Quantcast Best workflow for SECAM-L digitalization? - digitalFAQ Forum
  #1  
01-21-2021, 06:02 PM
qcalv qcalv is offline
Free Member
 
Join Date: Dec 2020
Posts: 8
Thanked 0 Times in 0 Posts
Hello all,

I want to digitalize VHS recorded for most of them in SECAM-L (France). Some are in PAL too.
I have read quite a lot on this forum regarding S-VHS decks, TBCs, capturing, restoring and so on.

I am interesting in the capturing process and restoring work (or at least some basic filtering to get simple optimization) but I do not look for perfect workflow/setup as I do not have unlimited time to spend on this project and it will make no real sense for 10 VHS that I could let digitalize professionnally .

Nevertheless, I want to loose through the capture process as mimimum quality as I can.
I did not manage to find easily (I live in Germany) low budget solutions (~200 EUR) compatible with SECAM-L. So I am interested in what I can do considering what gear I have. And maybe if there is easy update I can do which can bring a lot to my workflow.

I have a VHS player Samsung SV125FK (SECAM-L compatible).
I have a DVD recorder Pioneer DVR 320-S for PAL conversion (usable as pass-through).
I have a CANON MV530i video camera which I could use to convert PAL to DV (SECAM-L does not work as input)
I have a Pinnacle USB-500 that I can use to capture on my laptop.
I have a laptop which does not have any Firewire, but could get an Expresscard Firewire.
I have a TV Toshiba 32SL738G.

First question beforehand regarding what I observed:
1. When I plug my VHS player (Composite) to the Pinnacle USB-500 and capture with Virtual Dub, I get worse quality (on one VHS where I know it does not look too good - blue noise) than when I look at it on my TV. I cannot output the digital signal from my TV and capture it. I believe there is some filtering happening there. My question is: when I will be capturing in DV, will I be able to manage in software such nice filtering, or this can only happen with some hardware on analog signal before the ADC conversion?

2. What makes more sense between:
a. VHS (composite) > DVD recorder (composite) > Video camera (DV) > Laptop (Firewire)
b. VHS (composite) > DVD recorder (composite) > Video camera (DV) > Pinnacle USB-500 > Laptop (USB)
c. VHS (composite) > DVD recorder (composite) > Pinnacle USB-500 > Laptop (USB)
What happens with the Pinnacle USB-500 if I plug in some DV signal? It just make a pass-through to my laptop?
What is the difference in terms of performance between the ADC in the Canon video camera and the Pinnacle capture box?

3. I have a S-Video output on the DVD recorder. I tried to do following as I understood S-Video is better as composite. Is that correct for this case?
VHS (composite) > DVD recorder (S-Video) > Pinnacle USB-500 > Laptop (USB)
This did not work. I do not get any capturing working. Somehow it seems that the DVD recorder cannot get the SECAM-L on the S-Video output. Does it make sense, or am I doing something wrong?

3. I have various capture software available (Adobe Premiere Pro, Virtual Dub, and Pinnacle Studio 24). Does any of these solutions make any difference (in terms of performance) when capturing from the Pinnacle USB-500?

4. Is there a best solution between DV > Laptop with Expresscard Firewire or DV > Pinnacle USB-500 > Laptop with USB?

That would be it for now. I have some further questions regarding the bluish lines/noise I am seeing on one of my VHS but I will keep this for later .

Thanks in advance for any help.

Fran
Reply With Quote
Someday, 12:01 PM
admin's Avatar
Ads / Sponsors
 
Join Date: ∞
Posts: 42
Thanks: ∞
Thanked 42 Times in 42 Posts
  #2  
01-21-2021, 06:53 PM
hodgey hodgey is offline
Free Member
 
Join Date: Dec 2017
Location: Norway
Posts: 1,055
Thanked 250 Times in 215 Posts
Quote:
Originally Posted by qcalv View Post
My question is: when I will be capturing in DV, will I be able to manage in software such nice filtering, or this can only happen with some hardware on analog signal before the ADC conversion?
When capturing the video you lose the non-visible parts of the image, which are used to stabilize and decode the color signal. You can reduce noise and adjust levels and such after capture, but you can't recover color that wasn't decoded properly or fix wiggle or lost frames.

