It is currently 18 Oct 2021, 08:04




 Page 29 of 29 [ 580 posts ]  Go to page Previous  1 ... 25, 26, 27, 28, 29
Author Message
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 31 Jul 2021, 18:43 
Jedi Master
Jedi Master
User avatar

Joined: 03 May 2004, 19:05
Posts: 7413
Location: Dullaware
Has thanked: 824 times
Been thanked: 566 times
Real hard part is that LDs are not perfectly flat, they come up a little in the center.
So even if you have weight it will never be flat and that's why those adjustments are there to compensate for the irregularities in the disc.
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 04 Aug 2021, 17:38 
Serious fan
Serious fan
User avatar

Joined: 12 Sep 2015, 05:57
Posts: 199
Location: United States
Has thanked: 5 times
Been thanked: 51 times
titan91 wrote:
After another decode, it's hard to say if it's much better. I would assume so, as the SNR doesn't seem to drop as much as the disc goes on. SNR averages 35dB. A few screenshots after flattening from beginning to end:


Here's my take5 decode. SNR 35db and trailing. This is using --NTSC_color_notch_filter, at 28.6msps, and 10.7+ LPF.

Image

You can see i don't have the crosstalk issue. I wonder if the damage and then repair of that player whacked the alignment?
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 21 Aug 2021, 03:28 
Serious fan
Serious fan
User avatar

Joined: 12 Sep 2015, 05:57
Posts: 199
Location: United States
Has thanked: 5 times
Been thanked: 51 times
Here are some more detailed illustrations on how to wire the batteries to the OPA657 board.

Use the battery holders with a switch on them.
Image
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 08 Sep 2021, 12:43 
Honest fan
Honest fan
User avatar

Joined: 14 Jul 2019, 02:53
Posts: 75
Location: United States
Has thanked: 10 times
Been thanked: 3 times
I doubt it's the player, as you tested it yourself without issues after you repaired it. I can't reproduce it on my other discs. I'm trying to get my hands on more discs from a friend to borrow. Might be able to do that soon.
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 30 Sep 2021, 23:38 
Serious fan
Serious fan
User avatar

Joined: 05 Mar 2012, 03:43
Posts: 155
Location: Italy
Has thanked: 7 times
Been thanked: 2 times
A question I can't remember I already posed - memory lacks at my age... :D

Is an AC-3 output good to get raw signal to be decoded?
_________________
FanRes - restoring wrong movies, one title at the time!
blog.spoRv.com - a blog about video restoration…
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 01 Oct 2021, 00:02 
Serious fan
Serious fan
User avatar

Joined: 12 Sep 2015, 05:57
Posts: 199
Location: United States
Has thanked: 5 times
Been thanked: 51 times
laserdisc.ws wrote:
A question I can't remember I already posed - memory lacks at my age... :D

Is an AC-3 output good to get raw signal to be decoded?


no, that is just the AC3 RF.

you have to open the player and tap an RF test point. They are usually pretty easy to spot (but not always).
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 01 Oct 2021, 00:05 
Serious fan
Serious fan
User avatar

Joined: 05 Mar 2012, 03:43
Posts: 155
Location: Italy
Has thanked: 7 times
Been thanked: 2 times
9954tony wrote:
no, that is just the AC3 RF.

you have to open the player and tap an RF test point. They are usually pretty easy to spot (but not always).


But does it not include the whole RF signal, so an RF demodulator is needed to extract only the AC-3 track?
Well, if you have also a laserdisc player with an AC-3 out, I guess you already try it, right? ;)
_________________
FanRes - restoring wrong movies, one title at the time!
blog.spoRv.com - a blog about video restoration…
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 01 Oct 2021, 00:22 
Serious fan
Serious fan
User avatar

Joined: 12 Sep 2015, 05:57
Posts: 199
Location: United States
Has thanked: 5 times
Been thanked: 51 times
laserdisc.ws wrote:
9954tony wrote:
no, that is just the AC3 RF.

you have to open the player and tap an RF test point. They are usually pretty easy to spot (but not always).


But does it not include the whole RF signal, so an RF demodulator is needed to extract only the AC-3 track?
Well, if you have also a laserdisc player with an AC-3 out, I guess you already try it, right? ;)


it does not include the whole RF. it only includes the small RF that is stored on the one audio channel. AC3 discs will only have one analog audio channel, the other is used for AC3. Usually they put director commentary on there. The demodulator just changes the AC3 RF stored on the one audio channel to the format the receiver can handle.
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 01 Oct 2021, 17:30 
Serious fan
Serious fan
User avatar

Joined: 05 Mar 2012, 03:43
Posts: 155
Location: Italy
Has thanked: 7 times
Been thanked: 2 times
9954tony wrote:
it does not include the whole RF. it only includes the small RF that is stored on the one audio channel. AC3 discs will only have one analog audio channel, the other is used for AC3. Usually they put director commentary on there. The demodulator just changes the AC3 RF stored on the one audio channel to the format the receiver can handle.

