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 7th October 2013, 21:41   #1  |  Link
Music Fan
Registered User
 
Join Date: May 2009
Location: Belgium
Posts: 1,743
VHS capture (through dvd recorder) in Composite or S-video ? (Avisynth comparison)

Hi,
I have to capture a Pal VHS with an Hauppauge PVR-150 (I do this with Virtual VCR in Lagarith yuy2) and I have 3 possible connections between VCR and PVR-150 ;
-direct connection ;
1) composite using a scart to composite adapter (only scart output on this VCR)

-standalone dvd recorder between VCR and PVR-150 ; dvd not recording of course, just switched on to let pass the signal to use its TBC better than PVR-150's TBC (especially on a Secam VHS I captured some time ago, maybe less useful with Pal) ; VCR connected with Scart to Dvd recorder (VCR probably sending a composite signal on scart), but 2 solutions for dvd output to PVR-150 ;
2) composite
3) s-video

I read that s-video was not very useful for VHS but also that sometimes, the Y/C separation could be best done by the player than by the capture card.
So I made tests ; I don't see any differences but as my eyes are not perfect, I'd like to be sure which is the best with this equipment.

Here are 3 extracts in Lagarith yuy2 ;
1) composite without dvd ;
http://www15.zippyshare.com/v/78102589/file.html
There is a little header problem with this video due to the Hauppauge filter used by VirtualVCR ; the framerate detected is 29.97 while it's actually 25 fps, just change it with Virtual Dub (video, framerate, source rate ... and save in direct stream copy).

2) composite with dvd ;
http://www58.zippyshare.com/v/63861223/file.html

3) s-video with dvd ;
http://www6.zippyshare.com/v/85808976/file.html

You can find Lagarith codec there ;
http://lags.leetcode.net/codec.html

Why do I ask it in Avisynth section ?
Because I'd like to know how to analyze these videos with Avisynth to measure precisely differences between these captures (if there are differences, I think to luma and chroma levels) to choose the best way to capture this VHS.
I will also need to filter this video but let's begin with the capture connection's question.

Last edited by Music Fan; 7th October 2013 at 22:34.
Music Fan is offline   Reply With Quote
Old 7th October 2013, 23:57   #2  |  Link
Guest
Guest
 
Join Date: Jan 2002
Posts: 21,901
You should use s-video as the luma and chroma are carried on separate wires and are not combined and thus do not require separation.
Guest is offline   Reply With Quote
Old 8th October 2013, 09:05   #3  |  Link
Music Fan
Registered User
 
Join Date: May 2009
Location: Belgium
Posts: 1,743
I know but for VHS things are different and s-video is not always recommended, that's why I said that the Y/C separation could be best done by the player than by the capture card.
Does anyone know how to analyze videos ?
Music Fan is offline   Reply With Quote
Old 8th October 2013, 09:25   #4  |  Link
Asmodian
Registered User
 
Join Date: Feb 2002
Location: San Jose, California
Posts: 4,406
With s-video the Y/C separation is done by the S-VCR not the capture card and sent separately so it doesn't cause interference with the luma information. Why would s-video not be recommended for VHS? My quick tests always looked much better using s-video but the Black Magic Intensity I was using doesn't have a great comb filter. I was capturing in uncompressed 10-bit 4:2:2. S-video is nice because you don't need a comb filter at all.

As I understand it, unless damaged, s-video is always better than composite.

"Analyze" is pretty vague.
Asmodian is offline   Reply With Quote
Old 8th October 2013, 09:59   #5  |  Link
Music Fan
Registered User
 
Join Date: May 2009
Location: Belgium
Posts: 1,743
Quote:
Originally Posted by Asmodian View Post
Why would s-video not be recommended for VHS?
Because I read several times that VHS recorded video in a composite mode, Y and C are blended.
S-VHS record Y and C separately, thus a s-video cable is especially useful for S-VHS.
It means that for VHS, Y and C have to be separated by the capture card (if VCR connected in composite) or before, like I tried with the dvd burner used as intermediary.
And the quality of this separation varies from one device to another.
Topic on the same subject ;
http://forum.grassvalley.com/forum/s...ead.php?t=1389

Quote:
Originally Posted by Asmodian View Post
"Analyze" is pretty vague.
I got this answer on another forum ;
ColorYUV(analyze=true)
I tested but the results look very close and depend on the frames (values change from a frame to another), so it's hard to make a conclusion.
I cropped black borders ;
crop(16,2,-6,-24)

Don't forget that my question also concerns the TBC quality (you can compare both composite extracts).

