Best VP for Laserdisc playback? - Page 4 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #91 of 301 Old 06-04-2009, 12:51 PM
Advanced Member
 
Hunter's Avatar
 
Join Date: Dec 2000
Location: Houston, TX USA
Posts: 583
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 11
Quote:


The Y/C filter I am talking about are done in the digital domain. There is no perfect "comb filter".

I know. And the filter in the X9, for instance, is digital, also. And it is the fact of imperfection that makes the trade offs of the 3D adaptive process less acceptable than in the non-adaptive process, when fed into through any digital processing. No matter what the quality of the 3D adaptive circuitry.

Quote:


I can't support their tactics. I still have not seen any MEASURED technical specifications of the Entech decoders. Perhaps someone can point me to some.

And that's my ENTIRE point!!!
You haven't seen any specs nor have you seen the actual performance. Yet based upon a almost complete absence of information, you are concluding that the unit is flawed, and uncompetitive.

Don't get me wrong - I am NO supporter of Monster, from their inception in the early 1980's through today. Noel is a fantastic marketer, and has created a hugely successful business. It ends there for me - the products are "mid fi" at best, for the most part, and I have never purchased any at retail (the Entech and a few adapters I've purchased have been from the secondary market).

But to coin a phrase ... I am sorry if it offends some here but as an attorney with 30 years in the legal profession and 40 years in the audio and video areas, rendering judgments on no evidence is not advisable IMO. That's all.

And that CVSC-1? It's not supposed to compensate for insertion loss, resolution loss, color shift, overshoot, phase shift, or any other degradation caused by its use. It doesn't claim to, beyond the typical consumer puffery. It provides the option for people to get a quick, cheap S-Video signal from Composite. It does that!

And I can't let this go ...
Quote:


It (the CLD-95 et seq) doesn't get any cleaner than this.

Yes is does. No digital conversion. The X0.

Peace. Out.

Regards,

- Hunter
Hunter is offline  
Sponsored Links
Advertisement
 
post #92 of 301 Old 06-04-2009, 01:00 PM
AVS Special Member
 
Glimmie's Avatar
 
Join Date: Sep 2000
Location: Los Angeles, CA
Posts: 8,017
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 182 Post(s)
Liked: 251
Quote:
Originally Posted by HDgaming42 View Post

The player was a Pioneer LD-V4400 fed house sync. One decoder was a leitch, the others I don't remember. Composite out to a broadcast monitor did not show the same artifacting (very digital in nature) that any route into the digital domain produced.

I don't think anyone here is defending Monster, but the Entech deserves a closer look. Just ignore the Monster marketing speak that is based in anything but reality.

I think we'd all love to see some proper measurements for any of the methods discussed here. It would make the whole discussion a lot clearer.

So far we have a number of screencaps of different comb filters operating on the LD Video Essentials. I'd love to see some of your results! The Zone Plate would be of particular interest...

Well I'm not surprised the Leitch decoder was mediocre. They never made (or claimed to make) a reference decoder. Their products in this area were "utility" grade designed for picture only monitoring and news feed type material.

I know nothing about the Entech, in fact never heard of it until this thread. But looking at it and based on the profile of the company I can pretty much gather it is just a reference design of some OEM decoder chip. The lack of published specs is also suspicious. Do they even have the test equipment to measure the parameters?

I don't have a VE laser disk. I wish I did. I can make screen shots but I am not a beleiver in them. There are so many variables the end result here on the forum is practically useless. A zone plate however would be of some value here.

Glimmie's HT Page
Being redone - comming soon!

Glimmie is online now  
post #93 of 301 Old 06-04-2009, 01:13 PM
AVS Special Member
 
Glimmie's Avatar
 
Join Date: Sep 2000
Location: Los Angeles, CA
Posts: 8,017
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 182 Post(s)
Liked: 251
Quote:
Originally Posted by Hunter View Post

I know. And the filter in the X9, for instance, is digital, also. And it is the fact of imperfection that makes the trade offs of the 3D adaptive process less acceptable than in the non-adaptive process, when fed into through any digital processing. No matter what the quality of the 3D adaptive circuitry.


And that's my ENTIRE point!!!
You haven't seen any specs nor have you seen the actual performance. Yet based upon a almost complete absence of information, you are concluding that the unit is flawed, and uncompetitive.

Don't get me wrong - I am NO supporter of Monster, from their inception in the early 1980's through today. Noel is a fantastic marketer, and has created a hugely successful business. It ends there for me - the products are "mid fi" at best, for the most part, and I have never purchased any at retail (the Entech and a few adapters I've purchased have been from the secondary market).

But to coin a phrase ... I am sorry if it offends some here but as an attorney with 30 years in the legal profession and 40 years in the audio and video areas, rendering judgments on no evidence is not advisable IMO. That's all.

And that CVSC-1? It's not supposed to compensate for insertion loss, resolution loss, color shift, overshoot, phase shift, or any other degradation caused by its use. It doesn't claim to, beyond the typical consumer puffery. It provides the option for people to get a quick, cheap S-Video signal from Composite. It does that!

