Laserdisc capture, burned image is blocky/pixelated.

[Jack_Dante]

Hope this is in the correct forum, been trying to do a DVD backup of one of my old laserdiscs.

Have been using Wincoder and have set it to capture at NTSC DVD quality..

MPEG-2 format.
Capture size : (720 x 480 )
Video bitrate : 6400 kbit/sec
Audio bitrate : 224 kbit/sec
Total bitrate : 6817 kbit/sec
Total record time : 163 minutes

I've recorded side 1 and 2 as seperate Mpeg files and am using Ulead DVD movie factory 3 to join the files togethor and burn them to disc.

Although the mpeg files look OK when played on my hard drive, when burned to disc and played back the picture looks somewhat "blocky" and "pixelated". As it's a fairly long film (about 2hrs) I'm assuming its due to my software having to compress the image down to fit on the disc.

Films taken from a VHS source don't seem to have much pixelation, although the ones I've burned so far have only been about 90 odd mins in length. I was therefore wondering if altering the capture settings or using some other method would provide less pixelated results?

Hope you'll forgive the newbie question and thanks in advance.

[Epicurus8a]

Try this:

In Movie Factory there is a menu for selecting the video template you want to use. There is also an option to choose "do not convert mpeg compliant files" (or something like that). Be sure to check (tick) this box before burnning.

NOTE: You may have use a bit rate calculator to make sure your captures won't be too big for your DVDs.

Good Luck!

[spectroelectro]

Use a high quality MPEG-2 encoding program like TMPGenc or MainConcept or Quenc, and if the video's too large shave it down with the freeware ReJig. You can squeeze files by at least 25% with ReJig without causing any visible compression artifacts at all. Quenc and ReJig are free open source software, so your cost for those is zero.

[edDV]

Mostly you are seeing the difference between a 90 min and a 163 min compression.

Why not burn 2 DVD discs (or a Dual Layer) for this one?
Single side media is cheap.

[non-linear]

Looks like it's being reencoded. If you use the bitrate calculator on this site, you will see that 163 minutes (1hr 43min) with audio at 224Kbps, your video bitrate will have to be 3512Kbits/sec. Your video wouldn't fit on one single layer disc otherwise.

[ofbarea]

I capture laser disk using my ADS A/V link. It captures DV type 2 avi files. The resolution is lock at 720x480.

In VirtualDub and depending on the source movie I apply some filters. During this process I usually trim 8 pixels from each side and frame serve the movie to TMPGEnc. Usually the final MPEG has a resolution of 704x480. When I need to

If I notice that the AVI DV file is the one containing the pixilation (due capture errors and/or DV compression) I use “MSU De block” filter in Virtual Dub.

If I still get annoying pixilation in the final MPEG file, I check the Soften Block Noise control and increase the motion search precision to High quality in TMPGEnc MPEG settings.

[Rufus]

hi there,

i have a question about laserdisc transfer using virtual dub 1.4.9.2. sync,

what is the correct framerate to use: 30 fps or 29.97 fps ?

if i know correctly CAV laserdiscs have 30 fps
(30fps*1800rpm=54000frames/30min per side)

what about CLV/CAA lds ? 30 or 29.97 fps ?

does the player output 30 or 29.97 fps ?

after the transfer i want to restore the 24 progessive source (if possible)
..is there a good virtual dub filter or is it better to use tmpgenc ?

thanks in advance for your help

Rufus

p.s.: sorry if someone asked this before

[ofbarea]

I really think that NTSC Laserdisc players state that CAV disks spin at 1,800 rpm to facilitate calculations. This number is easier to use than 1,798.20 rpm; in the same way hard disk manufactures usually define Megabyte as 1,000,000 bytes instead of 1,048,576 bytes.

CLV has variable rpm, from 600 to 1800. This is the reason why ordinary LD players have less functions when playing CLV disks. This does not affect the NTSC output. It should produce standard 29.97 fps.

For most movies the TMPGEnc 3/2 pull down works properly, but there are exceptions, especially in Japanese made imports. Movie ratings are stricter in Japan than in the US, because of this, some movies have been edited to maintain the same US movie rating. Since these changes usually happens in the 30 fps version of the movie. This breaks the pull down sequence and if this is the case neither VirtualDub, nor TMPGEnc will be able to convert the movie to 24 fps properly.