Got it, thanks for the explanation!

Back to raw signal decoder: using my own capture cards, all PCI (from best to worst)
  • Aja Kona LS
  • Sweetspot (a.k.a. PDI Deluxe)
  • AverMedia BDA A16C (with Philips SAA713x chipset)
I noticed that I can get very high quality captures, comparable to the results of the raw decoder.

That's not to detract its quality - I think it's a fantastic project! - but for someone who could (or would) not, or simply has not the ability to ,open up his player and mod it, how good those "normal" captures could fare in comparison to the raw signal recording?

I'm talking mainly about video here - but you can add also analog audio (as we can capture lossless PCM AC-3 DTS tracks, they should be identical)

Let's state raw signal is 100%, how much a good analog video capture card like mine could score for resolution, colors, chroma crawl etc.? I guess they could be in the ballpark of 90/95, but it's only a supposition...
_________________
FanRes - restoring wrong movies, one title at the time!
blog.spoRv.com - a blog about video restoration…
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 01 Oct 2021, 17:48 
Serious fan
Serious fan
User avatar

Joined: 12 Sep 2015, 05:57
Posts: 199
Location: United States
Has thanked: 5 times
Been thanked: 51 times
laserdisc.ws wrote:
9954tony wrote:
it does not include the whole RF. it only includes the small RF that is stored on the one audio channel. AC3 discs will only have one analog audio channel, the other is used for AC3. Usually they put director commentary on there. The demodulator just changes the AC3 RF stored on the one audio channel to the format the receiver can handle.

Got it, thanks for the explanation!

Back to raw signal decoder: using my own capture cards, all PCI (from best to worst)
  • Aja Kona LS
  • Sweetspot (a.k.a. PDI Deluxe)
  • AverMedia BDA A16C (with Philips SAA713x chipset)
I noticed that I can get very high quality captures, comparable to the results of the raw decoder.

That's not to detract its quality - I think it's a fantastic project! - but for someone who could (or would) not, or simply has not the ability to ,open up his player and mod it, how good those "normal" captures could fare in comparison to the raw signal recording?

I'm talking mainly about video here - but you can add also analog audio (as we can capture lossless PCM AC-3 DTS tracks, they should be identical)

Let's state raw signal is 100%, how much a good analog video capture card like mine could score for resolution, colors, chroma crawl etc.? I guess they could be in the ballpark of 90/95, but it's only a supposition...


What's "good" can be rather subjective. Some people consider a "softer" picture with less overall video noise "better". Others may consider the picture with the "most detail, but slightly nosier" better. So there's that, and only you can decide that.

This concept comes up a lot, but one of the main points of the RF captures, is that once you have that, you can continuously improve the output you get as the code base gets better. Right now, a high quality NTSC composite capture may come really close to an ld-decode capture. That could change any minute though...

Right now, between PAL and NTSC in the *decode code base, the PAL matches the BBC transform decoder (the guy who originally designed it came in and contributed to the code: Jim Easterbrook). The NTSC decoding is really really good in *decode, but it is a (imo) a harder problem to solve than the PAL one, and no one like the BBC dedicated the resources to it the way the BBC did to the transform decode.

So for an "in the moment" point in time of right now, a person may say "composite capture is almost as good, so why go to the effort" when looking at the 2 side by side. With that choice, however, you are limiting yourself to never getting anything better than that. Whereas with *decode, you have the potential to continuously get a better decode from the same original RF capture.

Only the individual can make that choice...
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 01 Oct 2021, 18:01 
Serious fan
Serious fan
User avatar

Joined: 05 Mar 2012, 03:43
Posts: 155
Location: Italy
Has thanked: 7 times
Been thanked: 2 times
Yep, the PAL laserdisc players are not that great, as well as most editions... still, some are very good or even better than NTSC without the "Never Twice Same Color" problem, so getting the best.version.ever is very tempting... as well as NTSC, of course!!!

So, at the moment the NTSC analog capture could be on par or better than raw decoded, but as you wrote, the decoded one could improve only by updating the software - while the analog capture could, if possible, improve getting a better capture card and/or laserdisc player...

