logo Sign In

Warp99

User Group
Members
Join date
22-Dec-2004
Last activity
18-Dec-2005
Posts
38

Post History

Post
#169180
Topic
~French Coffret Trilogie LD to DVD transfer~ (Released)
Time
Originally posted by: Moth3r
Originally posted by: Karyudo
Hmm... interesting points.

The way that both NTSC and PAL versions have the same video flaw could be because they come from the same HD master tape. I'm not saying I'm sure they did, but they could have. There is a good chance that the master is HD. Even if that was dubbed down to sub-masters, those sub-masters could show the same flaw if it was added early on in the chain.

On the other hand, knowing that Faces has this same flaw, and that it's a frame before a cut, I'd have to agree that there is a stronger possibility that it's on the film itself.
I'd vote it's a video flaw. The locations of these dropouts do not generally coincide with cut/splice points, e.g.:

http://img202.imageshack.us/img202/7682/line2cy.jpg

And why do they not appear on the VHS? My '95 VHS set is from the same print (i.e. famous burn marks in first Tantive shot) but does not have these line dropouts present.

Yes the line you are showing must be a dropout, it's a bit strange though since most player are equipt with a dropout correction circuit that should replay the last correctly scanned line if the signal from the laser is lost. I've never seen it happen though but then I've never seen many (any) line dropouts on my discs either, but anyway it should be possible to see the duplicated lines in the capture if the player tried to correct a dropout, could be that this dropout maybe is not severe or long enough to trigger the correction circuit. Descratch should be able to safely remove most of them if they like this one is almost totally white or totally black.

The lines shown by Arnie.d in frames 20374 and 22939 could not be a video dropout, as they are in both the NTSC and PAL releases even the SE and I would guess even in your '95 VHS

Post
#168336
Topic
~French Coffret Trilogie LD to DVD transfer~ (Released)
Time
Originally posted by: Arnie.d Well, it's in my capture for sure and I can't remove it with Moth3r's script for some reason and I have absolutely no idea whatsoever what "minlen=700,asym=60,maxgap=50,maxwidth=1,mindif=7,blurlen=8,keep=0,border=0,maxangle=0,modeY=3" means so I can't adjust it.

Read all about it here but finding the right settings really is a tedious trail and error thing so if it's only these lines or maybe a few more you want to remove, I would use photoshop and remove them 'by hand' and then use imagesource to insert the edited frames, IMO it's faster and safer than descratch.

Although if your capture has many random line dropouts, descratch could be tuned to remove most of them if they are distinct enough.

Post
#168317
Topic
~French Coffret Trilogie LD to DVD transfer~ (Released)
Time
Originally posted by: Karyudo
Originally posted by: Citizen
They also appear on the German PAL THX discs too, the drop outs it would appear are on the actual celluloid footage and not a result of the film to LD process.


I'd have to disagree. That really looks like a video problem, not film. Film problems are more commonly vertical, because that's the direction film moves. A perfectly-horizontal (i.e. one scan line) flaw like that must be a telecine artifact.

The fact that these flaws appear on both German and French (and Spanish?) versions is almost certainly because they're from the same master tape.

They also appear on my US Faces discs, I guess then they would also be on the DC set.

The line in frame 22939 could be from a reflex or lights from the instrument panel.

The line in frame 20374 appears in four frames and clearly looks like a dropout of some sort, but how could the same dropout appear in both the PAL discs and the US discs ? ? - IMO not possible, the last frame containing the line is next to a cut and that would indicate that the line could be a horizontal scratch from the editing table and maybe it's been there since 1977.

Post
#165150
Topic
~French Coffret Trilogie LD to DVD transfer~ (Released)
Time
Originally posted by: Moth3r
Originally posted by: Arnie.d
I tested a piece of silence between the fox logo and the crawl.

Digital: RMS: -96.0, Peak: -68.0
Analoge: RMS: -96, Peak: -70.3 Hmmm, I would expect the digital peak to be lower than the analogue.
Originally posted by: Arnie.d
When I play the files (between the logo and crawl) and I can clearly see that the -db value is higher for the analoge capture, so it's more silent right?
No, a higher value means more noise. A lower value (by lower, I mean of course "more negative") indicates less noise.


Could it be that the 'silent piece' is not that silent at all but contains a wide spectrum background noise from the film and recording equipment and also the LD-player, so if/since the digital input has better bandwidth it will simply pick up more of this source noise. In any case I would be surprised if the unweighted signal to noise ratio of the SW movies soundtrack is much better than 65-70dB (higher S/N ratio = lower noise of course).

A digital input should also give less distortion, none if it really is bit-accurate ?

Post
#162704
Topic
.: The XØ Project - Laserdisc on Steroids :. (SEE FIRST POST FOR UPDATES) (* unfinished project *)
Time
I guess 'adaptive' could be line-adaptive or motion-adaptive or even system-adaptive in any case I don't think any of the philips saa71** chips has a 3D combfilter.

The Compro Videomate Ultra card i.e. is based on saa7134 but it uses a NEC 3D combfilter chip and only for NTSC. And also a 3D filter would require a frame buffer and this is usually a separate memorychip on the card.

Post
#162669
Topic
.: The XØ Project - Laserdisc on Steroids :. (SEE FIRST POST FOR UPDATES) (* unfinished project *)
Time
Originally posted by: nin
I don't have any BNC to RCA cable home right now, only a BNC to BNC so I cannot test it right now. Hope to do so soon.

Btw, I have just received info that the PDI delux card have "only" a Adaptive 2/4-line comb filter for two dimensional chrominance/luminance separation.


Yes the philips saa7118 capture chip used on the PDI card 'only' has a 2 line 2D combfilter for NTSC (the 4 line mode is for PAL), but still I think the saa7118 is quite good at least regarding the bandwidth (details and sharpness).

It'll be interesting to hear what you find out when you connect the X0 BNC output to the PDI card.

Post
#162048
Topic
.: The XØ Project - Laserdisc on Steroids :. (SEE FIRST POST FOR UPDATES) (* unfinished project *)
Time
And here is something I found here:

Back to the interesting subject of composite re-composition, the HLD-X0
and HIL-C2EX specifically mention that one of their composite outputs
(the one with a BNC plug) is a "direct out" composite with a different
path than the other composite output (with a regular pin plug). The
manuals explain that nothing but TBC is applied to this "direct"
output. In particular, the Sony manual says that screen info will not
be shown through the direct output. I thought it meant the regular
composite output was a recombination, while the direct would not be. As
other decks do not mention this, i also thought it meant the other
decks only had "regular" (recombining) outputs. Can someone
explain/confirm in opposition to what these outputs are "direct", what
else than TBC is applied to the composite "normal" outputs, and if
outputs on some hi-end models (X9 etc...) could be "direct" even though
the manual doesn't mention it ? The X0 even had a flow-chart explaining
this feature, maybe someone could post a scan of it ?


Post
#151395
Topic
.: The XØ Project - Laserdisc on Steroids :. (SEE FIRST POST FOR UPDATES) (* unfinished project *)
Time
Originally posted by: zionI don't know what we're planning at this point.

As you stated earlier ESB and ROTJ does not appear to need as much cleaning up, so maybe when you're done with ANH and all your skills and techniques are perfected, ESB and ROTJ could be quite quicker to do and the "when/if it is evident" thing might as well be for the complete trilogy. (Let me see, I used appear, maybe, could, if, and might so I guess I was just speculating.)

BTW, I think "Ø" is Danish or Swedish and means island but it could also be the sound of confusion

Post
#151032
Topic
.: The XØ Project - Laserdisc on Steroids :. (SEE FIRST POST FOR UPDATES) (* unfinished project *)
Time
Originally posted by: bongloads I am most impressed by the naturally clear and colourful capture of the HLD-X0 - breathtaking.