And I can't let this go ...

Yes is does. No digital conversion. The X0.

Peace. Out.


How can you have "no digital processing" today? Unless you are feeding your X9 into a CRT based display device, there is going to be digital processing. Does the X9 have a TBC? Then there is digital processing. Or do you perfer the Y/C smear from hetrodyne color stabilization? Now I do agree if that is what you are doing, then my processing chain most assured has more artifacts. However I also have no visible scan lines and far less noise due to the use of a high end image processor. These sold for $70K in the late 1990s and the only reason I have one is the SD models could not be upgraded to HD so most were scrapped. This is not even close to the crappy noise reducers in the current crop of videophile scalers.

As you are an attorney I'm sure you understand circumstantial evidence as well. Tests aside, do you really beleive some < $1000 consumer product made by a tiny esoteric audio company is superior to a $30,000+ product made by a comapny that is recoginized as a leading broadcast image processing manufacture? Yes I am aware of Moore's law but it doesn't close gaps this huge.

The CVS-C1 claims to remove "hanging dots". It's in the sales brochure. Now read the section on "hanging dots" by Virgle Lowe in that PDF I attached. Now do you think that cute little passive box is doing what he outlines as the problem? I'll bet it still does remove the hanging dots though. They simply roll off the luminance at 2mhz! This is why I don't subscribe to many of the consumer video equipment claims. They take a highly technical concept and just throw it into print.

Glimmie's HT Page
Being redone - comming soon!

Glimmie is online now  
post #94 of 301 Old 06-04-2009, 03:18 PM
Advanced Member
 
Hunter's Avatar
 
Join Date: Dec 2000
Location: Houston, TX USA
Posts: 583
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 11
Quote:


it is just a reference design of some OEM decoder chip

This is just plain incorrect. At a minimum please just open one up and look at it for 15 seconds before you keep making these totally unfounded claims.
Doesn't that seem fair?

Quote:


I'm sure you understand circumstantial evidence as well

Yes, and you have made it inescapably clear in at least two instances that you do not even have that with regard to the Entech under discussion for the past several years.

As I said I have almost no technical respect for Monster. I do not care, in this arena, about any aspect of the performance of that $119.00 cable-interrupter of a device. Why do you keep harping on that? It is virtually irrelevant to everything people have been otherwise discussing responsibly in this thread.

Quote:


there is going to be digital processing

THAT'S THE WHOLE POINT!!!!!!!!!!!
One more time - it is the combination of the 3D adaptive process PLUS digital processing that causes the problems we have all been writing around. And some of us have even seen.

Regards,

- Hunter
Hunter is offline  
post #95 of 301 Old 06-04-2009, 03:40 PM
AVS Special Member
 
Glimmie's Avatar
 
Join Date: Sep 2000
Location: Los Angeles, CA
Posts: 8,017
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 182 Post(s)
Liked: 251
Quote:
Originally Posted by Hunter View Post

This is just plain incorrect. At a minimum please just open one up and look at it for 15 seconds before you keep making these totally unfounded claims.
Doesn't that seem fair?
.......
THAT'S THE WHOLE POINT!!!!!!!!!!!
One more time - it is the combination of the 3D adaptive process PLUS digital processing that causes the problems we have all been writing around. And some of us have even seen.

Then show me technical specifications or at least a technical review of the product. I have searched and cannot find one.

I an a graduate EE as well as 25 years+ experience in broadcast engineering - mostly in high end post production. You are trying to tell me the Entech is superior to the high end professional decoders. If you bothered to read the technical papaer I povided you would see that 3D adaptive is not universally used in high end units. It is switched in where it will be most effective based on realtime analysis. There is no way the Entech is doing that for it's price point. Your 2D Entech has no knowledge of temporal video information. Only a field and frame based comb filter can compensate for that. You don't get that level of performance in consumer NTSC decoders.

I do image processing as a profession as well as a hobby. I have been involved in restoration of many old 2inch tape projects with the latest digital tools. I know how to get the best image from composite NTSC to component.

Glimmie's HT Page
Being redone - comming soon!

Glimmie is online now  
post #96 of 301 Old 06-04-2009, 04:01 PM
AVS Special Member
 
Glimmie's Avatar
 
Join Date: Sep 2000
Location: Los Angeles, CA
Posts: 8,017
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 182 Post(s)
Liked: 251
Let's try to cut the speculation.

Can somebody with an Entech post a photo if the interior. Try to get the whole interior to fill the screen. Based on what I see, I can tell you exactly what they are doing.

I have built analog encoders (component to NTSC) from the ground up in the mid 1980s. Yes, I can pretty much draw a block diagram form a picture if it's an analog design. With digital processing units, not a chance!

Glimmie's HT Page
Being redone - comming soon!

