Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion.

Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules.

 

Go Back   Doom9's Forum > Capturing and Editing Video > Avisynth Usage

Reply
 
Thread Tools Search this Thread Display Modes
Old 1st November 2014, 12:05   #1  |  Link
AVCHDfreak
Registered User
 
Join Date: Dec 2013
Posts: 30
One-inch Type C videotape characteristics and simulating them

Hi all,

I made a thread a while ago asking about the u-matic format, actually I'm trying to find an analog videotape format used in the 1980s to 1990s where the quality was somewhere in between u-matic and betacam, so far the closest I've found is one-inch Type C, will need some help with its characteristics and simulating them with Avisynth, has anyone had experience with Type C? Thanks in advance,

AVCHDfreak
AVCHDfreak is offline   Reply With Quote
Old 1st November 2014, 12:28   #2  |  Link
smok3
brontosaurusrex
 
smok3's Avatar
 
Join Date: Oct 2001
Posts: 2,395
from memory;
- bandwidth was slightly lower than with betacam sp (but the picture was probably visually the same)
- similar TBC as used in betacam sp (line repeater), at least on sony machines
- a lot better picture than U-matic
- machines were great to operate (fast and with 6 usable markers, again sony version)
__________________
certain other member
smok3 is offline   Reply With Quote
Old 1st November 2014, 12:44   #3  |  Link
AVCHDfreak
Registered User
 
Join Date: Dec 2013
Posts: 30
Hi smok3, people have told me that betacam and type c are visually similar, the major difference I know is that type c has visible dot crawl due to it being a composite video format, the source material I'm referencing from also has dot crawl but it definitely had better quality than u-matic. About the line repeater, is this the 'cloned lines' effect where good lines are repeated until the next good line?
AVCHDfreak is offline   Reply With Quote
Old 1st November 2014, 12:48   #4  |  Link
smok3
brontosaurusrex
 
smok3's Avatar
 
Join Date: Oct 2001
Posts: 2,395
Quote:
Originally Posted by AVCHDfreak View Post
About the line repeater, is this the 'cloned lines' effect where good lines are repeated until the next good line?
yes, thats the one.
__________________
certain other member
smok3 is offline   Reply With Quote
Old 1st November 2014, 12:58   #5  |  Link
AVCHDfreak
Registered User
 
Join Date: Dec 2013
Posts: 30
I've seen some shows from the '90s in East Asia where they still used u-matic tapes and they had consistent dropout (white 'bearding') in parts of the picture, if tbc wasn't used for type c, how would those dropouts look like? Also, what was the frequency of an artifact occurring (e.g. picture roll/loss of sync, loss of color phase)?
AVCHDfreak is offline   Reply With Quote
Old 1st November 2014, 13:40   #6  |  Link
smok3
brontosaurusrex
 
smok3's Avatar
 
Join Date: Oct 2001
Posts: 2,395
Picture was highly stable on a good machine, you could have a single dropout/hour (This was a huge tape).
__________________
certain other member
smok3 is offline   Reply With Quote
Old 2nd November 2014, 03:46   #7  |  Link
AVCHDfreak
Registered User
 
Join Date: Dec 2013
Posts: 30
That's no wonder why this format was also popular in Asia because of its quality output despite having much higher-quality alternatives, how would dropouts appear if say the machine was playing damaged tape (i.e. creasing or fingerprints on tape) or when the video heads need replacement, would it look something similar to VHS?

I'm using flaXen's Analog Noise Tool in Virtualdub which replicates most of the visual characteristics of the tape format, do you recommend any other plugin or script that may help in simulating the tape format?
AVCHDfreak is offline   Reply With Quote
Old 2nd November 2014, 17:32   #8  |  Link
raffriff42
Retried Guesser
 
raffriff42's Avatar
 
Join Date: Jun 2012
Posts: 1,375
Quote:
Originally Posted by AVCHDfreak View Post
would it look something similar to VHS?
It would vary, depending on the error concealment and TBC. Early (or inexpensive) models could only replace a bad line with a copy of the line above in the same field. Later models had a full frame of memory (!) and error concealment good enough to hide all but the nastiest tape creases.
raffriff42 is offline   Reply With Quote
Old 2nd November 2014, 23:23   #9  |  Link
AVCHDfreak
Registered User
 
Join Date: Dec 2013
Posts: 30
Hi raffriff42, I've read at the videohelp forums about South Park using Type C tape in their earliest episodes, how many generations can Type C manage before becoming unwatchable? Wouldn't the dot crawl from each generation lower picture quality quite quickly?

