processes video and even introduces artifacts when used as a pass-through? Or is that only if you actually use the ES10 as the authoring burner...
It all comes back to a "lesser of evils" situation. Restoration is about making the video look better, as opposed to the impossible task of making it look perfect. Noise filtering is rarely transparent and perfect. Sometimes you'll even have to stack filters (hardware and/or software) to mitigate damage done by a previous filter. While it's very intelligent to observe filters piecemeal, don't lose the forest in the trees. Keep in mind at all times that the end-goal is what matters. And that goal is to "make it better". Five steps forward, plus one step back, still equals progress!


And then replying to the more recent posts...
Quote:
the picture setting on the AG-1980 is "DETAIL."
From what I know of the AG-1980, this doesn't do anything. That was a failed feature, and was essentially a dead switch on the 1980. A later VCR (which never happened) would have actually used it. Maybe I'm wrong -- but that's the information I have, regarding this switch. Visually, I've never been able to see changes to the picture, regardless of the switch's setting.

Quote:
Picture 2: Median Filter on.
That's somewhat terrible, and therefore disappointing. It looks like the VirtualDub Static NR filter cranked all the way up. it just blurs the video by several pixels. It appears to be intraframe NR, as opposed to interframe NR. Interframe NR, or averaging temporally between frames, is what I would consider to be a "median" filter for video.

Quote:
Picture 3: VirtualDub NeatVideo plugin, for reference.
NeatVideo could use some manual tweaks to make it better. The default settings tend to be a wee bit too aggressive.

Still interesting, nonetheless.

- Did my advice help you? Then become a Premium Member and support this site.
- For sale in the marketplace: TBCs, workflows, capture cards, VCRs
Reply With Quote
  #18  
04-02-2011, 12:01 PM
jbd5010 jbd5010 is offline
Free Member
 
Join Date: Feb 2011
Location: Philadelphia, PA Area
Posts: 181
Thanked 15 Times in 11 Posts
I've always noticed that the "detail" switch on the 1980 was essentially the same as setting the witch to NOR and sliding the soft/sharp slider to the middle position. I guess I switched it to detail because I figured it would "lock in" the middle-of-the-road sharpening setting.

As far as the median filter, the instruction manual says:

Quote:
"The NRS30 is a full-featured multi-standard format interchange Timebase/Synchroniser incorporating a powerful 2-stage noise reducer comprising a non-linear adaptive recursive filter followed by a unique spatial-temporal median filter. The median filter has the ability to remove noise peaks without introducing smearing artefacts and is ideally suited to the suppressing of impulse noise originating from film dirt or ‘sparklies’ prevalent in pictures received via satellite sources.All recursive and median noise reduction coefficients are adjustable and a bargraph display shows the calculated noise floor of the signal being processed so the adaption threshold may be accurately set."
There are four settings for this filter, and they're described as follows:

The Median Filter switch selects the following filter characteristics:-

Quote:
1 Off
2 Luminance 5-point, 2 dimensional spatial filter
3 Luminance 7-point, 2 dimensional spatial filter (This setting has two configurations, SYMMETRIC and ASYMMETRIC which are selectable by a DIP switch on the noise reducer PCB. See ‘Noise Reduction’ section)
4 As position 3, plus additional 7-point filtering for the Chrominance channel.
Well, the front panel isn't very informative. It just has "Off" to the left of the knob and "max" to the right for the median filter knob. Obviously this filter merits some more testing/evaluation, since I thought what I did was just turn it "all the way up" for that screen capture. Better results might be possible.

Finally, yes I was very lazy when I ran NeatVideo that time. I need to become more familiar with it in order to get better results.

I'll come back with some more test results later.
Reply With Quote
  #19  
04-03-2011, 12:35 AM
lordsmurf's Avatar
lordsmurf lordsmurf is offline
Site Staff | Video
 
Join Date: Dec 2002
Posts: 13,508
Thanked 2,449 Times in 2,081 Posts
Lazy is okay, when you're just testing.

- Did my advice help you? Then become a Premium Member and support this site.
- For sale in the marketplace: TBCs, workflows, capture cards, VCRs
Reply With Quote
  #20  
04-03-2011, 05:25 PM
jbd5010 jbd5010 is offline
Free Member
 
Join Date: Feb 2011
Location: Philadelphia, PA Area
Posts: 181
Thanked 15 Times in 11 Posts
Jumping back to our SR-VD400U/HM-DH40000U tangent for a moment...

I popped in the tape used for the captures above, which has distinct time base errors. The 400U/40000U definitely has a solid TBC in it.

However, I'm finding with my unit that I can't capture more than 3 or 4 seconds of MPEG2 via firewire using CapDVHS. The capture just hangs up and stops after a few seconds... I never had that issue with my HM-DH30000U's before. Any issues like this NJRoadFan?

I have plenty of disk space, so that's not the problem.
Reply With Quote
Reply




Similar Threads
Thread Thread Starter Forum Replies Last Post
Seeking advice for VHS transfers (Equipment & Capture Methods) quarkz Project Planning, Workflows 18 01-08-2013 11:31 PM
Digital Camcorder needed - suggestions ? Canon Photo Cameras: Buying & Shooting 1 03-20-2009 02:52 AM
Digital cable reception bad, due to converter box ? mlaviolette Capture, Record, Transfer 1 12-21-2006 02:18 PM
Need the device that transfers tv shows to PC ty3 Capture, Record, Transfer 4 02-02-2006 06:42 AM

Thread Tools



 
All times are GMT -5. The time now is 04:28 AM