Quote:
c. VHS (composite) > DVD recorder (composite) > Pinnacle USB-500 > Laptop (USB)
Of those I would use c, the DVD-recorder should already output a signal that has been digitized and turned back to analog. Adding in the camcorder as well will only introduce DV compression for no reason. As far as I know, it's the DVR-x3x and newer Pioneer models that have decent video decoding that can help correct wiggle and such, I don't know how well the older ones work.

Quote:
3. I have a S-Video output on the DVD recorder. I tried to do following as I understood S-Video is better as composite. Is that correct for this case?
Yes, S-Video puts the color signal on a separate wire, so you avoid issues causing with the color and brightness signal interfering with each other. It is however a bit easier for the capture card to separate Y/C in a signal coming out of the dvd-recorder compared to directly from the VCR, as the signal from the DVR should be stable. I don't know how noticeable Y/C artifacts are with SECAM compared to PAL and NTSC, as the color is encoded in a very different way. It's odd that you don't get S-Video out though, maybe there is some setting for it? Does it work with PAL? I don't know whether it outputs SECAM or PAL on the composite/s-video outputs.
Reply With Quote
  #3  
01-25-2021, 12:27 AM
qcalv qcalv is offline
Free Member
 
Join Date: Dec 2020
Posts: 8
Thanked 0 Times in 0 Posts
Hi Hodgey,
thanks for your reply!
Regarding the Pioneer Model, I found it on another forum recommended for VHS digitalization. I will check the models you mention.
Regarding the S-Video, I thought the DVD Recorder is anyway giving output in PAL. But I will have a look again how to output S-Video on the Pinnacle USB-500.
Reply With Quote
  #4  
01-25-2021, 11:29 AM
latreche34 latreche34 is offline
Free Member
 
Join Date: Dec 2015
Posts: 1,435
Thanked 232 Times in 214 Posts
Quote:
I have a Pinnacle USB-500 that I can use to capture on my laptop.
I have a laptop which does not have any Firewire, but could get an Expresscard Firewire.
No need to any extra hardware, the USB-500 is a port converter, It takes a firewire device and connect it to the computer via USB, it even retains camcorder function controls such as play/stop/record ....etc.



Attached Images
File Type: jpg Diagram.jpg (34.6 KB, 40 downloads)
Reply With Quote
  #5  
01-26-2021, 02:57 PM
qcalv qcalv is offline
Free Member
 
Join Date: Dec 2020
Posts: 8
Thanked 0 Times in 0 Posts
Ok, got it for the USB / Firewire interface. Thanks.

Regarding the S-Video input, I retried today.
I get a signal out of the DVD recorder to my TV without any trouble (I do not have a S-Video input, so I use a scart adapter which gets S-Video in, plus Component Audio Red/White). That works.
When I plug it into the USB-500, does not work in Virtual Dub. Signal detected = 0. When I plug the component video (yellow) into the USB-500 then I get Signal detected = 1, 625 lines, and it is coded in PAL what I get from the DVD recorder.
1. What am I doing wrong in Virtual Dub? Is there any configuration to setup that the signal comes as S-Video in and not component for the USB-500?
2. Do I loose anything by getting the SECAM-L converted into PAL within the DVD recorder?

Thanks for the help.

fran


Attached Images
File Type: jpg VDUB_Component_PALB.JPG (27.2 KB, 0 downloads)
File Type: jpg VDUB_SVideo_PALB.JPG (25.1 KB, 2 downloads)
Reply With Quote
  #6  
01-26-2021, 03:18 PM
latreche34 latreche34 is offline
Free Member
 
Join Date: Dec 2015
Posts: 1,435
Thanked 232 Times in 214 Posts
Lookup crossbar for 500-USB. It switches from composite to S-Video since the driver lacks it.
Reply With Quote
The following users thank latreche34 for this useful post: qcalv (01-26-2021)
  #7  
01-26-2021, 04:05 PM
hodgey hodgey is offline
Free Member
 