Don't know how the head switching noise looks like but I assume it's already cropped off during playback.
AVCHDfreak is offline   Reply With Quote
Old 3rd November 2014, 00:55   #10  |  Link
raffriff42
Retried Guesser
 
raffriff42's Avatar
 
Join Date: Jun 2012
Posts: 1,375
The dot crawl didn't increase with generations, as long as all processing was done in composite mode - and it was, unless chroma key was needed. The main problem was noise.

I don't know the generation limit, but 3rd generation on-air master tapes were common - source, edited master (which stays in the vault) and air master.
raffriff42 is offline   Reply With Quote
Old 3rd November 2014, 09:21   #11  |  Link
smok3
brontosaurusrex
 
smok3's Avatar
 
Join Date: Oct 2001
Posts: 2,395
I do remember ampex presentation showing 1st and 16th generation in split screen with no visual difference, however in reality 4th generation was highly problematic already irc.
__________________
certain other member
smok3 is offline   Reply With Quote
Old 4th November 2014, 03:54   #12  |  Link
AVCHDfreak
Registered User
 
Join Date: Dec 2013
Posts: 30
Quote:
Originally Posted by raffriff42 View Post
The dot crawl didn't increase with generations, as long as all processing was done in composite mode
So the dot crawl would be removable in that case with dot crawl filters in Virtualdub, because they're somewhat "synchronized" in subsequent generations, right? It's not like the case where you edit a source in Type C format to a different analog composite video format, that would mess up the dots.

There are a lot of sources in composite video connections that can induce dot crawl into the output signal, what would be the biggest contributor in a typical Type C setup - i.e. noise in cables, demodulation in tape recorder?

Quote:
Originally Posted by smok3 View Post
...in reality 4th generation was highly problematic already irc.
Wow.

Do these machines have built-in color system converters? I've seen from my reference footage originating from 1992 where NTSC 60 fields were blended into PAL 50 fields for broadcast but I wasn't sure if this was done separately.

Last edited by AVCHDfreak; 4th November 2014 at 04:08.
AVCHDfreak is offline   Reply With Quote
Old 4th November 2014, 07:02   #13  |  Link
raffriff42
Retried Guesser
 
raffriff42's Avatar
 
Join Date: Jun 2012
Posts: 1,375
The 1/2" and 3/4" video cassette formats recorded color information separately using so-called color-under (heterodyne), so I guess dot crawl could get worse with each generation, but Type C recorded the composite video waveform exactly. This required an advanced TBC, as the signal off tape was very unstable compared to the precise (nanosecond-level) timing requirements of composite video. Without the TBC, the picture would have been horrible - nothing but rainbow artifacts and tearing - but with it, you got the sharpest possible image, given the standards of the time - and no added dot crawl.

So any dot crawl you see from a 1" source would be due to the limitations of NTSC or PAL composite video, not the recorder.
raffriff42 is offline   Reply With Quote
Old 4th November 2014, 10:40   #14  |  Link
AVCHDfreak
Registered User
 
Join Date: Dec 2013
Posts: 30
Just saw a demo about VHS generation loss and color information is lost first, which was a fright seeing the picture so unstable.

After looking at a few Type C demos online (US footage), the quality looks comparably good to Betacam, my reference footage was from Asian countries like Hong Kong and Taiwan, there wasn't much information available about what broadcast equipment the stations there used, I was last told they used Betacam from early '80s onwards, but the footage looked worse than the Type C demos, I don't think that the stations that produced the footage were wired for component-video yet, the quality of the camera can be a factor or they could be handling the Type C footage not as well using less-than adequate TBCs, dropouts occur every three seconds or so (they are full width black lines or little black specks usually, there was one frame where about 40% was just black, no line repeater here), doesn't look like Type C at all in that case, could there be another unheard format during the period?

Quote:
Originally Posted by raffriff42 View Post
...Type C recorded the composite video waveform exactly.
So, if noise affects the picture quality, would noise just be the random voltages in the waveform generated by the recorder's circuitry, or is there more to it?

Last edited by AVCHDfreak; 4th November 2014 at 10:58.
AVCHDfreak is offline   Reply With Quote
Old 4th November 2014, 14:49   #15  |  Link
raffriff42
Retried Guesser
 
raffriff42's Avatar
 
Join Date: Jun 2012
Posts: 1,375
The noise was mostly due to the limitations of magnetic recording. It was much, much lower than VHS or 3/4", but a little worse than Betacam - subjectively at least. Betacam used less tape, so the laws of physics say there should be *more* noise, but Betacam used metal-particle tape as opposed to Type C's iron oxide, so I dunno.

Typical S/N for a 1" VTR might have been around 47dB, according to this sales brochure (here's a screenshot), which is near the limit of ideal 8-bit digital video (20 * log10(256) or 48dB, if I am reading this correctly). *EDIT I believe the "56 dB" quoted is just sales-brochure-speak and would not be seen IRL.

To emulate it, just AddGrain() to taste, but since it was so low, and since noise gets mangled by lossy compression algorithms anyway, I don't see the point.

BTW, according to my old Handbook, 2" quad S/N was around 56 dB, but Type C superceded it for economic reasons. So I guess the deciders said 47dB ought to be enough for anybody

EDIT now that I think about it, much of the apparent noise was due to quantizing errors in the A-to-D->TBC->D-to-A process. This noise (-48dB) was about the same for any 8-bit digital TBC - whether connected to a 2" quad, Type C or Betacam.

Last edited by raffriff42; 18th March 2017 at 00:01. Reason: (fixed image link)
raffriff42 is offline   Reply With Quote
Old 5th November 2014, 03:53   #16  |  Link
AVCHDfreak
Registered User
 
Join Date: Dec 2013
Posts: 30
Sony seems to use Metal-particle for their 8mm/Hi8 range and advertised it as for professional use, may have stored data more densely, but it leaves enoough residue during playback in machines that clog heads up quickly and they don't seem to last very long, I thought Betacam tapes would have that kind of problem too.

Well, 48dB is the dynamic range for 8-bit audio so I get this makes sense to me, so to reduce quantizing errors would be to use TBC with higher bit-count.

After I looked further into my reference footage, things started to look more like U-matic again...
AVCHDfreak is offline   Reply With Quote
Old 6th November 2014, 12:26   #17  |  Link
AVCHDfreak
Registered User
 
Join Date: Dec 2013
Posts: 30
Just continuing as usual studying my footage, there were some shots where part of the vertical blanking interval is visible at the top of the image even with overscan, if the VBI part was small, there would be just one line of white dashes (I'm thinking line 21 captions), other times there would be three or four lines of flickering white dashes (VITC I believe), I'm not sure if it's due to camera or other system used in the process, but why would part of the VBI show up despite proper calibration of VBI offset?
AVCHDfreak is offline   Reply With Quote
Old 6th November 2014, 14:09   #18  |  Link
raffriff42
Retried Guesser
 
raffriff42's Avatar
 
Join Date: Jun 2012
Posts: 1,375
Quote:
Originally Posted by AVCHDfreak View Post
why would part of the VBI show up despite proper calibration of VBI offset?
Define "proper calibration." Closed captions were on line 21 (NTSC), the first line after the end of vertical blanking. Closed captions were not normally blanked. VITC and other VBI data can be on line 10 (NTSC) or later. These lines were often not blanked for over-the-air television, as the lines could contain commercial data (pre-Internet news feeds etc) or commands to the remote transmitter. (*EDIT and these lines would normally be invisible to analog TVs due to overscan)

As far as blanking during analog-to-digital capture, see "technical - which 480 of 486 lines are captured?"
http://forum.doom9.org/showthread.php?t=152179

Last edited by raffriff42; 6th November 2014 at 14:13.
raffriff42 is offline   Reply With Quote
Old 8th November 2014, 05:22   #19  |  Link
AVCHDfreak
Registered User
 
Join Date: Dec 2013
Posts: 30
I'm not too sure what was going on with the shots showing the VBI, it looked like that it had been "shrunken" slightly, probably by a separate video mixer, which embedded the original source image along with part of its VBI (?), I may never know what happened there. When there is text on plain backgrounds, there's an intermittent "wobble" on the text, is this normal for type c (with less advanced tbc)?

Off-topic, but I'm planning to get probably a u-matic (sp) or betacam vtr for obtaining the real effect, how long are these machines likely to last before requiring service like head replacement? Obsolescence makes it quite hard for trying out the old stuff...

Last edited by AVCHDfreak; 8th November 2014 at 09:14.
AVCHDfreak is offline   Reply With Quote
Old 8th November 2014, 14:44   #20  |  Link
raffriff42
Retried Guesser
 
raffriff42's Avatar
 
Join Date: Jun 2012
Posts: 1,375
Your video might have gone through a 3/4" generation at some point. There could have been a timing error - eg, VTR not synced to the character generator.
raffriff42 is offline   Reply With Quote
Reply

Tags
analog, simulate, videotape

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +1. The time now is 10:54.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, vBulletin Solutions Inc.