Glimmie is online now  
post #97 of 301 Old 06-04-2009, 04:35 PM
AVS Special Member
 
Glimmie's Avatar
 
Join Date: Sep 2000
Location: Los Angeles, CA
Posts: 8,017
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 182 Post(s)
Liked: 251
In reading through the posts here and the archived threads on Laserdisk I see there is misunderstanding of how the format works and why a "pure" unprocessed composite output is impossible. One of two things must be done - either a digital TBC that processes the entire composite signal OR a hetrodyne color processor. Here is why:

Laser disk uses "Direct NTSC Recording". That means the entire composite signal is FM modulated and recorded on the disk as FM. The frequencies used are the same as PAL composite analog videotape - roughly 6-9mhz. That locks in a S/N of 46db maximum.

On playback the FM is equalized, limited, and demodulated just as in a broadcast 1 or 2 inch analog VTR. However just like a VTR's rotating heads are mecanically unstable, the large spinning disk is far worse. The jitter far exceeds tolerances for NTSC color playback. So at least the color portion of the signal must be time stabalized. To do this you must seperate the chroma, by some form of filter. Do some analog frequency beating to remove the jitter, and then mix the chroma back to the luminance that has been delayed to match the chromance. Here just like in consumer tape, the chorma is stable but the luminance is not. This is Ok for a monitor but since the locked relationship betwnn the color subcarrier and lumimance sync is lost, you get this Y/C smearing of the edges. One is stable, the other not.

FWEIW, VHS and SVHS is even worse as it can't record FM frequencies more than about 5mhz do to the slow head to tape speed. So they must record the color seperatly to start with. This really causes the ugly VHS artifcats we all know and love, Why can't we record 4.2mhz NTSC with a 5mhz FM signal - all I can say is study FM modulation theory. It's too complex to get into here.

A TBC corrects BOTH the chromanace and luminance together. They are never seperated. But the analog must be digitized to go through a digital TBC. There were analog TBC at one time for 2 inch tape but they are obsolete and never could cope with the high jitter of a laserdisk player.

So anybody thinking they are getting as pure composite analog output from any laserdisk player is mistaken. You either suffer a very early Y/C seperation for the non TBC units or and A/D then D/A conversion for the TBC equipped units.

My idea was to grab the time base corrected video while still in digital format then feed it to a digital input decoder. I suffer one A/D conversion. The X9 feeding a decoder through composite video is going through and A/D and D/A as well as other possible processing so it can't be as clean a signal as a digital tap.

Glimmie's HT Page
Being redone - comming soon!

Glimmie is online now  
post #98 of 301 Old 06-04-2009, 04:36 PM
Advanced Member
 
nidi's Avatar
 
Join Date: Oct 2004
Location: Switzerland
Posts: 567
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by Glimmie View Post

Let's try to cut the speculation.

Can somebody with an Entech post a photo if the interior. Try to get the whole interior to fill the screen. Based on what I see, I can tell you exactly what they are doing.

I have built analog encoders (component to NTSC) from the ground up in the mid 1980s. Yes, I can pretty much draw a block diagram form a picture if it's an analog design. With digital processing units, not a chance!

I can do that, of both units tomorrow.

as I said before, the Entech is mediocre at best.

HDGaming, what are your findings of the test footage you saw from me ?

Michael

Hunter, I would like to see some test footage with your Entech.

THX Movies sometimes had test patterns on them.
nidi is offline  
post #99 of 301 Old 06-04-2009, 04:38 PM
Advanced Member
 
nidi's Avatar
 
Join Date: Oct 2004
Location: Switzerland
Posts: 567
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by Glimmie View Post

In reading through the posts here and the archived threads on Laserdisk I see there is misunderstanding of how the format works and why a "pure" unprocessed composite output is impossible. One of two things must be done - either a digital TBC that processes the entire composite signal OR a hetrodyne color processor. Here is why:

Laser disk uses "Direct NTSC Recording". That means the entire composite signal is FM modulated and recorded on the disk as FM. The frequencies used are the same as PAL composite analog videotape - roughly 6-9mhz. That locks in a S/N of 46db maximum.

On playback the FM is equalized, limited, and demodulated just as in a broadcast 1 or 2 inch analog VTR. However just like a VTR's rotating heads are mecanically unstable, the large spinning disk is far worse. The jitter far exceeds tolerances for NTSC colorplayback. So at least the color portion of the signal must be time stabalized. To do this you must seperate the chroma, by some form of filter. Do some analog frequency beating to remove the jitter, and then mix the chroma back to the luminance that has been delayed to match the chromance. Here just like in consumer tape, the chorma is stable but the luminance is not. This is Ok for a monitor but since the locked relationship betwnn the color subcarrier and lumimance sync is lost, you get this Y/C smearing of the edges. One is stable, the other not.

A TBC corrects BOTH the chromanace and luminance together. They are never seperated. But the analog must be digitized to go through a digital TBC. There were analog TBC at one time for 2 inch tape but they are obsolete and never could cope with the high jitter of a laserdisk player.

So anybody thinking they are getting as pure composite analog output from any laserdisk player is mistaken. You either suffer a very early Y/C seperation for the non TBC units or and A/D then D/A conversion for the TBC equipped units.