Join Date: Dec 2017
Location: Norway
Posts: 1,055
Thanked 250 Times in 215 Posts
Quote:
Originally Posted by qcalv View Post
2. Do I loose anything by getting the SECAM-L converted into PAL within the DVD recorder?
The video is decoded into digital component (Y/Cb/Cr) internally, so it doesn't really make any difference what tv standard the output is as long as you are using s-video.
Reply With Quote
  #8  
01-26-2021, 05:21 PM
qcalv qcalv is offline
Free Member
 
Join Date: Dec 2020
Posts: 8
Thanked 0 Times in 0 Posts
I managed now to capture from S-Video bei either using Pinnacle Studio 24 which sets the right setting on the USB-500 or by using the CrossbarThing utility before starting VirtualDub. Thanks a lot!

My target is to be able to edit, restore in the future and I do not have real restrictions with capturing lossless (from HDD size point of view).
In Pinnacle Studio 24 I can use DV, MJPEG or MPEG2.
In VirtualDub I would use Lagarith.

1. Is there any best solution between Lagarith and huffyuv? Or it does not matter for what I want to do?
I read in other posts that DV compared to MJPEG is not so good. MJPEG is getting old and I remember having issues in the past to handle easily the files. MPEG2 would be the best as I read.

I would rather capture without any "live" compression, does it make sense for editing/restoring/reworking afterwards, or it is not a big deal since VHS quality is anyway not tremendous?

2. Is there any cons of archiving anyway the lossless captured file (eg. between Lagarith or huffyuv)?

3. I started this project because my dad made some quick transfer from VHS to DVD using a VHS/DVD deck all-in-one. I found the quality bad, seeing compression blocks, and I was assuming I can get better results by capturing on laptop and then compressing myself with a better codec.

I attached one sample from the DVD VOB file my dad created with the combo deck.
I attached the same sample that I captured in Lagarith with the USB-500.
I converted both with H.264 (Single pass, CQP, 23, Tuning Film) to post small file size.
It is OK for artefacts analysis or do I compress too much the samples?

a. In general, at beginning a lot of bluish lines appears, then it becomes better.
What are these? Are they from the VHS recording? Or VHS playback?
I see them a lot on some of my VHS.
Can I eliminate them later on using filters with Avisynth and scripts?
Or shall I take care of this before hand through my capture workflow?
b. Will I get better results when capturing with Lagarith, editing, and then compressing the file with another codec than MPEG2?
c. I tried to capture with Pinnacle Studio 24 directly in MPEG2 and the result seemed anyway much better that what the DVD VOB shows. So was the setting on the VHS/DVD combo deck just bad?

4. When I look at the samples with VLC, I see "Z-lines" on the left hand. Is it because of Interlacing? Do I need to deinterlace before archiving the file? I understood this has to be done before applying restoring filters and stuff, but I am not sure for keeping the "raw" file? Is it just a matter on how the player handle the frames?


Attached Images
File Type: jpg Interlacing.JPG (14.4 KB, 2 downloads)
Attached Files
File Type: avi Sample_VobToH264.avi (36.28 MB, 1 downloads)
File Type: avi Sample_LagarithToH264.avi (54.80 MB, 1 downloads)

Last edited by qcalv; 01-26-2021 at 05:31 PM. Reason: Added next sample
Reply With Quote
  #9  
01-26-2021, 07:03 PM
hodgey hodgey is offline
Free Member
 
Join Date: Dec 2017
Location: Norway
Posts: 1,055
Thanked 250 Times in 215 Posts
1+2:
Lagarith compresses a bit better, but is also a bit more cpu-heavy than huffyuv. DV vs MJPEG vs MPEG2 would depend on compression level of mjpeg and mpeg2 (the compression level of DV is fixed). There isn't really any downside to using lossless codecs for capture if you have the hard-drive space for it other than that the aren't well supported in big name commercial video editors (though there are more modern codecs you can convert to later if that's needed.)

