Moth3r

@resonator:
A "comb filter" simply separates the Y (luma) and C (chroma) data from a composite video signal. It does not deinterlace. (Sometimes interlacing artefacts are referred to as "combing" - confusingly, this is in no way related to a comb filter!)

Since digital video is stored as YUV or RGB, an analogue composite signal must be separated into Y/C before it can be digitized. 

@Orinoco_Womble:
This might be a useful read:
http://www.videohelp.com/forum/archive/hardware-video-processors-and-filters-t239205.html

That's because this was never released.

JD has been - understandably - preoccupied with having a new baby and moving house.

No point in re-torrenting a VHS-rip when the DVD is to be released on 4 August. Buy the DVD. (And don't ask for DVD rips on this forum.)

Unfortunately, even if you buy from an online store which tells you the media code, that's no indication of the country of manufacture (discs from both Singapore and India have the same code). The only way to be confident of what you are getting is to examine the packaging.

People say that the India discs are inferior. I wouldn't know - I'm a cheapskate who doesn't pay the premium for Verb discs.

1. It has been said that the V8000 is a very good choice; no fancy features like dual side play or AC3 output, just a clean composite output.

You have to put the video through a comb filter at some point in the capture process. In my experience, it is preferable to use a high-quality 3D comb filter, rather than using a basic filter then trying to fix the resulting rainbows and dot crawl in software.

2. I think I read on Doom9 somewhere that the Intensity Pro is designed primarily for HDMI capture, and the analogue capability is almost an "afterthought". Looking on their website I can't see any specs for what kind of comb filter is on the card. The X0 project used the PDI Deluxe, but  this card is quite old now.

Possibly the ideal would be to use a video scaler to convert the composite signal into component outputs (look on ebay, scalers are not as important in the digital age), and get a basic card that can capture component (Compro VideoMate series have 10-bit ADCs, the E500 can capture component).

3. Capture 720 x 480 using a lossless codec.

4. Avisynth/NNEDI.

5. It's a simple theory - analogue noise adds a random variance to the image, if you average out a number of different captures then you - in theory - end up removing the random noise and keeping the detail on the laserdisc.

Unlike other noise reduction methods, this should not be destructive - traditional spatial algorithms may destroy fine texture, temporal algorithms can cause motion smearing (which admitted can be reduced using more advanced "motion-compensated" filters, if you wanted to go that route instead).

MrBrown said:

As you can see the Cinematic Grindhouse Versions are very ugly.

I thought that look was intended by the directors, to give the impression of an out-of-focus, worn, faded and damaged print?

http://avisynth.org/warpenterprises/files/colourlike_25_dll_20050825.zip

Off topic, but - the BBC gave a similar reason as to why Doctor Who is not yet available in HD. Apparently, creating the required amount of CGI and effects shots in HD would have doubled the cost and time of post-production. (Even with the "serious hardware" that the BBC probably has access to...)

Sorry, when I said "allowed" I meant from the POV of the forum rules, not copyright law.

(Half your post title is missing. You may have suffered from the "editing a post with quotes in the title" bug.)

I like the idea in principle - instead of a title related to how many posts a user has made, they categorize themselves into the type of Star Wars fan they are. The drawback is that we are no longer solely a SW site, a wide variety of other movies are represented in the second fan edit/preservation forum.

Ripplin said:

Hey there.

Can anyone describe to me what "Addicted to Star Wars" is? Sounds interesting. :)

 

http://originaltrilogy.com/forum/topic.cfm/WOOKIE-GROOMER-PRESENTS-ADDICTED-TO-STAR-WARS/topic/7619/

Bobby Jay said:

I'm not sure if posting the link here is allowed so if anyone else is interested, please PM me.

 

Since it's an audio recording of something that's never been released, you are allowed to post the links.

zombie84 said:

Are these still the ones from way-back that were 96 kbps? I was always really dissapointed that the encoding was done so low. Would be great to hear these at 196 or higher.

 

Belbucus' recording of the 70mm mix - with the '93 mix in the other channel - was uploaded as an uncompressed WAV. (See this thread.) There was also a newer file synched to the GOUT video.

However, unlike morgands1's audio, the 70mm mix in Belbucus' file is only a mono recording.

Thanks for the answer - I've since found this comparison which seems to show the Fortune Star Region 3 DVD is better quality than the HKL release.

BTW is it just me or is http://www.dvdcompare.net/ having problems at the moment?

If the topic title contains quote marks ("), and you edit the first post, the text contained within the quotes is missing from the edited title of the topic (unless you put it back in manually).

Seems that it fizzled out some time last year, which is a shame.

Anyway, your post reminded me of a related question, for anyone who might know. Which DVD contains the highest-quality version of the Bruce Lee/Kareem Abdul-Jabbar footage?

I suggest you select a short scene, and record it several times with different noise reduction settings and different connections. Then compare the recordings frame-by-frame (copying screenshots your computer may help you with the comparison).