Also movies slightly longer that 2 hours are "electronically time compressed" by dropping a field here and there to reduce the play time to fit in a standard CLV disk. In this case the 3/2 pull down will not work properly either. For more details check this link: Time compression

PAL disks have more capacity than NTSC, 36 minutes for CAV and 64 minutes for CLV.

[vhelp]

Laserdisc's are a little more sensitive (IMO) when it comes to
archiving them to DVD through capturing.

When you start on a LD project, its wise to make sure your LD
lens are clean. (Thats hard for me, since I don't have a LD
cleaner, but TBO, I'm just too lazy to pick one up) anyways.

After the above, then your next thought should be weather or
not to capture your LD through Composite (RCA) or S-Video wires.
IMO, its best to capture from S-Video connection IN/OUT, but
some users claim prefered results w/ the Composite connection.
It's a matter of taste, but I suspect its Trial 'n Error with
a little eye'ing at the source to arrive at a final conclusion
which to use.

Then, after the above two, you want to capture each side of the
disks, and tie them together by joining the AVI (never MPEG)
and encode them.
.
The problem with this approach of joining, (if you want to IVTC
with perfect results) is that you have to make sure of the
following:

* no dropped frames during capturing, and
* with-in the video source (from LD) the frames are perfectly
telecined.

If the frames are not perfectly telecined, then you will have
trouble with the IVTC step (from the joining of AVI files) during
the encoding phase. The Telecine pattern should always be:

* 3,2

That is ...

PPP II PPP II PPP II PPP II ... thoughout the AVI source files.
When you join your AVI's, they (3,2 patter) should flow evenly
throughout the join process. If they don't, you'll know, when
you starting seeing Interlacing patterns during play. Thus..
given two capture AVI projects (Side A, and Side B) and your
AVI files flows like this:

Side A - PPP II PPP II PPP II PPP II
Side B - PPP II PPP II PPP II PPP II

When you join them, they would become:

PPP II PPP II PPP II PPP II : PPP II PPP II PPP II PPP II

The ":" is where you joined the Side B to Side A 's captured AVI
file.

Here's another scenario. Given:

Side A - PPP II PPP II PPP II PPP II
Side B - II PPP II PPP II PPP II PPP

When you join them, they would become:

PPP II PPP II PPP II PPP II : II PPP II PPP II PPP II PPP

..would throw off the IVTC pattern process and you would altimately
have Interlace patterns (I sometimes call, misfires) And, if you
come up against this situation, then you will have problems with
your bitrate (and pixelation) in your final MPEG project, because
it takes more bitrate to (try) and overcome the Interlace (which
would now be seen as noise in your AVI, during the MPEG encoding
phase. This will be more evident in VBR encodings. CBR encodings
are a little more forgiving, if your bitrate is high enough, say
9000 or higher. I prefer to use a high bitrate and CBR becuase I
want to reproduce with closest to duplication of the source as
possible.., and only a CBR 9000 bitrate (or higher) can do this.
VBR (in the end) will leave you with too many pixels in your final
MPEG project.
.
But most of this is all dependant upon which Encoder you use in
your process.

* I prefer TMPGenc for all my Film projects. As long as the source
is properly (and perfectly) telecined, TMPGenc is *the* encoder to
use (IMO)

* For pure Interlace-only sources (ie, DV) Procoder is the way to go.
Pure interlace is when every frame is Interlaced. You can *not*
miss them. Frames will look distored (mostly in any motion areas)
where Film sources will have 3 progressive clean frames and 2 Interlace
(dirty) frames. But, again, pure interlace is where you see every
frame is interlace, and NO SIGN of progressive frames.

Now, as for your Laserdisc sources, these are more than likely, Film.
so, as long as *your* given movie is not screwed up in the Telecine
process, *and* your capture produce ZERO frame drops, *and* your
segmented AVI(s) are able to flow smothly ..

( ie, SA:SB - PPP II PPP II : PPP II PPP II ... )

Then you should be very good to go. But I warn you.., this is really
for experienced users. nebies will always have questions.., because
most sources are all difference (with the issues I laid out above) and
then some.

Laserdisc transfers can be very fun (and easy) to do. Enjoy your
endeavor.

From the Video Workstation of,
-vhelp 3324