Last edited by Music Fan; 8th October 2013 at 10:03.
Music Fan is offline   Reply With Quote
Old 8th October 2013, 13:22   #6  |  Link
pandy
Registered User
 
Join Date: Mar 2006
Posts: 1,049
Quote:
Originally Posted by Music Fan View Post
Because I read several times that VHS recorded video in a composite mode, Y and C are blended.
S-VHS record Y and C separately, thus a s-video cable is especially useful for S-VHS.
It means that for VHS, Y and C have to be separated by the capture card (if VCR connected in composite) or before, like I tried with the dvd burner used as intermediary.
This is not true - Y and C in each color VCR are recorded separately - "The luminance (Y) and color (C) components of the composite video signal are recorded differently. Luminance, which is in effect the black and white picture with all the high resolution components but no color, is frequency modulated on a carrier at around 3.4 MHz. The deviation is about 1 MHz and the maximum frequency recorded on a VHS tape is a little over 5 MHz (Beta is slightly different and S versions of Beta and VHS extend some of these to achieve higher bandwidths. The color signal is separated from the composite video and is amplitude modulated on a 629 kHz carrier. This is called the "color under' system."

http://www.repairfaq.org/sam/vcrfaq.htm#vcrvhsvid
http://www.modeemi.fi/~leopold/AV/VideoFormats.html

main difference between S-VHS and VHS (except resolution) is fact that VHS VCR usually have no separate output for C and Y signal (S-Video) - S-VHS VCR can read VHS tape and S-Video providing Y and C will be better than CVBS.
pandy is offline   Reply With Quote
Old 8th October 2013, 13:43   #7  |  Link
Music Fan
Registered User
 
Join Date: May 2009
Location: Belgium
Posts: 1,743
Quote:
Originally Posted by pandy View Post
This is not true - Y and C in each color VCR are recorded separately - "The luminance (Y) and color (C) components of the composite video signal are recorded differently. Luminance, which is in effect the black and white picture with all the high resolution components but no color, is frequency modulated on a carrier at around 3.4 MHz. The deviation is about 1 MHz and the maximum frequency recorded on a VHS tape is a little over 5 MHz (Beta is slightly different and S versions of Beta and VHS extend some of these to achieve higher bandwidths. The color signal is separated from the composite video and is amplitude modulated on a 629 kHz carrier. This is called the "color under' system."

http://www.repairfaq.org/sam/vcrfaq.htm#vcrvhsvid
http://www.modeemi.fi/~leopold/AV/VideoFormats.html
Thanks, interesting !
It means that a lot of people on several forums are wrong about that subject because all research I made (maybe not enough !) lead to the information that VHS record a composite signal.

Quote:
Originally Posted by pandy View Post
main difference between S-VHS and VHS (except resolution) is fact that VHS VCR usually have no separate output for C and Y signal (S-Video) - S-VHS VCR can read VHS tape and S-Video providing Y and C will be better than CVBS.
Ok, but as my VCR does not have S-Video output but only SCART, and as the manual does not mention if this SCART can send a s-video signal, how to know what signal is sent on SCART to the dvd burner ? Could it be RGB ?
Let's admit that this is CVBS ; in this case I still have to know if the Y/C separation by the dvd is useful or not.
That's why I gave some extracts that can be analyzed.
Music Fan is offline   Reply With Quote
Old 8th October 2013, 14:12   #8  |  Link
Guest
Guest
 
Join Date: Jan 2002
Posts: 21,901
Quote:
Originally Posted by Music Fan View Post
Thanks, interesting !
It means that a lot of people on several forums are wrong about that subject because all research I made (maybe not enough !) lead to the information that VHS record a composite signal.
You need to research original specifications and other technical materials, not the bloviations of random people on the internet.

Quote:
Ok, but as my VCR does not have S-Video output but only SCART, and as the manual does not mention if this SCART can send a s-video signal, how to know what signal is sent on SCART to the dvd burner ? Could it be RGB ?
SCART is capable of carrying composite, s-video, RGB, and YPbPr component. To determine which are actually present on your SCART connection, you can either examine a schematic for the VCR, or, if that is absent, look at the signals with an oscilloscope.

Quote:
Let's admit that this is CVBS ; in this case I still have to know if the Y/C separation by the dvd is useful or not.
Of course it is useful. How do you think your DVD recorder could avoid doing this separation?
Guest is offline   Reply With Quote
Old 8th October 2013, 14:25   #9  |  Link
Music Fan
Registered User
 
Join Date: May 2009
Location: Belgium
Posts: 1,743
Quote:
Originally Posted by neuron2 View Post
SCART is capable of carrying composite, s-video, RGB, and YPbPr component. To determine which are actually present on your SCART connection, you can either examine a schematic for the VCR, or, if that is absent, look at the signals with an oscilloscope.
Ok, not simple, I don't have oscilloscope but I can search a schematic for the VCR (which is 15 years old).

Quote:
Originally Posted by neuron2 View Post
Of course it is useful.
That's not so evident, especially if my VCR send only CVBS on its SCART. In this case, I have to choose between an Y/C separation by the card of before, by the dvd burner, but I need to know which one does it best.
Did you analyze my videos and saw differences ?
Music Fan is offline   Reply With Quote
Old 8th October 2013, 14:39   #10  |  Link
Guest
Guest
 
Join Date: Jan 2002
Posts: 21,901
Without the schematics I can't know what you are getting through a DVD player interposed. How are you connecting the DVD player to the PVR? You'd have to have s-video or component to the PVR for this scenario to potentially improve things (if the DVD player Y/C separation is better than that of the PVR).

But can't you assess your test results yourself? Why do you need us to decide for you which appears better?

Last edited by Guest; 8th October 2013 at 14:42.
Guest is offline   Reply With Quote
Old 8th October 2013, 14:59   #11  |  Link
Music Fan
Registered User
 
Join Date: May 2009
Location: Belgium
Posts: 1,743
Quote:
Originally Posted by neuron2 View Post
Without the schematics I can't know what you are getting through a DVD player interposed.
The first aim was to use its TBC.

Quote:
Originally Posted by neuron2 View Post
How are you connecting the DVD player to the PVR?
I have the choice between CVBS and S-video and I try to know which one give the less damaged results in my configuration (and not only in absolute) and if S-video is useful or if my Hauppauge will handle better than the dvd recorder the CVBS signal coming from the VCR.

I do this capture for a friend (it's his wedding tape) and I wanna do it the best I can (with the equipment I have).

Last edited by Music Fan; 8th October 2013 at 16:40.
Music Fan is offline   Reply With Quote
Old 8th October 2013, 16:33   #12  |  Link
Guest
Guest
 
Join Date: Jan 2002
Posts: 21,901
It's possible the interposed DVD could degrade the signal as well and so somewhat negate any improvement from the Y/C separation (assuming there is such an improvement). Anyway, assuming the VCR outputs only composite, it's clear that you must have an s-video from the DVD to the PVR to possibly gain anything. My eyes are probably worse than yours, so let's see if some others can comment on your captures.