My idea was to grab the time base corrected video while still in digital format then feed it to a digital input decoder. I suffer one A/D conversion. The X9 feeding a decoder through composite video is going through and A/D and D/A as well as other possible processing so it can't be as clean a signal as a digital tap.



I have a service manual of the X0, try to post a scematic tomorrow.

and it definately isn't analog all the way.

Michael
nidi is offline  
post #100 of 301 Old 06-04-2009, 04:50 PM
AVS Special Member
 
Fudoh's Avatar
 
Join Date: Jun 2001
Location: Munich, Germany
Posts: 1,345
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 21
Here are a few shots from the CVSI-1's inside:






@nidi: maybe the SVSI-1 would be interesting for comparison. And I'm still wondering about the PSUs. Can you by any chance take a digicam photo of the original SVSI and CVSI power supplies ? You emailed me the specs (16V DC output), but there have been bundled PSUs with 12V as well (thanks to sidb for the the picture) and mine has been running on 9V lately.
Fudoh is offline  
post #101 of 301 Old 06-04-2009, 07:37 PM
Advanced Member
 
HDgaming42's Avatar
 
Join Date: Sep 2006
Location: Alberta, Canada
Posts: 633
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 130 Post(s)
Liked: 18
Noooow we're getting somewhere! Thanks for the pics Fudoh!

Below are a few crops of Zone Plates blown up to 200% featuring different comb filters. Thanks to Nidi for the caps! My only ? is--did you use two different sources? The fonts don't match and there is a luminance difference...

Sorry for them not being inline--I don't have webspace nor the patience to sign up for a free image hosting service. Feel free to do so if you wish!

I've ordered them in (IMHO) descending quality.
LL
LL
LL
LL
LL
HDgaming42 is offline  
post #102 of 301 Old 06-04-2009, 07:49 PM
AVS Special Member
 
Glimmie's Avatar
 
Join Date: Sep 2000
Location: Los Angeles, CA
Posts: 8,017
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 182 Post(s)
Liked: 251
Ok, this is a DIGITAL processor. The ADV7123 chip is a triple channel DAC which is most likely providing RGB or YcBcR output. The ADS822E is a flash converter. What I find curious is that there are three of them. The only thing that makes sense to me is one for composite and two for Svideo. I would have shared two of the between Svideo and composite but whatever.

Those Xilinx FPGAs are rather low density by todays standards but big enough to implement a 2D/3D design. There is not enough memeory visible there to hold a field or frame so this rules out temporal combing.

The resistors in the analog section are all 1% so they didn't cut corners there.

I must say it is better built than I expected.

OK, Here's what I don't see. INDUCTORS! Or at least not enough. There looks to be one possibly per ADC but that could also be a power rail choke. There seems to be little or no anti-aliasing filters before the A/D. Likewise there seems to be lacking reconstruction filters on the DACS. Do the guy's that designed this understand Nyquist? You can pull this stunt with oversampling CD players. It won't work very well with video. Any HF noise in the input signal over the Nyquist limit is going to fold back and create aliasing.

Glimmie's HT Page
Being redone - comming soon!

Glimmie is online now  
post #103 of 301 Old 06-05-2009, 05:28 AM
Advanced Member
 
Hunter's Avatar
 
Join Date: Dec 2000
Location: Houston, TX USA
Posts: 583
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 11
Yes, Glimmie, we now ARE somewhere!!
Overall ... good for undertaking a view.

Per the "it has to be digital" and TBC postings ... there is no digital TBC in the BNC Composite output of the HLX-X0. Either that or the research (or memory) of several people + me on AVS several years ago is wrong. We studied the schematics.
I would truly like to provide you with those. You have NO idea how much I would like to as it would mean that I still have my X0 Service manual. Somehow it has been misplaced over the years.

The 3D adaptive artifacts that NIN74, tkmedia, Josh, myself and others finally found too much to put up with are with any digital processing that occurs after the player's output. Meaning after the filter.

As to the inductors issue ... the unit may not be perfect, according to the most rigorous EE theoretical standards, but it's head and shoulders above most in terms of design, including industry units. (I've seen and heard pro stuff that certainly will survive misuse in any condition, but the results are mediocre.) But I'd be quite surprised, for example, if there is NO anti-aliasing filter of some sort. If nothing else the output just doesn't have the friggin' cr*p that would result.

PLEASE view the results of the Entech before you tell anyone that it just can't ultimately be useful. Once you do - after the best possible Composite signal you can find and before any deinterlacing, scaling, etc. - and compare it to other alternatives, then please tell us of your conclusion. I would welcome that. What I can tell you is that with a CLD-95, CLD-97, CLD-99, CLD-704, CLD-703, LD-S2, HLD-X9, HLD-X0, and others, together with close to two dozen Y/C filters and way over a dozen scalers/processors.

Quote:


Do the guy's that designed this understand Nyquist?