Yes it really is good, I think the X0 must more or less be the only player that can reproduce the full bandwidth of laserdiscs without much (any) noise or artefacts in the picture.

Also very nice cleanup work, by removing as much dirt as you can the picture will be much more enjoyable even though a laserdisc will never look as sharp as a dvd. But by all means keep the original ('glithces') if you can distinguish them.

I'm very sure when we get to see the fully color and brightness corrected result of the X0 project, it will almost look like a real dvd and certainly be just as enjoyable

Post
#150509
Topic
.: The X0 Project Discussion Thread :. (* unfinished project *)
Time
Originally posted by: zion
I've uploaded the last newsletter article to the web site. The next issue will be out in the next couple days, so sign up now at the site if you haven't already or you'll miss it.

The first of the new articles we've been talking about will be up this week as well, so stay tuned.

I almost missed your announcement Zion I've been waiting for this newsletter you know
Post
#150079
Topic
Original Trilogy.com in the Press
Time
Originally posted by: MeBeJedi I don't doubt that - but notice how the name is changed.

Besides, there's all sorts of ways to skirt laws and regulations, but just because you can buy it in a store doesn't prove it is legal. Hell, a person may unknowingly buy it because they think it is a "video signal cleaners", and never be aware that it removes the macrovision signal.

All I want is to clean up my video signal so I can watch my legally bought movies with my legal dvd player on my legal video projector without any strange interference that I have no way of knowing where it is coming from.
Post
#149844
Topic
.: The X0 Project Discussion Thread :. (* unfinished project *)
Time
Originally posted by: ripa
Originally posted by: Laserman
I thought I covered most of this before?


I'm sorry, but the search function seems to be a bit limited (only searches thread titles).

Anyway, I thought that the rotational speed difference would only yield a different frame rate but otherwise compliant PAL signal. I didn't think of the effects on the modulated signal. Since both PAL and NTSC video signals are FM modulated onto a 7,1MHz carrier, playing that back at a wrong speed would shift the carrier frequency, probably causing demodulating to fail. Also, I just noticed the technical discussion forum, where I should have posted this instead. I guess I got too excited about the prospect of playing back PAL discs on the X0. Sorry about that.

I don't think you should worry about it, finding or making a X0 equivalent PAL player has been a reoccuring topic since the beginning of the presevation efforts.

You can find detailed information of the NTSC and PAL laserdisc formats here:

The LaserDisc Technical Page

As you can see the NTSC carrier is 8.1Mhz and the PAL is 7.1Mhz, also the BW and sound carriers are different so very likely also the filters before the demodulation stage need to be changed.

Post
#149712
Topic
.: The X0 Project Discussion Thread :. (* unfinished project *)
Time
Originally posted by: Karyudo
Originally posted by: Warp99
if the 2950 or the 925 circuitry by LSI design does not provide for a 'real' composite output.


(For those following along, "LSI" is "large scale integration," which means putting a lot of discrete electronic components functions together in one package. This is typically what happens in mature electronics, and is one large reason why the prices of electronic equipment (say, DVD players) start to plummet once that equipment hits the mainstream. It's also a reason why sometimes late-model equipment isn't quite as robust or tweakable as earlier-generation stuff -- you just can't get in there and swap out individual components to try and improve or change things.)

Thanks Karyudo for filling in, yes you're right a LSI design would be difficult or impossible to alter. Only if Pioneer for some reason included a composite DAC (digital to analog converter) in the TBC (time base corrector) LSI circuit, i.e. if the same TBC circuit also was used for the players that only has composite output, it could be possible to connect a simple filter and video amplifier and get access to the composite signal before it enters the combfilter board.

Also other tweaks would be limited to the few discrete components surrounding the LSI circuits, some improvement could be made but requires some trial and error and it's timeconsuming and sometime you realize that the design was pretty optimal to begin with. The power supplies can usually be improved though, either with higher quality components or by altering the design but mistakes here can be very costly and you might end up turning the unit into a 10 kg paperweight.