Last edited by Guest; 8th October 2013 at 17:51.
Guest is offline   Reply With Quote
Old 8th October 2013, 17:01   #13  |  Link
Music Fan
Registered User
 
Join Date: May 2009
Location: Belgium
Posts: 1,743
Quote:
Originally Posted by neuron2 View Post
Anyway, it's clear that you must have an s-video from the VHS to the DVD, and from the DVD to the PVR to possibly gain anything.
Actually, I'm convinced now (after what I read above) that S-video is the best output on a VCR when one has it (I mean to connect to a capture card having it and CVBS), but as mine has only Scart, the decision about this choice is more difficult to take.
Anyway, I continue my research about the VCR Scart's ouptut to know what signal passes through it (the VCR is a Sony SLV-E830 and the dvd recorder is a Philips DVDR-70 for those who wonder).

edit : I found SLV-E830's service manual, still have to read it ;
http://elektrotanya.com/sony_slv-e830.pdf/download.html

Last edited by Music Fan; 8th October 2013 at 17:13. Reason: link added
Music Fan is offline   Reply With Quote
Old 8th October 2013, 17:14   #14  |  Link
sven_x
Registered User
 
Join Date: Oct 2011
Location: Germany
Posts: 39
Record a still image

You could record a still image like it is done here and then compare the output of that recording via s-video against SCART.

In my opionion at VHS recorders the SCART output carries mostly only composite signal, RGB are not connected.
sven_x is offline   Reply With Quote
Old 8th October 2013, 17:17   #15  |  Link
Guest
Guest
 
Join Date: Jan 2002
Posts: 21,901
According to the service manual that ancient VCR has only composite out on the SCART connector.

Maybe you can borrow/purchase a better VCR for this.
Guest is offline   Reply With Quote
Old 8th October 2013, 17:30   #16  |  Link
Music Fan
Registered User
 
Join Date: May 2009
Location: Belgium
Posts: 1,743
Thanks for this information, you were fast ! On which page did you find that ? I searched (rapidly) but didn't see it.
I don't know anybody who has a better VCR than mine and as I don't capture a lot of VHS, I'm not gonna buy a new one just for a few tapes.
But now we know what signal arrives to the dvd, one problem solved.
Still have to see if there is a difference between dvd and capture card's CVBS treatment.

Last edited by Music Fan; 8th October 2013 at 17:31. Reason: grammar
Music Fan is offline   Reply With Quote
Old 8th October 2013, 17:30   #17  |  Link
Music Fan
Registered User
 
Join Date: May 2009
Location: Belgium
Posts: 1,743
Quote:
Originally Posted by sven_x View Post
You could record a still image like it is done here and then compare the output of that recording via s-video against SCART.
Ok, I will read this.
Music Fan is offline   Reply With Quote
Old 8th October 2013, 17:49   #18  |  Link
Guest
Guest
 
Join Date: Jan 2002
Posts: 21,901
Quote:
Originally Posted by Music Fan View Post
Thanks for this information, you were fast ! On which page did you find that ? I searched (rapidly) but didn't see it.
The specs at the start define the pins used and there is only one pin for the video out, so that's that. But it's also confirmed by looking at the schematic (schematic page 3-2).

Quote:
Still have to see if there is a difference between dvd and capture card's CVBS treatment.
Yes, it will be interesting to know your results.

Last edited by Guest; 8th October 2013 at 17:51.
Guest is offline   Reply With Quote
Old 8th October 2013, 19:21   #19  |  Link
Music Fan
Registered User
 
Join Date: May 2009
Location: Belgium
Posts: 1,743
Quote:
Originally Posted by neuron2 View Post
The specs at the start define the pins used and there is only one pin for the video out, so that's that. But it's also confirmed by looking at the schematic (schematic page 3-2).
Wow, this schematic is chinese for me. Thanks, I wouldn't have found or understood this alone.
Music Fan is offline   Reply With Quote
Old 8th October 2013, 21:55   #20  |  Link
ChiDragon
Registered User
 
ChiDragon's Avatar
 
Join Date: Sep 2005
Location: Vancouver
Posts: 600
You need captures of the same frames in order to compare your options. There is no reason to include VCR <cvbs> DVDR <cvbs> PVR-150 in the comparison though.

To compare the Y/C separation (comb filter), the most useful test footage has low movement and high spatial frequency (sharp definition).

Last edited by ChiDragon; 8th October 2013 at 21:58.
ChiDragon is offline   Reply With Quote
Reply

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 18:21.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.