... sigh ... yeah I think so. I would find it almost impossible to believe you haven't heard of the designers/engineers/theoretical personnel I named earlier: Martin, Fryer, Johnson, and the Spectral staff. You may not agree with some of the details of their choices (there are always tradeoffs, right?), but their respective knowledge, experience and results are unassailable in terms of technical knowledge and application.
For PCM see the SDR-1000, et seq series. From the first effort in the early 1990's the digital products have been excellent at worst. Reference - I am still kicking and wriggling about the sampling/filtering and digital signal/clock interface failings of CD, since having come unglued during one of the first presentations of Sony's horrid-sounding CDP-101 in Washington, D.C., in 1983. In spite of my not having been a supporter of many things Spectral, I have always thought their digital products to be among the finest. The people and company DO understand the issues.

Regards,

- Hunter
Hunter is offline  
post #104 of 301 Old 06-05-2009, 11:10 AM
AVS Addicted Member
 
Josh Z's Avatar
 
Join Date: Jul 2002
Location: Planet Boston, source of the spice, Melange.
Posts: 20,292
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 461 Post(s)
Liked: 411
Quote:
Originally Posted by Glimmie View Post

On playback the FM is equalized, limited, and demodulated just as in a broadcast 1 or 2 inch analog VTR. However just like a VTR's rotating heads are mecanically unstable, the large spinning disk is far worse. The jitter far exceeds tolerances for NTSC color playback. So at least the color portion of the signal must be time stabalized. To do this you must seperate the chroma, by some form of filter. Do some analog frequency beating to remove the jitter, and then mix the chroma back to the luminance that has been delayed to match the chromance. Here just like in consumer tape, the chorma is stable but the luminance is not. This is Ok for a monitor but since the locked relationship betwnn the color subcarrier and lumimance sync is lost, you get this Y/C smearing of the edges. One is stable, the other not.

Glimmie, I would be really interested in your insight into the so-called (for lack of a better name) "white smearing" artifact that affects almost every laserdisc player ever produced, except the HLD-X0 and HLD-X9.

The problem manifests only on CLV format discs, not CAV.

I've never heard a good explanation for it. Nor has anyone ever found a fix (aside from buying an X9 or X0).

Josh Z
Writer/Editor, High-Def Digest (Blog updated daily!)
Curator, Laserdisc Forever

My opinions are my own, and do not necessarily reflect those of my employers.

Josh Z is online now  
post #105 of 301 Old 06-05-2009, 11:35 AM
Advanced Member
 
nidi's Avatar
 
Join Date: Oct 2004
Location: Switzerland
Posts: 567
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by Hunter View Post

Yes, Glimmie, we now ARE somewhere!!
Overall ... good for undertaking a view.

Per the "it has to be digital" and TBC postings ... there is no digital TBC in the BNC Composite output of the HLX-X0. Either that or the research (or memory) of several people + me on AVS several years ago is wrong. We studied the schematics.

signal path of the X0
LL
nidi is offline  
post #106 of 301 Old 06-05-2009, 11:55 AM
AVS Special Member
 
Glimmie's Avatar
 
Join Date: Sep 2000
Location: Los Angeles, CA
Posts: 8,017
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 182 Post(s)
Liked: 251
Quote:
Originally Posted by nidi View Post

signal path of the X0

Aside fromthe Kanji characters I can read "A/D" and "TBC". So this particular unit has a DIGITAL TBC. It also shows that the composite output remains in a composite form and is not Y/C seperated and recombined as in some other Pioneer units. There is a "VNR" block however and this can be trouble. If it can't be bypassed via s switch or menu item, it WILL induce some artifacts.

Glimmie's HT Page
Being redone - comming soon!

Glimmie is online now  
post #107 of 301 Old 06-05-2009, 12:13 PM
AVS Special Member
 
Glimmie's Avatar
 
Join Date: Sep 2000
Location: Los Angeles, CA
Posts: 8,017
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 182 Post(s)
Liked: 251
Quote:
Originally Posted by Hunter View Post

Yes, Glimmie, we now ARE somewhere!!
Overall ... good for undertaking a view.

Per the "it has to be digital" and TBC postings ... there is no digital TBC in the BNC Composite output of the HLX-X0. Either that or the research (or memory) of several people + me on AVS several years ago is wrong. We studied the schematics.
I would truly like to provide you with those. You have NO idea how much I would like to as it would mean that I still have my X0 Service manual. Somehow it has been misplaced over the years.

As I explained above, it either has a digital TBC or a Hetrodyne type color corrector. There has to be something there to correct for the mechanical jitter of the disk and optical head. Sure there's a servo system but that's not good enough for color playback. And if it has a hetrodyne color corrector, there has to be Y/C seperation at the front end. Furthermore a hetrodyne color correction system is inferior to a TBC system. As I said, hetrodyne color correctors are what VHS and Beta machines used.

Also note the posted block diagram, the X0 clearly has a digital TBC.

Quote:


The 3D adaptive artifacts that NIN74, tkmedia, Josh, myself and others finally found too much to put up with are with any digital processing that occurs after the player's output. Meaning after the filter.