3

h.264 with decent bitrate can be fine for displaying artifacts. Since the video is interlaced you do however want to tell the encoder to encode the video as interlaced so it displays correctly in e.g VLC and the compression doesn't mess with the interlacing. There should be a setting for it somewhere in the encoder. You would also want to use 4:2:2 or 4:2:0 chroma sub sampling. 4:4:4 H.264 is not well supported, and the captured video will be 4:2:2 (or 4:2:0 for the mpeg2 from the DVD) anyhow.

a. It's normal for there to be lots of color noise at the start of a VHS recording if it was recorded over something. There is a little distance between the video head that erases the video track and the video heads on the drum that write to it. That short part would not be erased before recorded on and as such some of the old signal will be left and cause flashing colors. High end editing VCRs had what was called a flying erase head on the video head drum itself that helped avoid this, but consumer decks did not. It looks a little different to what I'm used to in PAL/NTSC, but I presume it's the same thing you are seeing.

Color on VHS tapes is inherently noisy due to how the format works, though VCRs sometimes had extra noise reduction circuitry built in to reduce it, especially high-end SVHS decks. SECAM is a bit different to PAL/NTSC though so I don't know if it was as effective for that. DVD-recorders sometimes had fancy noise reduction as well, I think the pioneer ones gives some options for it in the menus. I don't really know what degree of noise is normal for SECAM though. You can reduce color noise a fair bit with post-processing with e.g avisynth in any case.

I believe the solid color at the right of the image is something typical of SECAM, I'm not sure if it is inherent to the format or something to do with the video decoding hardware though.

b. Yeah, generally you want to capture to a lossless format, do processing/filtering/editing, and then export to H.264 (or similar) for viewing.

c. That's very possible, DVD-recorders have different quality settings, which would determine how much you could fit on one DVD. The sample looks heavily compressed. Additionally the quality of the mpeg2-encoders would vary.

d. With lagarith and huffyuv VLC won't really know if the video is interlaced, for arching the raw file that doesn't matter, but for final playback you either want deinterlace with a quality deinterlacer before encoding to e.g H.264, or skip that and encode the video as interlaced (ok if it's for viewing on a TV or similar, but not suggested for e.g youtube upload).
Reply With Quote
  #10  
01-27-2021, 01:00 PM
qcalv qcalv is offline
Free Member
 
Join Date: Dec 2020
Posts: 8
Thanked 0 Times in 0 Posts
Thanks Hodgey for your clarifications.

Now I am facing desynchronization audio/video.

To summarize I have:
Samsung SV125FK > [SCART] > DVD recorder Pioneer DVR 320-S > [S-Video + Composite Red/White Audio ]> Pinnacle USB-500 > USB 2.0 (I think) > Laptop > VirtualDub 1.10.4 32bit > USB 3.0 > SSD 128GB

My laptop is:
OS Windows 7 Pro SP1 64 bits
MOTHERBOARD Lenovo 2355CTO
CPU Intel i5-3230M CPU @ 2.6 Ghz
RAM 8,00GB Dual-Channel DDR3 @ 798MHz
AUDIO Realtek High Definition Audio / Intel Display Audio

I shutdown internet with TinyFirewall, disabled Antivirus Avast, saw CPU usage < 40%.

Reading the other posts on this forum concerning the configuration of the timing menu in VirtualDub, I tried three configurations (see attached).
All three of them does not work.
Config1: I get totally crazy audio (accelerating/decelerating). The result show me huge amount of dropped and inserted frames.
Config2: Similar as Config1
Config3: Audio seems shifted after 15min few seconds at least
No Audio playback, Overlay.

Any idea where I should tweak or change configuration?
Can it come from the external SSD I use?
Sometimes I see the counter in VirtualDub freezing for few seconds and starting again during Capture.
What does it mean? Just display wrong or audio/video capturing is delayed then as well?

I understand the USB-500 is doing the sync audio/video, so what do I need to configure in VirtualDub?

Thanks.


Attached Images
File Type: jpg CaptureResult1.JPG (29.5 KB, 3 downloads)
File Type: jpg CaptureResult3.JPG (29.1 KB, 3 downloads)
File Type: jpg TimingConfig1.JPG (51.2 KB, 1 downloads)
File Type: jpg TimingConfig2.JPG (50.1 KB, 1 downloads)
File Type: jpg TimingConfig3.JPG (50.2 KB, 1 downloads)
Reply With Quote
  #11  
01-27-2021, 01:17 PM
latreche34 latreche34 is offline
Free Member
 
Join Date: Dec 2015
Posts: 1,435
Thanked 232 Times in 214 Posts
Try AmarecTV, it is a lean capture app, this is just to rule out a CPU/memory problem.
Reply With Quote
  #12  
01-27-2021, 02:32 PM
hodgey hodgey is offline
Free Member
 
Join Date: Dec 2017
Location: Norway
Posts: 1,055
Thanked 250 Times in 215 Posts
Hm, in your screens you get like average 24 fps, so you are getting drops/inserts in the 3rd capture as well.
Reply With Quote
  #13  
02-01-2021, 12:38 AM
qcalv qcalv is offline
Free Member
 
Join Date: Dec 2020
Posts: 8
Thanked 0 Times in 0 Posts
I tried now AmarecTv, but I can't get it working. Always throwing this error message.
I tried few different options in the graphs panels, the Pinnacle Box is recognized.
I checked that it is still capturing over VirtualDub fine.
Why would I get less than 25fps? Because of CPU? Even if it is around 30% in use only during capture?


Attached Images
File Type: jpg AmarecTv.JPG (25.5 KB, 2 downloads)
Reply With Quote
  #14  
02-14-2021, 01:15 PM
qcalv qcalv is offline
Free Member
 
Join Date: Dec 2020
Posts: 8
Thanked 0 Times in 0 Posts
I realize that the problem comes (among others) from the external SSD drive.
I managed to get 25fps by capturing 30min on my internal SSD.
However, when I tried to extend this to 1h, I get again desynchronization (24.98fps).
Is there any tool to check when this desynchronization happens, and why, like ressource monitor tools?
Can I easily correct this desynchronization afterwards, or am I stuck to capture only 30min by 30min packets of data? I attached both results and the configuration I used.

Other questions concerning the workflow:
I am now playing with AvsPmod to test few Avisynth scripts with the huffyuv avi file (like crop/adding borders, removing blue chroma noise).
I want later on to be able to edit with Adobe Premiere CS4 (cut/encode) and probably make some additionnal corrections (color), finally encode with Adobe Media Encoder for long term storage.

Do I need an intermediate file after the Avisynth processing, if yes what kind of codec?
Can I do this in one run from Adobe Premiere, like editing the original huffyuv file, then running it through Avisynth, then encode it?

What makes sense in this case?

Thanks.
Reply With Quote
  #15  
02-14-2021, 01:26 PM
themaster1 themaster1 is offline
Free Member
 
Join Date: Feb 2011
Posts: 241
Thanked 51 Times in 43 Posts
Try Virtualvcr (different than virtualdub), i still use it personnally, always have great results (on XP/7).

There is an option resample audio dynamically, you should try this

but 24.98fps that's close to 25fps with an avisynth script that should do it

avisource()
assumetff()
assumefps(25.000)
Reply With Quote
Reply




Tags
secam, vhs capture

Similar Threads
Thread Thread Starter Forum Replies Last Post
MSECAM (SECAM B) and TBC? enas.nikos Capture, Record, Transfer 2 05-27-2020 01:00 PM
Best starting method for newbie (VHS digitalization)? chrisbati General Discussion 2 03-28-2019 02:41 PM
CVBS capturing, VHS digitalization, best capture card? PMVcr Capture, Record, Transfer 30 11-28-2018 03:58 AM
SECAM-L conversion to DVD? photoguy1234 General Discussion 3 10-28-2016 03:15 AM

Thread Tools



 
All times are GMT -5. The time now is 09:58 AM