Post
#149679
Topic
.: The X0 Project Discussion Thread :. (* unfinished project *)
Time
Originally posted by: Laserman You would be far better off taking an excellent PAL player like the 2950 or even the 925 and redesigning its circuitry to minimise noise than trying to turn the X0 into a PAL player.


I don't want to turn this thread into X0-PAL discussion, not at this moment anyway, but I'm curious if you did get hold of the 2950 service manual ?

IMO if the 2950 or the 925 circuitry by LSI design does not provide for a 'real' composite output, there could be other players to consider.
Post
#149638
Topic
.: Citizen's NTSC DVD / PAL DVD / XviD project :. (Released)
Time
Originally posted by: Citizen The noise you're seeing is what's on the laserdisc, the 5x capture-merge system meant I could do away with temporal softening as the picture you get is as close to what's on the laserdisc as possible.
I intentionally left the noise there for a couple of reasons; if I were to add a temporal softener filter to reduce this noise it would mean even more motion smearing and I hate that kind of smearing so I don't use temporal softeners unless I really have to, secondly to maintain an overall even quality picture feel to it because removing the noise through temporal softening would mean when there's no camera movement the picture would look extremely crisp & clean but when things started moving the noise level would increase and look worse than if the static scene already had some noise.

The dotcrawl could be because of the capture device I use, a Canopus ADVC-100 which converts the analogue video to DV so there's some compression artifacts, the 5x capture-merge and subsequent filtering helped reduce some of the compression artifacts but probably not entirely.


Yes laserdiscs are noisy and also keep in mind that the PAL standard is aprox 3 dB noisier than the NTSC standard due to the higher bandwidth and no multi-capture-averaging process can change that, it can only reduce random noise in the capture chain, which of course is still very important.

But I wonder if the noise could be reduced even further by averaging 4-5 french or 4-5 german discs of the same release through the same capture chain, intheory manufacture imperfections like noise and dropouts would then also be reduced, but maybe it would require discs that are manufactured at the same plant and almost at the same time. Anyway it was just a thought and the discs would also need to be captured 4-5 times each so we talking about 16-25 captures and who has 4-5 discs of the same ANH release..?..

Regarding the dotcrawl, I think the ADVC-100 uses a Philips saa7114 capture chip and this has a 2D 2-4 line adaptive combfilter which is quite good, better results could possibly be achieved with a 3D combfilter but those are not so common for PAL capture devices.

BTW, Citizen keep up the great work.
Post
#149632
Topic
.: The X0 Project Discussion Thread :. (* unfinished project *)
Time
Originally posted by: ripaSo what does happen when you put a PAL disc in an NTSC player? I guess this is a bit farfetched


Most (all?) NTSC only LD players would not know what to do with a PAL disc and simply reject it like it was a damagade disc.

Although the signal is stored 'composite' on the discs it still needs quite a bit processing before it becomes a signal suited for capture. And I'm pretty sure that the X0 player is filled with unique and special circuits that are all by design made for NTSC only and trying to tweak them for PAL playback would be extremly difficult, and you also need to find the X0 service manual and schematics first.

Possibly with the schematics in hand the X0 could be made to accept and sync a PAL disc but IMO you would have to process the video through other circuits and maybe even the demodulation stage needs to be different.

But hey, who would like to have their X0 player dissected?, I'll have a go at it

Post
#140246
Topic
.: The X0 Project Discussion Thread :. (* unfinished project *)
Time
Yes, zion usually announces it, but it was mentioned that it was delayed because of the mess after Katrina.

But now there's a new hurricane, I sure hope it will not be as bad this time and that everybody will be safe.

EDIT:
no, so far no newsletter #2 and I would have expected an announcement if it really was out.

can't seem to find the info in this thread...