Again, with a high end decoder you don't have this issue. At least I don't using an Innovision/Tekniche DX210. The Accom DB221 is just as good.
Quote:


As to the inductors issue ... the unit may not be perfect, according to the most rigorous EE theoretical standards, but it's head and shoulders above most in terms of design, including industry units. (I've seen and heard pro stuff that certainly will survive misuse in any condition, but the results are mediocre.) But I'd be quite surprised, for example, if there is NO anti-aliasing filter of some sort. If nothing else the output just doesn't have the friggin' cr*p that would result.

PLEASE view the results of the Entech before you tell anyone that it just can't ultimately be useful. Once you do - after the best possible Composite signal you can find and before any deinterlacing, scaling, etc. - and compare it to other alternatives, then please tell us of your conclusion. I would welcome that. What I can tell you is that with a CLD-95, CLD-97, CLD-99, CLD-704, CLD-703, LD-S2, HLD-X9, HLD-X0, and others, together with close to two dozen Y/C filters and way over a dozen scalers/processors.

Just as with the TBC/color corrector requirement, there are other laws of physics and electrical engineering that cannot be bypassed without penalty. I don't see a component nest on the board that looks to be a filter anywhere. I can't be sure without a schematic. The fact that you don't see any alaising could be due to other factors. I could show you countless schematics of broadcast equipment with A/D and D/A converters and the associated filters. Many of these filters are at least 5 poles. If the Entech does have filters, there are not more than a single pole. There just isn't any parts there to support anything beyond that. Intesteringly those filters are also the root of other evils. Ringing is high up on the list. Non linear gain and phase shift is another biggie. Differential gain and phase fidelity is very important in NTSC. But simply leaving them out is just as bad. This is why I went to an all digital path as early as possible. right after the TBC in the CLD95. I then only have one analog filter in front of the TBC to induce aritfacts.

Quote:


... sigh ... yeah I think so. :rolleyes I would find it almost impossible to believe you haven't heard of the designers/engineers/theoretical personnel I named earlier: Martin, Fryer, Johnson, and the Spectral staff. You may not agree with some of the details of their choices (there are always tradeoffs, right?), but their respective knowledge, experience and results are unassailable in terms of technical knowledge and application.
For PCM see the SDR-1000, et seq series. From the first effort in the early 1990's the digital products have been excellent at worst. Reference - I am still kicking and wriggling about the sampling/filtering and digital signal/clock interface failings of CD, since having come unglued during one of the first presentations of Sony's horrid-sounding CDP-101 in Washington, D.C., in 1983. In spite of my not having been a supporter of many things Spectral, I have always thought their digital products to be among the finest. The people and company DO understand the issues.

No I haven't heard of them. Probably becaused I never subscribed to "Stereo Review" or "The Absolute Sound" I am a member of SMPTE, SBE, and IEEE and can get access to AES documents as well. Perhaps you could direct me to some of their papers? I mean these great guys must have published some PROFESSIONAL papers regarding their work?

Glimmie's HT Page
Being redone - comming soon!

Glimmie is online now  
post #108 of 301 Old 06-05-2009, 12:29 PM
AVS Special Member
 
Glimmie's Avatar
 
Join Date: Sep 2000
Location: Los Angeles, CA
Posts: 8,017
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 182 Post(s)
Liked: 251
Quote:
Originally Posted by Josh Z View Post

Glimmie, I would be really interested in your insight into the so-called (for lack of a better name) "white smearing" artifact that affects almost every laserdisc player ever produced, except the HLD-X0 and HLD-X9.

The problem manifests only on CLV format discs, not CAV.

I've never heard a good explanation for it. Nor has anyone ever found a fix (aside from buying an X9 or X0).

I don't know exactly what you are refering too but it sounds like an FM modulation issue.

Do you have a screen shot?

Glimmie's HT Page
Being redone - comming soon!

Glimmie is online now  
post #109 of 301 Old 06-05-2009, 12:37 PM
Advanced Member
 
nidi's Avatar
 
Join Date: Oct 2004
Location: Switzerland
Posts: 567
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
I got the 2 discs today and did a capture of the mentioned examples

of Austin Powers' Phone Booth scene

and 5th Element's LeeLoo assemble scene.

PM me for the download links


Michael
nidi is offline  
post #110 of 301 Old 06-05-2009, 01:31 PM
AVS Addicted Member
 
Josh Z's Avatar
 
Join Date: Jul 2002
Location: Planet Boston, source of the spice, Melange.
Posts: 20,292
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 461 Post(s)
Liked: 411
Quote:
Originally Posted by Glimmie View Post

I don't know exactly what you are refering too but it sounds like an FM modulation issue.

It's the issue, on CLV discs, where a faint "ghost" image of an object will seem to smear to the right of that object. The most famous example of this is at the beginning of Goldeneye. When agent 006 points his gun at 007, the barrell of the gun very clearly smears to the right.

The severity of the artifact varies from scene to scene. Some scenes may not have it at all, and on some it's very minor. But it affects every CLV disc to some extent. Meanwhile, the CAV edition of the same movie and transfer will be free of the problem.