I did the same when I first got my D925. The noise reduction (not variable on the 925 - just a button labelled "HQ" that can be set on or off) softens the image too much for my liking. The s-video out gave better results than the composite (the Y/C separation in the player being better than the capture card).

Also, I assume I should I should use an S-video (instead of a composite) cable since the comb filter is probably better in my newer Sony DVD-recorder (RDR-GX7).... is that a correct assumption?

The second half is correct - the DVD-recorder probably has a better comb filter than the LD player. But if you use an s-video cable then you are using the comb filter in the LD player to do the Y/C separation. Unfortunately, even if you use the composite connection the signal will still go through the LD players comb filter as part of the players Digital Video Processing circuitry - the composite output is simply the Y/C of the s-video output recombined.

The only way to find out which gives the best results is to test different options.

Marshallator said:

I have a progressive scan DVD player and a progressive projector (10' screen), so that is likely not the problem.

Just a quick test: whack THIS as a DVD-video onto a re-writable disc or something and try playing it. See any flickering?

bkev said:

Regarding other "GOUT" changes, wasn't it discovered that at times fine details were blurred out?

That's not an alteration; just a mastering fuck up.

zombie84 said:

...This, I argue, is because the shot was re-composited for the foreign French/German prints because of the foreign crawls.

The French laserdisc (used in the Citizen and Mother transfers) has the English crawl (with French subtitles in the lower black bar). The German disc I believe does have a German-language crawl, and I do remember seeing screenshots from someone'e capture - but I don't think they went as far as the planet reveal.

Rainbowing is a comb filter artefact (caused by hi-frequency luminance information affecting the chroma demodulator). When I looked at the specs for the SAA7135 chipset it said it supported 3D (motion-adaptive) y/c separation, but also went on to mention a built-in 4-line adaptive (2D) comb filter.

What is the actual make/model of your capture card?

You could try using a comb filter in another device. For example, feed the composite signal from the LD player into a DVD recorder, then connect the s-video output from that to your capture card.

Max is right. If you capture a reduced luma range, then extend it afterwards with AviSynth, you can get banding effects. Conversely, the risk in calibrating the capture card brightness (black point) and contrast (white point) to grab the full range is that if something changes you may end up clipping the waveform. So it's important to check several sections, and to re-check when capturing the next side of the laserdisc.

You should definitely use the waveform monitor in the colortools plugin, because then you are not relying on your visual judgement or your monitor being correctly calibrated. I would say use the starfields as a reference for black rather than the letterbox bars - I remember a spot in Empire when the black level suddenly increased at a scene change, so the blackness of space became instead a washed out dark grey. And blaster flashes or explosions are a good reference for peak white.

See also:

http://en.wikipedia.org/wiki/Progressive_scan

Broadcast/professional equipment uses BNC connections for video cabling, to minimise signal degradation from back-reflection caused by impedance mismatch. The X0 team actually modified their PDI Deluxe capture card to swap the RCA with a BNC connector.

Arnie, in those clips you posted, did you crop then add the black bars back in? I'm asking because the black level of the bars is pure black, whereas the black in the actual video is elevated a fair bit higher. There is no clipping of the whites - in fact, I would recommend increasing the contrast.

Orinoco_Womble said:

Just been reading through the older posts again and came across this post from Laserman:

Laserman said:

Thanks for the PM Moth3r.
By that I meant the best way to fix the problem is to do it electronically rather than trying to filter it later.
Basically the problem with Laserdisc is that even the best players crap out at around 5 to 5.5MHz. This leaves laserdisc with a relatively slow rise time (the time taken for a signal on the video path to change from one colour to another) and can cause the 'ringing' or halos as well as awful transitions between colours.
To solve the problem somewhat you can map the Chroma signal to the Luma signal. The Luma signal is relatively quick, around 150nanoseconds, whereas the chroma signal lags with a rise time of around 800nanoseconds. By devising a circuit to lock the chroma signal rise time to the Luma's faster rise time the effect can be pretty much eliminated, resulting in a DVD like picture from laserdisc.

If you have a laserdisc with a set of colourbars on it you can see the problem between the Magenta and Green vertical bars quite clearly.
If you were smarter than me you could then work out a way to lower the rise time of the Luma signal with some sort of rise time acceleration and nail the halos forever.

 

Now, this sounds really interesting. Can anyone shed any further light on this whole 'remapping' business and how it might be achieved please?

The upshot of this discussion was that if you don't know how to build a rise time accelerator yourself, you should look on ebay for a Faroudja VS-50 or Crystal Vision VPS-1, both of which include similar bandwidth-expansion circuitry.

I brought this up a while ago; here is the post. Mike reckoned the same starfield was used when it was re-composited for the 1981, only flipped and positioned differently. (I could never see that , myself.)

Although hard to see, the telecine also confirms the GOUT has the original starfield.