It is currently 18 Aug 2019, 06:35




 Page 1 of 2 [ 28 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: Y/C separation happening in players / how to recombine.
PostPosted: 15 Jan 2019, 12:58 
True fan
True fan
User avatar

Joined: 11 Oct 2015, 16:35
Posts: 475
Location: United States
Has thanked: 1 time
Been thanked: 24 times
I have had this discussion in the past with some people, not sure I completely understand the theory so hoping someone can explain.

So inside of players (with the exception of players without S-Video) the signal gets separated internally (using the players built in comb filter), TBC and NR are applied and then the separated signal is sent out via S-Video but the signal is recombined (again using the players built in comb filter) to be sent out the composite. First off is this statement correct?

Assuming that statement is correct could some players "damage" the signal with the separation / recombining or is this a 1:1 transition because the same comb filter is recombining it (this is where I have heard conflicting reports, some say the damage is done, some say its a 1:1 transition)?

Again assuming that first statement is correct then does the quality of the players comb filter actually matter even if using composite out (since the comb filter is being used internally, IE CLD-99 is better than CLD-D704 though everyone claims they are the same via composite out)?

And finally here is the main thing I am working on and wondering about (have experimented before and just started again), Player S-Video out -> DPS-575 (which corrects Y/C delay, and uses proc amp to correct colors / brightness / contrast / etc) then composite out of DPS-575 (which in theory does a better job of recombining the signal than the player) to video processor (be it Lumagen, C2, whatever) so that the high end comb filter in the processor gets the cleanest signal possible to work with. Does it make any sense to do it this way?

It has long been stated (by nearly everyone) to just disregard the S-Video out of any player and use the composite (since LD is native composite signal), the overall question is does this actually make sense or has the "damage already been done" internally by the player?
_________________
Panasonic LX-HD20
Sony HIL-C2EX
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 15 Jan 2019, 14:24 
Advanced fan
Advanced fan
User avatar

Joined: 16 Jun 2015, 15:40
Posts: 672
Location: Australia
Has thanked: 33 times
Been thanked: 24 times
I pondered over the recombined nature of the composite out in most players which led a while back to spending a bit of time comparing the s vs comp output of the players I've had more experience with (S9, R7G, H9 & X9). What I can say from that is that running composite to a better external comb filter did improve the issues I was seeing with the internal filters. However on all but the X9 there was some loss of sharpness/clarity in the composite image compared to the s-video out. The composite out of the X9 through external filter was the only one that looked as clear compared to its s-out.

I took from this that how well the signal is recombined does influence the picture, and of those players the only one that does it really well is the X9. I've also read that in theory if the recombination mirrors the separation (what I assume you mean by 1:1) then errors/issues should cancel out. This seemed to happen for all the players though; comb filter issues of the Mitsubishi filter were equally absent in the composite output of both the S9 and X9 but still the S9 lost something that the X9 didn't through composite out.

Whether this is because the X9 is closer to a perfect 1:1 recombination or there's other factors at play in the nitty gritty of the circuitry I don't really have the technical understanding to say. I'm not even sure the comb filter does the recombining - can they work in both directions simultaneously? Anyway I think if you have a player where like me you noticed a loss in the Composite out compared to the S-vid, then the idea is at least worth trying.
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 15 Jan 2019, 14:31 
Honest fan
Honest fan
User avatar

Joined: 29 Oct 2013, 15:56
Posts: 116
Location: United Kingdom
Has thanked: 2 times
Been thanked: 1 time
I have found this to be a better signal chain for the LD-X1 (S2). I feed the S-Video from the player into a Snell & Wilcox TBS 190 to recombine and output composite to C2. See thread:

[LD-X1] composite problem

The X1 doesn't do a good job at recombining the S-video signal back to composite.

I imagine it depends on the player and how accurately it recombines the Y/C. I've not tried this with other players but I will do some tests to find out if any improvements are made.
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 15 Jan 2019, 15:47 
True fan
True fan
User avatar

Joined: 11 Oct 2015, 16:35
Posts: 475
Location: United States
Has thanked: 1 time
Been thanked: 24 times
kencol wrote:
I have found this to be a better signal chain for the LD-X1 (S2). I feed the S-Video from the player into a Snell & Wilcox TBS 190 to recombine and output composite to C2. See thread:

[LD-X1] composite problem

The X1 doesn't do a good job at recombining the S-video signal back to composite.

I imagine it depends on the player and how accurately it recombines the Y/C. I've not tried this with other players but I will do some tests to find out if any improvements are made.


This is shockingly close to what I'm seeing with my Kenwood Z1, went to try out the SVideo for the heck of it and was amazed to see an uptick in clarity and sharpness (the Kenwood does have a 3d comb of some sort https://youtu.be/PLvt891nECM). I seem to be getting very good results with the DPS recombining and sending to C2 also.....
_________________
Panasonic LX-HD20
Sony HIL-C2EX
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 15 Jan 2019, 17:44 
Jedi Initiate
Jedi Initiate
User avatar

Joined: 16 May 2009, 18:05
Posts: 3011
Location: MA USA
Has thanked: 6 times
Been thanked: 43 times
chroma(color) and luma(white/brightness) overlaps in the form of ntsc in LD.

Let's say you did an absolutely terrible Y/C separation such as 1-D (notch filter). You cut the signal into two separate signals right in the middle of where Y and C overlaps. The end results would be horrendous on your TV. You would have some of that Luma leaked into Chroma and the vice versa.

How about now you combine the two signals together? Now you are back to your original signal since one signal had the entire upper portion and the other the lower portion. This is the worst case scenario but a good explanation of how recombining is none-destructive.

CLD-99 has a flaw. Instead of using the mitsubishi 3D comb filter , it uses the 2D comb filter found on the CLD-704 before the TBC. It later re-combines this for the composite output and uses the mitshubishi 3D comb filter before the S-video output.

Two reasons Pioneer did this. 1, they simply used CLD-704 and added a 3D comb filter at the end instead of completely redesigning the board. 2, when combined the effects of the 2D filter is nulled.


On players like HLD-X0 and X9, they use different components for each path. S-video path for instance gets futher noise reduction (and possibly edge enhancement) ICs. You can verify this by playing with the adjustments on composite output. You will see that some of those sliders won't alter the picture but work on S-video output.


Lumagen already has a frame based tbc and chroma delay function. CII has a line based TBC which is adequate for this sort of NTSC signal. In electronics, every component has its own signature on the output. Daisy chaining a bunch can fix some issues but may create others. Minimalistic is the safest approach for the most. But for those who have the time and resources it is not the defacto option. Also a lot of this is matter of personal opinion. I personally don't like artificially enhanced and sharpened video (or audio). For me X0 + Radiance 2144 yields the best picture.
_________________
Wanted
HD-VMD Discs
Hi-Vision/Muse Discs Only complete/mint!
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 15 Jan 2019, 17:47 
True fan
True fan
User avatar

Joined: 11 Oct 2015, 16:35
Posts: 475
Location: United States
Has thanked: 1 time
Been thanked: 24 times
substance wrote:
chroma(color) and luma(white/brightness) overlaps in the form of ntsc in LD.

Let's say you did an absolutely terrible Y/C separation such as 1-D (notch filter). You cut the signal into two separate signals right in the middle of where Y and C overlaps. The end results would be horrendous on your TV. You would have some of that Luma leaked into Chroma and the vice versa.

How about now you combine the two signals together? Now you are back to your original signal since one signal had the entire upper portion and the other the lower portion. This is the worst case scenario but a good explanation of how recombining is none-destructive.

CLD-99 has a flaw. Instead of using the mitsubishi 3D comb filter , it uses the 2D comb filter found on the CLD-704 before the TBC. It later re-combines this for the composite output and uses the mitshubishi 3D comb filter before the S-video output.

Two reasons Pioneer did this. 1, they simply used CLD-704 and added a 3D comb filter at the end instead of completely redesigning the board. 2, when combined the effects of the 2D filter is nulled.


On players like HLD-X0 and X9, they use different components for each path. S-video path for instance gets futher noise reduction (and possibly edge enhancement) ICs. You can verify this by playing with the adjustments on composite output. You will see that some of those sliders won't alter the picture but work on S-video output.


Lumagen already has a frame based tbc and chroma delay function. CII has a line based TBC which is adequate for this sort of NTSC signal. In electronics, every component has its own signature on the output. Daisy chaining a bunch can fix some issues but may create others. Minimalistic is the safest approach for the most. But for those who have the time and resources it is not the defacto option. Also a lot of this is matter of personal opinion. I personally don't like artificially enhanced and sharpened video (or audio). For me X0 + Radiance 2144 yields the best picture.


So basically you are saying no point in using SVideo at all?
_________________
Panasonic LX-HD20
Sony HIL-C2EX
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 15 Jan 2019, 19:03 
Jedi Master
Jedi Master
User avatar

Joined: 03 May 2004, 19:05
Posts: 5706
Location: United States
Has thanked: 77 times
Been thanked: 106 times
pbiancardi wrote:
So basically you are saying no point in using SVideo at all?


I thought this was the norm for the past 10 years? To not use S-video.
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 15 Jan 2019, 19:09 
True fan
True fan
User avatar

Joined: 11 Oct 2015, 16:35
Posts: 475
Location: United States
Has thanked: 1 time
Been thanked: 24 times
rein-o wrote:
pbiancardi wrote:
So basically you are saying no point in using SVideo at all?


I thought this was the norm for the past 10 years? To not use S-video.


No doubt, I'm just curious if this correct in every case as my eyes (and apparently other's eyes) seem to say different in some cases? I've also heard confilcting info on the internal recombining of the signals so I was curious about this.

I mean let's face it a lot of the things people say about LD are just regurgitated and incorrect, how many times has someone said "the 1010 has a red laser" only to have someone here finally debunk it (and I still hear people say it does).
_________________
Panasonic LX-HD20
Sony HIL-C2EX
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 15 Jan 2019, 19:41 
Jedi Master
Jedi Master
User avatar

Joined: 03 May 2004, 19:05
Posts: 5706
Location: United States
Has thanked: 77 times
Been thanked: 106 times
The only way I ever used S-video was when I had an either higher or lower quality TV and needed to use it for a player like the R7G and I'm sure the same
for an S9 or any other player where at the time the insides were better than TVs or the only way you could get the bells and whistles was to use S-Video on a CRT set.

Well, the 1010 is an entirely different issue, I've only heard that single person ever say it wasn't and then I believe they are gone unless they are trolling FB.
That red laser issue was posted on laserdiscarchive and other top LD player guys always said it was true, even techs when they were around.
But like I posted about that, no matter as even when it was sought after it was only fetching 250 so it really wasn't a big issue, I sold mine for 150 about 4 years ago and listed that it had a red laser. So its not like anybody got took, and I really do wish I still had it, as I want as many players as possible now. :lol:
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 15 Jan 2019, 20:46 
Jedi Initiate
Jedi Initiate
User avatar

Joined: 16 May 2009, 18:05
Posts: 3011
Location: MA USA
Has thanked: 6 times
Been thanked: 43 times
Two separated signals can be recombined without damage no matter how bad the separation has been. If you make changes to these signals after separation then the recombined signal will be changed as well. Noise reduction and sharpness make changes to these signals. Naturally low noise players are desired for this reason. Even with all this, if the processing on the player isn’t extremely destructive the recombined signal should as good as original or better(due to processing).


You ideally want to separate the signal with the best comb filter then move further with each step (convert to component then rgb etc.) the only reason we go back a step is because no LD player has a good enough comb filter built in. We recombine to redo the work with a modern more capable comb filter.

You could technically tap into the signal before the comb filter and skip the inferior processing. The major issue is that no modern (consumer grade) work with this signal that is misaligned and jittered(due to rotation of the disc). Outboard tbc are designed for professional grade equipment and unable to correct the raw ntsc feed from LD. The tbc in ld players are custom tuned for LD. It only makes sense to tap the signal after its been separated, time aligned and put back together. I see very little point in tapping the signal after this stage because video dacs in most Ld players are really high speed. I doubt there would be a visible difference. Sorta like using the component out from the cable box vs hdmi debate.
_________________
Wanted
HD-VMD Discs
Hi-Vision/Muse Discs Only complete/mint!
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 15 Jan 2019, 22:18 
Young Padawan
Young Padawan
User avatar

Joined: 14 Jan 2010, 09:44
Posts: 3579
Location: United States
Has thanked: 61 times
Been thanked: 151 times
When I compared composite and Y/C on my CLD-99 I remember the color being better with the s-video. However with my XBR-960 there are a lot of settings...I use the composite anyway because motion looks better when the 960 separates it.
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 16 Jan 2019, 06:12 
Young Padawan
Young Padawan
User avatar

Joined: 30 Aug 2005, 15:38
Posts: 3079
Location: Pennsylvania
Has thanked: 20 times
Been thanked: 31 times
I think if you have a really good comb filter like the the one in Sony XBR's (like Zeta mentioned) the recombined part becomes a non issue.

Using a CLD-D704 via composite into my KV-40XBR800 yielded better results than an R7G via S-Video by far--despite the recombined Y/C.

But really I think the best solution would be what that one poster from a couple years or so ago who doctored his CLD-D604 to output composite before Y/C separation by soldering a wire from a point on the PCB before the video hit the comb filter to an RCA jack (wish I could find that thread.)

Are Panasonic's also recombined composite? Because I think the composite output from the Runco LJR-II is pretty darn clean running to my Panasonic Plasma (which has a 3D comb filter.)
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 16 Jan 2019, 07:56 
Young Padawan
Young Padawan
User avatar

Joined: 14 Jan 2010, 09:44
Posts: 3579
Location: United States
Has thanked: 61 times
Been thanked: 151 times
I don't think there are probably too many players with pure composite out after a certain year. I think the on screen display works in the Y/C domain so in most Pioneer players the pure composite would probably have no OSD. It for sure wouldn't have any trick play.
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 16 Jan 2019, 12:08 
True fan
True fan
User avatar

Joined: 11 Oct 2015, 16:35
Posts: 475
Location: United States
Has thanked: 1 time
Been thanked: 24 times
I'm not actually saying use the S-Video out as the end game I'm saying S-Video out to another unit (in my case DPS-575) which recombines the signal instead of the player and then composite from that unit to your end game display / processor. According to what substance says that should be a waste of time, I've had others tell me it works well (see kencool above, others have told me the same, I didn't come up with the idea).
_________________
Panasonic LX-HD20
Sony HIL-C2EX
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 16 Jan 2019, 14:18 
Advanced fan
Advanced fan
User avatar

Joined: 16 Jun 2015, 15:40
Posts: 672
Location: Australia
Has thanked: 33 times
Been thanked: 24 times
pbiancardi wrote:
According to what substance says that should be a waste of time,

That's not the impression I got from what he said:
substance wrote:
Daisy chaining a bunch can fix some issues but may create others. Minimalistic is the safest approach for the most. But for those who have the time and resources it is not the defacto option.

I got from that it could be worth trying if you have the capability.
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 16 Jan 2019, 14:42 
True fan
True fan
User avatar

Joined: 11 Oct 2015, 16:35
Posts: 475
Location: United States
Has thanked: 1 time
Been thanked: 24 times
audioboyz1973 wrote:
pbiancardi wrote:
According to what substance says that should be a waste of time,

That's not the impression I got from what he said:
substance wrote:
Daisy chaining a bunch can fix some issues but may create others. Minimalistic is the safest approach for the most. But for those who have the time and resources it is not the defacto option.

I got from that it could be worth trying if you have the capability.


Yeah I guess you're right, only the crazy people like me should bother trying :lol: .
_________________
Panasonic LX-HD20
Sony HIL-C2EX
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 16 Jan 2019, 15:03 
Advanced fan
Advanced fan
User avatar

Joined: 16 Jun 2015, 15:40
Posts: 672
Location: Australia
Has thanked: 33 times
Been thanked: 24 times
pbiancardi wrote:
Yeah I guess you're right, only the crazy people like me should bother trying :lol: .


haha perhaps!!

As an X9 user I know there's no point for me to even try, I've tested its recombination and apparently no issues. I'm sure Substance mostly uses his X0 so same deal (I assume so anyway with no personal experience of the player).

I think though if your particular player is doing a poor job of recombining then this may be worth a shot, and kencol reports a positive outcome.

Almost everyone has this constant mantra "composite is better because LD is natively composite", while the second half of that is irrefutably true, the whole separation/recombination process introduces another element meaning overall this statement may not always hold true.

By recombining externally using a better recombination than built into the player may very well produce a better recombined composite signal than the player does. Feed that better composite into your superior external comb filter and better result overall! (Assuming chaining the extra gear doesn't introduce its own problems!!)
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 16 Jan 2019, 17:23 
Jedi Initiate
Jedi Initiate
User avatar

Joined: 16 May 2009, 18:05
Posts: 3011
Location: MA USA
Has thanked: 6 times
Been thanked: 43 times
As I tried to describe above, recombination is none destructive. There is no penalty in doing so even with the worst comb filter. Perfectly separating two overlapping signal is near impossible but recombination of the two signal although might have been imperfectly separated yields the original signal. On more modern players this entire event happens in digital therefore it’s a mathematical calculation.

LD has 7mhz luma and a fraction of that in chroma. These signals need to be over upsampled (stretched) and aligned there a separation is required. Also ld signal is too unstable without a tbc tuned for signal with such jitter. Very old player and low end players do this in the analog domain with major loss. Digital conversion and separation is necessary.

If someone tapped into anything in a ld player and getting an analog video output, it must be welll after the signal have been digitized, separated and converted back to analog again. Otherwise he would have to build his own video dac and/or build the entire board if tapped earlier in the chain. He probably by passed some added circuitry after the video dac which at this stage alll above has happened already.

Don’t forget the placebo effect. Your eyes wants to see what they want to see sometimes. Spending a ton of money or being too enthusiastic about some device bias you into thinking you are getting better results. Being subjective is really difficult.
_________________
Wanted
HD-VMD Discs
Hi-Vision/Muse Discs Only complete/mint!
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 16 Jan 2019, 21:34 
Young Padawan
Young Padawan
User avatar

Joined: 14 Jan 2010, 09:44
Posts: 3579
Location: United States
Has thanked: 61 times
Been thanked: 151 times
On a lot of old video cards that had only s-video out they just gave you a pigtail that summed the two signals...by soldering them together basically. I’m pretty sure that’s all it did anyway. Presumably you’d need to attenuate.
Offline
 Profile  
 
 Post subject: Re: Y/C separation happening in players / how to recombine.
PostPosted: 16 Jan 2019, 21:38 
Jedi Initiate
Jedi Initiate
User avatar

Joined: 16 May 2009, 18:05
Posts: 3011
Location: MA USA
Has thanked: 6 times
Been thanked: 43 times
signofzeta wrote:
On a lot of old video cards that had only s-video out they just gave you a pigtail that summed the two signals...by soldering them together basically. I’m pretty sure that’s all it did anyway. Presumably you’d need to attenuate.



Not necessarily. Two signals no longer have overlapping frequencies and occupy separate freq. bands therefor they don’t cancel each other out or multiply when added together.
_________________
Wanted
HD-VMD Discs
Hi-Vision/Muse Discs Only complete/mint!
Offline
 Profile  
 
Display posts from previous:  Sort by  
 Page 1 of 2 [ 28 posts ]  Go to page 1, 2  Next


Who is online

Users browsing this forum: No registered users and 2 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  

cron