This is present on every laserdisc player I've ever owned (and I've owned a lot) except the HLD-X9. I haven't owned an X0, but hear it is also free of the problem. Generally, players with sharper pictures (like the CLD-D704) will show it off more clearly than those with softer picture, where it might be masked (but is still present).

It's visible on any type of display. It's visible with any and all of the LD players' NR and other processing features turned off. I've eliminated every other variable, and am certain that it's not something else in my signal chain. Also, numerous other viewers have confirmed seeing it.

It's the type of thing that once you see it, you can't not ever see it again.

Quote:
Do you have a screen shot?

Not at the moment.

Here's an old thread about the subject at HTF:

http://www.hometheaterforum.com/htf/...isc-smear.html

The post by "ChristopherDAC" is the closest I've ever heard to a plausible explanation, but I don't have the engineering background to know how accurate that theory may be.

Josh Z
Writer/Editor, High-Def Digest (Blog updated daily!)
Curator, Laserdisc Forever

My opinions are my own, and do not necessarily reflect those of my employers.

Josh Z is online now  
post #111 of 301 Old 06-05-2009, 01:44 PM
AVS Special Member
 
Glimmie's Avatar
 
Join Date: Sep 2000
Location: Los Angeles, CA
Posts: 8,017
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 182 Post(s)
Liked: 251
Quote:
Originally Posted by Josh Z View Post

It's the issue, on CLV discs, where a faint "ghost" image of an object will seem to smear to the right of that object. The most famous example of this is at the beginning of Goldeneye. When agent 006 points his gun at 007, the barrell of the gun very clearly smears to the right.

The severity of the artifact varies from scene to scene. Some scenes may not have it at all, and on some it's very minor. But it affects every CLV disc to some extent. Meanwhile, the CAV edition of the same movie and transfer will be free of the problem.

This is present on every laserdisc player I've ever owned (and I've owned a lot) except the HLD-X9. I haven't owned an X0, but hear it is also free of the problem. Generally, players with sharper pictures (like the CLD-D704) will show it off more clearly than those with softer picture, where it might be masked (but is still present).

It's visible on any type of display. It's visible with any and all of the LD players' NR and other processing features turned off. I've eliminated every other variable, and am certain that it's not something else in my signal chain. Also, numerous other viewers have confirmed seeing it.

It's the type of thing that once you see it, you can't not ever see it again.



Not at the moment.

Here's an old thread about the subject at HTF:

http://www.hometheaterforum.com/htf/...isc-smear.html

The post by "ChristopherDAC" is the closest I've ever heard to a plausible explanation, but I don't have the engineering background to know how accurate that theory may be.

Sounds like ringing or non-linear delay in a filter. Any filter causes a time delay. Some frequencies are delayed more than others so at the output these time differewnces are displayed as the ghost images you see.

Filters are not limited to the A/D and D/A being discussed above either. There are a few filters in the RF (FM) section of the playback chain. These too can cause the problem. As I noted above the RF signal chain in laserdisk is the same as the old broadcast 1 and 2 inch VTRs. Just like a tape head, the laser diode and detector requires some equalization. This is where non linear phase shift is the worst - in a video signal carried as FM. The broadcast tape machines used a "transversal" filter for this stage. These do not induce phase shift but are complex circuits. I doubt any of the Pioneer CLV series had them. In the X series, they may have gone there but I don't know. A simple series type equalizer here will cause frequency dependant phase shifts which when demodulated can cause "ghosts".

Glimmie's HT Page
Being redone - comming soon!

Glimmie is online now  
post #112 of 301 Old 06-05-2009, 02:51 PM
AVS Addicted Member
 
Josh Z's Avatar
 
Join Date: Jul 2002
Location: Planet Boston, source of the spice, Melange.
Posts: 20,292
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 461 Post(s)
Liked: 411
Have you seen this artifact on your own player? Can anything be done to eliminate it?

Josh Z
Writer/Editor, High-Def Digest (Blog updated daily!)
Curator, Laserdisc Forever

My opinions are my own, and do not necessarily reflect those of my employers.

Josh Z is online now  
post #113 of 301 Old 06-05-2009, 02:57 PM
AVS Special Member
 
Fudoh's Avatar
 
Join Date: Jun 2001
Location: Munich, Germany
Posts: 1,345
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 21
Just downloaded nidi's sample files. Stunning overall picture quality from a LD. The composite output (captured using a Sweetspot card and therefore the Sweetspot's Y/C filter) exhibits several rainbow artefacts all over the place. You can see them at the cast lettering on Austin powers or MM's shirt in the very first scene and nearly everywhere on the construction robot in the 5th Element scene.

In THIS comparison the S-Video output of the X0 looks superior by all means. Now the question remains if there's any external device which will boost the X0's composite output beyond the captured S-Video quality sample...
Fudoh is offline  
post #114 of 301 Old 06-05-2009, 03:06 PM
AVS Special Member
 
Glimmie's Avatar
 
Join Date: Sep 2000
Location: Los Angeles, CA
Posts: 8,017
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 182 Post(s)
Liked: 251
Quote:
Originally Posted by Josh Z View Post

Have you seen this artifact on your own player? Can anything be done to eliminate it?


I'll check mine. I failed to mention the artifact could be recorded in during the disk mastering as well. Not likely as Pioneer had good QC but still a possibility.

FWEIW My laser disk collection is mainly limited to concerts. Reason being they are forever locked into NTSC and a DVD version is worse based on how much care was taken in the transfer. Furthermore Laserdisk will never have MPEG artifacts.

Movies OTOH, can be remastered to DVD from component video tape. So there is no NTSC involved and I find the lack of NTSC artifacts a favorable compromise to the MPEG artifacts. Still don't think every feature on DVD is from a component source. Mastering was done to analog 1 inch tape up until the early 1990s when component D1 took over. So an older vidoemaster in the studio vault could still be analog NTSC. If it's not a big title the studio will just cut the DVD from the NTSC tape versus the $80K average cost of a new component transfer from film.

Glimmie's HT Page
Being redone - comming soon!

Glimmie is online now  
post #115 of 301 Old 06-05-2009, 05:40 PM
AVS Special Member
 
dr1394's Avatar
 
Join Date: May 2002
Location: Mizar 5
Posts: 3,172
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 21 Post(s)
Liked: 21
This is just FYI for the folks on this thread. The best comb filter I've seen is from this company:

http://www.videoq.com/

http://www.videoq.com/Downloads/VideoQ%20VQP.pdf

Unfortunately, it's only a development board. Videoq is Victor Steinberg's (formerly of Snell and Wilcox) company.

Ron

HD MPEG-2 Test Patterns http://www.w6rz.net
dr1394 is online now  
post #116 of 301 Old 06-06-2009, 07:22 AM
AVS Addicted Member
 
Josh Z's Avatar
 
Join Date: Jul 2002
Location: Planet Boston, source of the spice, Melange.
Posts: 20,292
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 461 Post(s)
Liked: 411
Quote:
Originally Posted by Glimmie View Post

I'll check mine. I failed to mention the artifact could be recorded in during the disk mastering as well.

If that were true, the problem wouldn't vanish when the disc is played on an X0 or X9. This is definitely a hardware problem, not a software problem.

Josh Z
Writer/Editor, High-Def Digest (Blog updated daily!)
Curator, Laserdisc Forever

My opinions are my own, and do not necessarily reflect those of my employers.

Josh Z is online now  
post #117 of 301 Old 06-06-2009, 09:34 AM
Advanced Member
 
HDgaming42's Avatar
 
Join Date: Sep 2006
Location: Alberta, Canada
Posts: 633
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 130 Post(s)
Liked: 18
Quote:
Originally Posted by dr1394 View Post

This is just FYI for the folks on this thread. The best comb filter I've seen is from this company:

http://www.videoq.com/

http://www.videoq.com/Downloads/VideoQ%20VQP.pdf

Unfortunately, it's only a development board. Videoq is Victor Steinberg's (formerly of Snell and Wilcox) company.

Ron

thanks for this! I noticed that the 2D portion of the comb filter uses the SAA chipset.

Glimmie, or anyone else with an opinion on the matter: which is generally regarded as the superior chipset in this regard? The SAA's or the TI's? I was trying to gather this information in another thread but didn't have a lot of takers...

For instance:

- Holo3DGraph II using SAA7118
- Crystalio II using TI-TVP5160

Now I realize there will always be differences in implementation, but soley as a chipset, does one perform better than the other?
HDgaming42 is offline  
post #118 of 301 Old 06-06-2009, 03:08 PM
Advanced Member
 
nidi's Avatar
 
Join Date: Oct 2004
Location: Switzerland
Posts: 567
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
the Saa is a 2D design and the TI is a 2D/3D design.

I can't capture the output of the C2, but as I said previously, the TI
implementation in the C2 is really good, it's really close to the
X0's comb filter

Michael
nidi is offline  
post #119 of 301 Old 06-07-2009, 02:29 PM
Advanced Member
 
nidi's Avatar
 
Join Date: Oct 2004
Location: Switzerland
Posts: 567
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Josh , HDGaming, Fudoh, have you compared the test clips
I did yet?

what is your verdict ?


Michael
nidi is offline  
post #120 of 301 Old 06-07-2009, 04:04 PM
AVS Special Member
 
Fudoh's Avatar
 
Join Date: Jun 2001
Location: Munich, Germany
Posts: 1,345
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 6 Post(s)
Liked: 21
I sure did, I downloaded the additional SVSI-1 sample as well and while the SCVI-1 might have an advantage when it comes to actual color resolution and to feeding the output to an outboard processor for additional postprocessing (especially sharpening), the X0's direct S-video output actually wins. It has much less rainbow artefacts and vivid, but still solid colors. Also dotcrawl is minimized using the X0's internal comb filter.

So, in the end, the question remains if there's any outboard comb filter which will exceed the X0's s-video quality. I'll try to do the same or similar tests on my X9 soon. Hell, some people mentioned that the X9 had the edge over the X0 when it comes to S-Video output...

Quick comparison:

Fudoh is offline  
Reply Video Processors

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off