Well, for somone like me that knows what a soldering iron is but have no idea how to use it (properly), what are your hints to get the whole thing works? I'd "sacrifice" a Pioneer LD-V4300D for that! ;)
_________________
FanRes - restoring wrong movies, one title at the time!
blog.spoRv.com - a blog about video restoration…
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 01 Oct 2021, 18:08 
Serious fan
Serious fan
User avatar

Joined: 05 Mar 2012, 03:43
Posts: 155
Location: Italy
Has thanked: 7 times
Been thanked: 2 times
Forgot to mention: I can have access to various players, from the cited LD-V4300D to the DVL-909, CLD-D925, CLD-2950, Philips LDW600, up to a Panasonic LX-HD20 (clone of Sony HIL-C2EX) and the HLD-X9 - plus a Pioneer and Panasonic DVD-R to eventually use as TBC and/or comb filter.

Of course I'd use the top only for to capture the best needed discs, while reserving the others to capture "normal" ones.
_________________
FanRes - restoring wrong movies, one title at the time!
blog.spoRv.com - a blog about video restoration…
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 01 Oct 2021, 18:19 
Serious fan
Serious fan
User avatar

Joined: 12 Sep 2015, 05:57
Posts: 199
Location: United States
Has thanked: 5 times
Been thanked: 51 times
laserdisc.ws wrote:
Yep, the PAL laserdisc players are not that great, as well as most editions... still, some are very good or even better than NTSC without the "Never Twice Same Color" problem, so getting the best.version.ever is very tempting... as well as NTSC, of course!!!

So, at the moment the NTSC analog capture could be on par or better than raw decoded, but as you wrote, the decoded one could improve only by updating the software - while the analog capture could, if possible, improve getting a better capture card and/or laserdisc player...

Well, for somone like me that knows what a soldering iron is but have no idea how to use it (properly), what are your hints to get the whole thing works? I'd "sacrifice" a Pioneer LD-V4300D for that! ;)


That's not quite how i'd say it, but again, it is subjective. Ld-decode probably beats all but the "best" regular capture hardware now (in my opinion). The problem with the idea of
"improving hardware for future composite capture" is that all the hardware is aging and dying. If you take the idea that "a better capture card could come out, and i will just recapture on my x9" doesn't take into account these 3 truths: #1 Who's going to make a better hw capture card at this point? There is no market. #2 your discs are currently and continuously degrading (some faster than others, and maybe some so slowly they will last your lifetime, but how to know which is which) from rot and mediocre manufacturing. #3 one day, you will put a disc in your X9, and smoke will come out the top. All the players are aging too.

Tapping the players is actually pretty straightforward. You can even buy premade cables now. The official instructions are to drill a hole in the back of the player, but some of us have just snaked a small cable through one of the RCA jacks (so no drilling).

I'm not the expert on HW tapping though... just a apprentice/journeyman maybe?
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 01 Oct 2021, 18:53 
Serious fan
Serious fan
User avatar

Joined: 05 Mar 2012, 03:43
Posts: 155
Location: Italy
Has thanked: 7 times
Been thanked: 2 times
About PAL: as I tried all best players, I can say that none are better than average NTSC players - someone wrote about a CLD-704 IIRC; hence, the raw decoder "should" do better - I guess.

About everything you wrote about re-capturing: you are right!
1) nobody would make a better capture card - still I can find one better than mine!
2) after many decades, I guess the ones prone to laser rot are already rotten; the other should go fine if properly made, but as usual, "who knows"?
3) sure any player would die sooner or later - I hope later - but hoping to find another still working well is not impossible - for now

Still, as you wrote, once raw captured and properly stored, work is definitive.

An idea: as the LD-V4300D has a lot of outputs, I can always capture a disc in several ways at once:
- raw decoder
- capture card 1 from BNC output 1
- capture card 2 from BNC output 2
- capture card 3 from RCA output

Of course, I'd need four PCs - or a PC with PCI (and not PCI-e) good enough to be able to capture three signals from three different capture cards + raw decoder at once, but I guess it does not exist! :D
At least I have the old PC with PCI with which I could capture analog, and a new one for raw capture.
_________________
FanRes - restoring wrong movies, one title at the time!
blog.spoRv.com - a blog about video restoration…
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 01 Oct 2021, 18:59 
Serious fan
Serious fan
User avatar

Joined: 12 Sep 2015, 05:57
Posts: 199
Location: United States
Has thanked: 5 times
Been thanked: 51 times
laserdisc.ws wrote:
About PAL: as I tried all best players, I can say that none are better than average NTSC players - someone wrote about a CLD-704 IIRC; hence, the raw decoder "should" do better - I guess.

About everything you wrote about re-capturing: you are right!
1) nobody would make a better capture card - still I can find one better than mine!
2) after many decades, I guess the ones prone to laser rot are already rotten; the other should go fine if properly made, but as usual, "who knows"?
3) sure any player would die sooner or later - I hope later - but hoping to find another still working well is not impossible - for now

Still, as you wrote, once raw captured and properly stored, work is definitive.

An idea: as the LD-V4300D has a lot of outputs, I can always capture a disc in several ways at once:
- raw decoder
- capture card 1 from BNC output 1
- capture card 2 from BNC output 2
- capture card 3 from RCA output

Of course, I'd need four PCs - or a PC with PCI (and not PCI-e) good enough to be able to capture three signals from three different capture cards + raw decoder at once, but I guess it does not exist! :D
At least I have the old PC with PCI with which I could capture analog, and a new one for raw capture.


I'm pretty sure the consensus is that all discs rot/degrade, just a matter of how fast. So on a long enough timeline, the survival rate for every disc is "0".

There is more talk on the discord sight. Look back through this thread and find the link, and head over there. A lot more people, tapping a lot of different players, in a lot of different ways.
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 01 Oct 2021, 19:02 
Serious fan
Serious fan
User avatar

Joined: 05 Mar 2012, 03:43
Posts: 155
Location: Italy
Has thanked: 7 times
Been thanked: 2 times
For sure any laserdisc will fail - but probably I'll die before them, even if it will happen in 50 years from now! :P

OK I leave the topic now, until at least I could have my hands again of my whole hardware rig!

Thanks for the info! :thumbup:
_________________
FanRes - restoring wrong movies, one title at the time!
blog.spoRv.com - a blog about video restoration…
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 02 Oct 2021, 13:43 
Jedi Master
Jedi Master
User avatar

Joined: 03 May 2004, 19:05
Posts: 7413
Location: Dullaware
Has thanked: 824 times
Been thanked: 566 times
9954tony wrote:
I'm pretty sure the consensus is that all discs rot/degrade, just a matter of how fast. So on a long enough timeline, the survival rate for every disc is "0".


But this is true with everything, even books, comics, people. Welcome to planet Earth, us and our stuff.
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 03 Oct 2021, 03:03 
Hardcore fan
Hardcore fan
User avatar

Joined: 13 Aug 2018, 03:18
Posts: 1181
Has thanked: 305 times
Been thanked: 382 times
The theory that LDs will degrade is very plausible, but I haven't seen any proof for discs that were properly manufactured. They seem stable, but no telling what they'll look like in 50 years when I'm dead.

If you are just now getting into RF capturing, go for a better player than the V4300/V4400. I think they were chosen because they both can play PAL. (V4400 requires an IC swap to enable PAL playback).
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 03 Oct 2021, 05:56 
Serious fan
Serious fan
User avatar

Joined: 12 Sep 2015, 05:57
Posts: 199
Location: United States
Has thanked: 5 times
Been thanked: 51 times
cplusplus wrote:
The theory that LDs will degrade is very plausible, but I haven't seen any proof for discs that were properly manufactured. They seem stable, but no telling what they'll look like in 50 years when I'm dead.


In truth, that's my experience too. i should probably walk that back a bit, as i was just passing on information i read from others on discord.

cplusplus wrote:
If you are just now getting into RF capturing, go for a better player than the V4300/V4400. I think they were chosen because they both can play PAL. (V4400 requires an IC swap to enable PAL playback).


If you are only doing NTSC, then yeah, the MUSE players seem to provide a better RF capture. The 4300/4400 are simply well documented. The X9/X0 have been tapped, but it isn't as straightforward (but it isn't impossible either).

@laserdisc.ws not a huge deal, but the 4300 only has 1 bnc and 1 rca composite out, not 2 bnc composite outs. There are other bnc connectors on the back, but those are for sync in/out only.
Offline
 Profile  
 
 Post subject: Re: (WIP) Laserdisc software image decoder from raw signal
PostPosted: 05 Oct 2021, 20:08 
Serious fan
Serious fan
User avatar

Joined: 05 Mar 2012, 03:43
Posts: 155
Location: Italy
Has thanked: 7 times
Been thanked: 2 times
Yep, my bad - I was thinking about the fact it has three video outputs, but one of them is RGB! :angel:

Sure a MUSE player with its red laser could in theory grab more info - in particular with laser rot discs; still, even if I was an experienced welder (and I'm not) I would think not twice, but dozen times, to open up a MUSE player to mod it! :shock:
_________________
FanRes - restoring wrong movies, one title at the time!
blog.spoRv.com - a blog about video restoration…
Offline
 Profile  
 
Display posts from previous:  Sort by  
 Page 29 of 29 [ 580 posts ]  Go to page Previous  1 ... 25, 26, 27, 28, 29


Who is online

Users browsing this forum: No registered users and 1 guest


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: